JP4153798B2 - Safe driving diagnosis method and safe driving diagnosis device - Google Patents

Description

【００５６】
また、加速度時系列でなく、予め加減速についてそれぞれ定めた閾値以上の加速減速の加速度（急加減速）を記憶している場合の、加速度の標準偏差の求め方について述べる。この場合の計算方法の一例を数式２に示す。
【００５７】
【数２】

【００５８】
数式２は、最小二乗により経験的に求めた近似式とする。ここでの式中にある重み係数Ａ（１），Ａ（２），Ａ（３），Ａ（４）は、予め採取した（初速と終速から計算した）１秒毎の加速度から求めておく（以下同じ）。具体的には、予め１秒毎の加速度を採取（計算）しておき、その値により数式１で求めた加速度分散の真値，加速度時系列データから加速減速についてそれぞれ定めた閾値以上の値を取り出してそれぞれ急加速データ，急減速データとし、このデータから急加減速頻度，急加速平均，急減速平均，急加速２乗平均，急減速２乗平均の値を計算し、最小二乗法によりこれら重み係数Ａ（１），Ａ（２），Ａ（３），Ａ（４）を予め求めておく。また重み係数Ａ（１），Ａ（２），Ａ（３），Ａ（４）は、ニューラルネットワークを用いて関数近似を行って求めても良い。
【００５９】
次に、加速度の歪度についての他の計算方法について説明する。この計算を数式３に示す。
【００６０】
【数３】

【００６１】
ここで、加速度標準偏差とは数式１で計算した値である。また、加速度の時系列でなく、予め定めた加速減速の閾値以上の加速度を記憶している場合の、加速度の歪度の求め方について述べる。
【００６２】
この場合の計算方法の一例を数式４に示す。
【００６３】
【数４】

【００６４】
数式４は、最小二乗により経験的に求めた近似式とする。ここでの式中にある重み係数Ｂ（０），Ｂ（１），Ｂ（２），Ｂ（３），Ｂ（４），Ｂ（５）は、予め採取した１秒毎の加速度から求めておく。具体的には、予め１秒毎の加速度を採取し、その値により数式３で求めた加速度分散の真値、加速度時系列から閾値以上の値を取り出し、急加減速頻度，急加速平均，急減速平均，急加速２乗平均，急減速２乗平均，急加速３乗平均，急減速３乗平均の値を計算し、最小二乗法により重み係数Ｂ（０），Ｂ（１），Ｂ（２），Ｂ（３），Ｂ（４），Ｂ（５）を予め求めておく。また重み係数Ｂ（０），Ｂ(１），Ｂ(２），Ｂ(３），Ｂ(４），Ｂ（５）は、ニューラルネットを用いて関数近似を行って求めても良い。
【００６５】
図１８は、休憩法令違反情報処理をＰＡＤ表現した図である。法令では、４時間の連続運転毎３０分の休憩が義務付けられている。
そこで、走行車両１０における走行開始後の走行中断時刻と走行開始時刻を取り込み、その差分により連続運転時間を算出し、連続運転時間内における作業開始時刻と休憩時刻との差分により休憩時間を算出して、連続運転時間内における休憩時間と、休憩法令で規定された休憩時間との照合を行うことにより認定される法令違反の有無を判定する。なお、法令違反があった場合、その日数（回数）カウントして診断項目として用いても良い。
【００６６】
具体的には、図１８（ａ）は、連続運転時間を算出する方法を示している。休憩法令違反情報処理部３２５は、まず、連続運転時間最大を０に設定する（Ｄ２３１）。そして、走行軌跡時系列ＴＢＬ３５２に記録されている始業のコード検索を行い、運行車両の走行開始を検出したときに、荷積み、荷卸し、休憩、終業等、走行開始後から次の走行開始後の走行中断まで（Ｄ２３４）、走行開始後の走行中断時刻を検出する（Ｄ２３５）。
そして、連続運転時間を、走行開始後の走行中断時刻と走行開始時刻の差分とし（Ｄ２３６）、連続運転時間最大としてメモリにセットする（Ｄ２３７）。前記の動作（Ｄ２３３〜Ｄ２３７）は、走行軌跡時系列ＴＢＬ３５２に始業コードが設定されている全てのデータ分について繰り返される。
【００６７】
図１８（ｂ）に休憩法令違反設定のための処理が示されている。ここで、休憩法令違反情報処理部３２５は、始業から終業に渡って、Ｔ＝ｔ＋４時間３０分を計算し、Ｔが終業を越えた場合には終業時刻とする（Ｄ２３９）。そして、時刻ｔからＴ内における休憩時間を集計し（Ｄ２４０）、その休憩時間の集計値が３０分以下の場合に法令違反と設定する（Ｄ２４１、Ｄ２４２）。
休憩時間の集計処理は図１８（ｃ）に示されている。すなわち、休憩法令違反情報処理部３２５は、時刻ｔ〜Ｔの範囲で（Ｄ２４３）、走行軌跡時系列ＴＢＬ３５２に休憩コードが記録されていることを検索し（Ｄ２４４）、休憩時間を走行開始時刻と休憩時刻の差分を、休憩から次の走行（走行、運転自動検出コード）まで累算する（Ｄ２４５、Ｄ２４６）。
【００６８】
図１９は、アイドリング時間合計、アイドリング回数、アイドリングヒストグラムの集計方法についてＰＡＤ表現した図である。図９にアイドリングヒストグラムＴＢＬ（３５６１）のデータ構造が示されている。
ここでは、エンジンＯＮで位置情報に変化のないアイドリング状態が所定時間経過したときに、その回数と時間が計測され設定される。
【００６９】
具体的に、アイドリング集計処理部３４１は、まず、プログラムに割り付けられる全アイドリングヒストグラム回数メモリを０に設定し、また、アイドリング時間合計値を０にセットする（Ｄ２５１）。そして、走行軌跡時系列ＴＢＬ３５２を参照することにより、アイドリング開始が検出された場合（Ｄ２５３）、次のアイドリング終了まで（Ｄ２５４）、アイドリング時間をアイドリング終了時刻とアイドリング開始時刻の差分として求め（Ｄ２５５）、かつ、アイドリング時間からアイドリングヒストグラム回数メモリを＋１カウントアップする（Ｄ２５６）。
そして、アイドリング集計処理部３４１は、アイドリング回数をアイドリングヒストグラム回数メモリの総和としてセットする（Ｄ２５７）。
省燃費運転診断コンテンツ生成部３４による省燃費運転診断は、他に、加速度統計情報ＴＢＬ作成部３２２に記録された縦加速度平均値と、物理データＴＢＬ３５４に記録された平均速度、および走行距離と、運送業者１の労務管理者が管理端末１２を操作することにより入力される実燃料消費率とを統計処理することにより行われる。ちなみに、本実施形態の車載端末１１は、エンジンＯＮの状態で機能して、車両速度情報および位置情報を送信するようになっている。このため、ドライバがイグニッションスイッチによりエンジンを停止すれば、アイドリングの時間は計測されなくなる。
【００７０】
次に、安全運転診断の評価、およびその度合いの計算方法について述べる。安全運転診断の度合いは、交通事故を起こす確率に逆比例するように設定される。これは、予めドライバの事故歴（一年辺り何回事故を起こしたかの回数。以下、事故頻度と呼ぶ）が得られる場合はその情報と、加速度標準偏差、加速度歪度のペアの情報を集めておき、事故頻度を加速度標準偏差と、加速度歪度の値とで説明するための関係式を求める。
これにより、加速度標準偏差と加速度歪度を計測すれば、事故の頻度を推定することが可能になる。ここで加速度標準偏差は、走行地域の影響を受けやすい物理量であるが、ドライバの性格もある程度反映した量であるため、この物理量をも含める。この関係の一例として次の、数式５による方法がある。
【００７１】
【数５】

【００７２】
ここで、この式中にある重み係数Ｃ（０），Ｃ（１），Ｃ（２）は、予め集めておいたドライバの事故歴、加速度標準偏差、加速度歪度のペアの情報から、事故歴データを事故頻度と見なして、最小二乗で求めておく。また数式５の事故頻度の逆数を安全運転の度合いとする。またここで、最小二乗でなくとも、ニューラルネットを使用して、事故頻度を出力、加速度標準偏差、加速度歪度を入力とした関数近似を行って求めても良い。また、事故歴が得られ難い場合には、事故に至る可能性の高いと考えられる急減速を事故頻度の代わりのデータとして用いて加速度標準偏差、加速度歪度をもとに前記（数式５）により同様に事故頻度を推定する。
【００７３】
次に、安全運転の度合いの精度を向上させる方法について述べる。この精度は、数式５の最小二乗またはニューラルネットによる学習により決定される。このため、被験者のデータ数が多くなるにつれ、精度が向上することになる。このため図１の安全運転・省燃費運転診断サーバ３０に蓄えられた加速度の歪度と標準偏差の値の群を用いて再学習を行えば良い。
【００７４】
また、これらの学習データが全く無い場合には簡便な方法として、事前に複数の被験者の走行データの加速度時系列データを分析して得られた閾値を用いて、一例として以下のように設定する。加速度標準偏差がｘkm/h/sec以上でかつ、加速度歪度の絶対値がｙ以上のときには、安全運転の度合いを０とする。またそれ以外のときには、安全運転の度合いを１と簡易的にする。
【００７５】
最後に、警告画面とそのメッセージ内容について述べる。ドライバの運転の加速度の分布を元に、ドライバの性格を診断して評価を行った安全運転の度合いが、予め設定された閾値以上の場合警告を与えることにする。ここで閾値の値は１つ以上でも良い。本実施形態では５段階のレベルを用意し、各レベルの境界となる４つの閾値を用意する。そして安全なレベルから順にＡ，Ｂ，Ｃ，Ｄ，Ｅと名前を付けることにする。
また、加速度の加速側，減速側の分布の偏りを求めるため、加速度の歪度を求めておき、この加速度の歪度の値が減速側に大きな値を持つ場合には、そのドライバを、加速がゆったりで、減速が急であると判定し、加速側に大きな値をもつ場合には、そのドライバを、加速が急で減速がゆったりしていると判定する。例えば、加速度の歪度が−ｙ以下の場合、ｙ以上の場合、それ以外の場合の３通りに分ける。この加速度の歪度が−ｙ以下の場合、ｙ以上の場合、それ以外の場合の３通りに分けた組み合わせと、安全なレベルの組み合わせにより、診断メッセージを用意する。
【００７６】
ここで、加速度歪度が−ｙの場合には、加速度分布の減速の裾野が加速より広いことを意味するため、「ブレーキがきつめです」のメッセージを入れる。また加速度歪度がｙ以上の場合には、加速度分布の加速の裾野が減速より広いことを意味するため、「アクセルがきつめです」のメッセージを入れる。
さらに、加速度歪度が−ｙを超え、ｙ未満の場合には、加速度の分布が左右対称に近いことを意味するため、アクセルがきつい、ブレーキがきついのメッセージは入れない。また、安全レベルがＡの場合には「安全な運転をしています」、Ｂの場合には「安全運転に気をつけています」、Ｃの場合には「普通の運転をしています」、Ｄの場合には「やや荒っぽい運転をしています」、Ｅの場合には「荒っぽい運転をしています」のメッセージを入れる。また、安全レベルが２つしか設定されない場合のメッセージは、レベルＡとＥのみの表示とする。
【００７７】
すなわち安全運転の度合いが閾値より安全サイド（値が大きい場合）にある場合はレベルＡのメッセージ群の中から歪度の値に応じてメッセージを出す。一方、閾値より危険サイド（値が小さい場合）にある場合はレベルＥのメッセージ群の中から歪度の値に応じてメッセージを出す。
なお、この診断メッセージは、安全運転サービスセンタの管理者が変更できるようにする。
【００７８】
次に、安全レベルをＡ〜Ｅの５段階に分けた場合における閾値の決め方について説明する。この閾値は予め、ドライバの被験者を集めておき、主観的な安全のレベル分けをしておく。そして各々の安全レベルにおける、安全運転の度合いの平均値を求める。この各安全レベル毎の平均値の中間値を、各安全レベルの閾値とする。例えば、レベルＡの安全運転の度合いが１００、レベルＢの安全運転の度合いが８０とする。この場合レベルＡとＢの閾値は９０となる。
【００７９】
また安全のレベルが２つである場合の閾値の決め方について説明する。ここで、安全の度合いが交通事故頻度に基づいて計算している場合には、平均的なドライバの交通事故の頻度の逆数とする。また、安全の度合いが急減速の回数に基づいて計算している場合には、平均的なドライバの急減速の頻度の逆数とする。ここでの平均的なドライバは、日本全国のドライバの平均を使用するか、または安全運転・省燃費運転診断サーバ３０に蓄えられているドライバの平均値とする。
【００８０】
ここで、本発明により作成される帳票の一例について、図２０〜図２６を参照しながら説明する。
図２０に示す日々帳票によれば、各診断項目を総合的に判断して指導のコメントが付されると共に、安全、省燃費運転度合いが点数化され表現される（Ａ）。また、走行日の各診断項目値が記され、前月の平均値と比較することで運転の改善の度合いが把握できる（Ｂ）。
更に、１運行（始業〜終業）の走行経路が示される。ここでは、作業場所の他に、危険な運転を行った場所をマークすることで、これらが多発する地域を特定し、指導に役立たせることができる（Ｃ）。また、速度変化グラフも表示される。速度違反、速度ムラのある運転の時間帯をマークすることで、これらが発生しやすい時間帯を把握し、指導に役立てることができる（Ｄ）。
【００８１】
図２１に示す帳票は、月次帳票であり（ドライバの１ヶ月の月平均評価と１週間の走行情報の履歴）、各診断項目のうち、特に重要な項目について帳票を作成し、ドライバの月平均での安全運転、省燃費の傾向を報告するために用いられる。
ここでは、１ヶ月間における各診断項目の値の変化がグラフ表示されている。ここで、通常でない運転と思われる場合、図２０に示す日々帳票により詳細を確認することができる（Ｅ）。また、１ヶ月間の各診断項目値の平均値が記録される。また、前月の値を表示することにより、その傾向を明示することができる（Ｆ）。
【００８２】
なお、診断項目とは、運転者、走行日時、運行時間［ｈ］、ハンドル時間［ｈ］、最大連続ハンドル時間［ｈ］、走行時間［ｈ］、アイドリング時間［ｈ］、休憩時間［ｈ］、作業時間［ｈ］、走行距離［ｋｍ］、安全運転点数、省燃費運転点数、指導コメント、速度［km/h］、平均速度［km/h］、瞬間最高速度［km/h］、速度違反時間割合［％］、加速度、急加減速回数［回/h］、ヒヤリハット回数［回/h］、最大加減速度［Ｇ］、加減速分布［回/h］、加減速傾向分布をいう。
【００８３】
図２２に示すドライバ比較データは、走行距離、走行時間についてＡ氏、Ｂ氏、Ｃ氏、Ｄ氏各ドライバを対比して示したものであり、１運行当りの平均運行時間および走行距離は、安全運転診断の前後でほぼ同様である。図２３に示すドライバ比較データは、速度に関する評価が示されており、Ｂ氏とＤ氏は、指導の前後で速度違反の割り合いが減少していることが確認できる。
【００８４】
図２４に示すドライバ比較データは、安全指導前後における急加減速に関する評価が示されている。ここでは、Ａ氏とＢ氏には若干急加減速の傾向が見受けられ、Ｃ氏とＤ氏は、指導の結果、急減速の傾向が減少していることが確認できる。また、図２５に示すドライバ比較データは、安全指導前後における運転傾向に関する評価が示されている。ここでは、原点に近いほど安定した速度（定速度）で無駄な加速がない（アクセルワークが安定した）ドライバであることを示している。Ｄ氏は、安全指導により、加速、アクセルワーク共に改善が認められる。補足すると、図２５において、データが上方にあるほどドライバは急加速しやすい傾向にあり、データが下方にあるほどドライバは急減速しやすい傾向にある。また、図２６に示すドライバ比較データは、指導前後における燃費に関する評価（アイドリング時間、燃費）が示されている。ここでは、運行車両として冷凍車が例示されているため、アイドリング回数についてはその特殊事情（冷凍機を動かす関係上アイドリングストップをできない）により改善は見られないが、通常の車両では指導による改善が見られた。また、燃費については、全員改善が行われている。
【００８５】
なお、図２２〜図２６に示すドライバ比較データは、法人データとしても活用できる。すなわち、Ａ氏を社内Ａグループ、Ｂ氏を社内Ｂグループとし、比較することで、安全運転、省燃費運転に関して社内で競争意識を持たせ、活動を活性化することができる。勿論、グループとグループ内個人を並べて比較する帳票も作成可能である。このことにより、安全運転管理者は最小限の労力で各種帳票を入手でき、効率的な指導が可能となる。勿論、会社（企業）を１つのグループとみなしてデータを取り扱うようにし、会社同士の比較を行うことができるようにしても良い。
【００８６】
以上説明のように本実施形態は、車両から取得される物理データや労務データを一元管理してその統計的性質を分析することにより、ドライバに合致した格付けや帳票を自動作成するものであり、また、安全指導員が、前記した一元管理されるデータを参照して安全運転指導、あるいは省燃費運転指導を行う際に、そのコンテンツを容易に作成でき、安全指導員の負担軽減をはかるものである。
なお、図４に示す安全運転診断コンテンツ生成部３２、帳票出力部３３、省燃費運転診断コンテンツ生成部３４、速度違反情報ＴＢＬ作成部３２１、加速度統計情報ＴＢＬ作成部３２２、急ハンドル回数・最大ハンドル加速度算出部３２３、加速度標準偏差・歪度算出部３２４、休憩法令違反情報処理部３２５、アイドリング集計処理部３４１のそれぞれで実行される手順をコンピュータ読取り可能な記録媒体に記録し、この記録媒体に記録されたプログラムをコンピュータシステムに読み込ませ、実行することにより本発明の安全運転、省燃費運転診断装置が実現されるものとする。ここでいうコンピュータシステムとは、ＯＳや周辺機器等のハードウェアを含むものである。
【００８７】
また、本発明を実施するに際して、加速度センサは必須のセンサではなく、加速度を車両速度情報や車両位置情報（位置情報）に基づいて計算により算出するようにしても良い。また、横Ｇ（横加速度）については、横Ｇを用いないでも安全運転診断、省燃費運転診断を行うことができる。もちろん、横Ｇを用いて安全運転診断や省燃費運転診断を行っても良い。また、横Ｇは、専用のセンサを用いて計測することとしても良いし、例えば位置情報、車両速度情報等から計算により求めることとしても良い。また、車両速度を測定する車速パルスセンサは一例であり、本発明が車速パルスを生成するセンサに限定されることはない。また、速度規制の情報に、社内的な速度規制を加味するようにしても良い。
更に、運行車両１０に搭載される通信ユニット１１５はパケットに制限されず、携帯電話を用い、通常のモデムを使用して無線通信を行っても同様の効果が得られる。また、本実施形態では、車両にＧＰＳセンサ１３と車両センサ１４（車速）を設け、運行車両１０の現在位置と車速を計測することとしたが、ＧＰＳセンサ１３が持つ性能によっては車両センサ１４を不要とすることが可能である。このように、運行車両１０に搭載するハードウェア（車載端末１１を含む）を簡易な構成としながらも安全運転診断、省燃費運転診断を実現するものである。ちなみに、車両速度を測定する車速パルスセンサとＧＰＳセンサ１３は、広く普及しているので、このような安全運転診断等を行うに際して、利用者やサービス提供者に対する金銭的な負担が低減される。
【００８８】
【発明の効果】
以上説明のように本発明によれば、ドライバに合致した格付けや帳票を自動作成することができる。また、安全指導員が、管理（一元管理）されるデータを参照して安全運転指導、あるいは燃費指導を行う際に、そのコンテンツを容易に作成できるため、そのためにかかる安全指導員の負担軽減をはかることができる。また、安全運転診断の結果通知を適宜行い、あるいは安全指導により、ドライバの安全運転啓蒙につながり、交通事故の低減が見込まれる。
【図面の簡単な説明】
【図１】 本発明の安全運転診断方法および省燃費運転診断方法を実現するビジネスモデル、ならびにその利用主体との関係を明するために引用した図である。
【図２】 本発明で使用される運行車両に搭載される車載端末の構成を説明するために引用した図である。
【図３】 図１、図２に示す車載端末の内部構成を示すブロック図である。
【図４】 図１に示す安全運転・省燃費運転診断サーバの内部構成を機能展開して示したブロック図である。
【図５】 本発明実施形態で使用される車両ＤＢの各種テーブルのデータ構造を説明するために引用した図である。
【図６】 本発明実施形態で使用される走行軌跡時系列ＴＢＬの種別コードと記録情報との関係をマトリクス表現した図である。
【図７】 本発明実施形態で使用される速度違反情報ＴＢＬのデータ構造を説明するために引用した図である。
【図８】 本発明実施形態で使用される加速度統計情報ＴＢＬのデータ構造を説明するために引用した図である。
【図９】 本発明実施形態で使用されるアイドリングヒストグラムＴＢＬのデータ構造を説明するために引用した図である。
【図１０】 本発明実施形態の動作を説明するために引用した動作シーケンス図である。
【図１１】 本発明実施形態の動作を説明するために引用した動作シーケンス図である。
【図１２】 本発明実施形態で用いられる帳票出力のためのＧＵＩ画面の一例を示す図である。
【図１３】 本発明の安全運転診断プログラムの一部構造をＰＡＤ表現して示した図である。
【図１４】 本発明の安全運転診断プログラムの一部構造をＰＡＤ表現して示した図である。
【図１５】 本発明の安全運転診断プログラムの一部構造をＰＡＤ表現して示した図である。
【図１６】 本発明の安全運転診断プログラムの一部構造をＰＡＤ表現して示した図である。
【図１７】 本発明の安全運転診断プログラムの一部構造をＰＡＤ表現して示した図である。
【図１８】 本発明の安全運転診断プログラムの一部構造をＰＡＤ表現して示した図である。
【図１９】 本発明の省燃費運転診断プログラムの一部構造をＰＡＤ表現して示した図である。
【図２０】 本発明の安全運転診断装置により出力される帳票のフォーマットの一例を示す図である。
【図２１】 本発明の安全運転診断装置により出力される帳票のフォーマットの一例を示す図である。
【図２２】 本発明の安全運転診断装置により出力される帳票のフォーマットの一例を示す図である。
【図２３】 本発明の安全運転診断装置により出力される帳票のフォーマットの一例を示す図である。
【図２４】 本発明の安全運転診断装置により出力される帳票のフォーマットの一例を示す図である。
【図２５】 本発明の安全運転診断装置により出力される帳票のフォーマットの一例を示す図である。
【図２６】 本発明の省燃費運転診断装置により出力される帳票のフォーマットの一例を示す図である。
【符号の説明】
１…運送業者、２…データ収集事業者、３…安全運転指導事業者、４、５、６…通信網、１０…運行車両、１１…車載端末、１２…管理端末、２０…データ収集サーバ、３０…安全運転・省燃費運転診断サーバ（安全運転診断装置、省燃費運転診断装置）、３１…通信制御部、３２…安全運転診断コンテンツ生成部、３３…帳票出力部、３４…省燃費運転診断コンテンツ生成部、３５…車両ＤＢ[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a safe driving diagnosis method, a fuel-saving driving diagnosis method, an apparatus, a safe driving diagnosis program, and a fuel-saving driving diagnosis program.
[0002]
[Prior art]
There are a digital tachograph and a drive recorder as devices that are mounted on a vehicle and collect and store the behavior of the vehicle when the vehicle is running.
[0003]
The former digital tachograph collects vehicle physical data related to vehicle speed, vehicle position, mileage, etc., and stores it in flash memory (CF) or other memory that can be attached to and detached from the device for offline safe driving guidance and fuel consumption guidance. Used to do
In this safe driving instruction, the above-mentioned memory is taken into a personal computer, etc., the physical data and labor data are uniquely rated, and a form with a message such as a comment on driving is created and sent to the driver. This is realized by presenting (see Non-Patent Document 1).
[0004]
The latter drive recorder also collects physical data during vehicle travel and stores it in a built-in memory, which is used for accident situation analysis when an unexpected situation such as a traffic accident occurs (non-patented). Reference 2).
[0005]
[Non-Patent Document 1]
HP, Yazaki Sogyo Co., Ltd., Type Certification No. 1 Digital Tachograph, [Search January 24, 2003], Internet, <http://www.wbs.ne.jp/bt/yazaki-keiso/NewPages/Fdigi.htm >
[Non-Patent Document 2]
Japan Automobile Research Institute, Car Research (Vol. 23, No. 9, pp. 41-46), Yuji Arai et al., “Accident Analysis with Drive Recorder”
[0006]
[Problems to be solved by the invention]
As mentioned above, the safe driving instruction or the fuel-saving driving instruction by the conventional digital tachograph, the safe driving instructor makes a subjective rating on the obtained physical data and labor data, creates a form, and based on the form It was something to do. For this reason, the burden on the safe driving instructor is large, and the rating varies depending on the skills of the safe driving instructor.
Furthermore, the original data for safe driving diagnosis is recorded in a detachable memory and the data is taken off-line, so there is a risk of falsification, and a correct diagnosis result may not be obtained. In addition, when creating a form necessary for a safe driving instructor to give a safe driving instruction, it was difficult to search and edit physical data and labor data.
[0007]
The present invention has been made in view of the above circumstances, and can safely create a rating and a form that match a driver, a safe driving diagnosis method and a fuel saving driving diagnosis method, and an apparatus, a safe driving diagnosis program, and a fuel saving driving method. The purpose is to provide a diagnostic program.
In addition, by creating an environment that allows safe driving instructors to easily create content that is referred to when performing safe driving guidance or fuel saving guidance with reference to physical data and labor data that are centrally managed, Another object of the present invention is to provide a safe driving diagnosis method and a fuel-saving driving diagnosis method, a device, a safe driving diagnosis program, and a fuel-saving driving diagnosis program that reduce the burden on the safe driving instructor.
[0008]
[Means for Solving the Problems]
In order to solve the above-mentioned problems, the physical data and labor data acquired through the communication network are managed, the physical data is statistically analyzed, and the rating and the form matching the driver are provided in association with the labor data. It was decided. In addition, in order to enable safe driving instructors to refer to physical data and labor data that are managed and to provide content easily when performing safe driving instruction, it is possible to collect physical data from vehicles, and Equipped with an in-vehicle terminal that can input labor data (car navigation device can be substituted), and transfers it to a safe driving / fuel-saving driving diagnostic server (safe driving diagnostic device, fuel-saving driving diagnostic device) via a communication line (wireless) It was decided. In this way, as will be described later, when a safety instructor accesses the safe driving / fuel-saving driving diagnosis server, for example, the physical data and labor data for a certain period of the target driver are taken in, and the statistics for each diagnosis item are collected. A form showing the tendency of safe driving and fuel-saving driving is automatically generated.
[0009]
DETAILED DESCRIPTION OF THE INVENTION
Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings as appropriate.
This embodiment aims to enlighten safe driving and improve fuel efficiency using automatically generated forms, reducing the labor required for safety diagnosis and fuel-saving driving diagnosis by safety instructors that were previously performed offline. Is done. In addition, since data collection and form creation are performed on-line consistently, the possibility of data alteration is reduced, and accurate safe driving diagnosis and fuel-saving driving diagnosis can be realized.
Here, the “physical data” is data relating to the behavior of the vehicle, which is recorded every predetermined time and transmitted every predetermined time unit. The vehicle speed (vehicle speed information), the current vehicle position (position information), the travel Running data such as distance, and data generated when the vehicle is in a certain set state, which is recorded and transmitted along with time and vehicle position at the time of occurrence, status data such as sudden acceleration, sudden deceleration, idling, etc. Say. “Labor data” is data generated by pressing the operation panel installed on the vehicle terminal. When it occurs, it is recorded along with the time and position of the vehicle, and is sent and sent, starting work, closing work, loading, unloading, and resting. Event data input by the driver. In addition, data entered by operating a management terminal by a labor manager (operating manager of a blue number car → road transport law, etc.) such as labor data and fuel consumption data that cannot be entered by the driver, and a safe driving manager Additional data such as comments input by (safe driving manager of white number car → road traffic law) is also included.
[0010]
In addition, the “diagnostic item” includes a speed violation time ratio [%] that is a percentage of a travel time exceeding the legal speed, and an initial speed and an end speed of the sudden acceleration / deceleration state data. Of the acceleration estimated by the difference between them, the number of times of acceleration / deceleration that is an average value of the number of acceleration / deceleration times greater than a predetermined value generated per 1 hour of travel time [times / h], Average number of near-miss times [times / h], maximum acceleration / deceleration [G], distribution of acceleration / deceleration times [times / h] per hour of travel time, acceleration / deceleration trend distribution, number of safe driving points, saving Refers to the number of fuel-efficient driving points, instruction comments, etc.
[0011]
FIG. 1 is a diagram quoted for explaining the business model for realizing the safe driving diagnosis method and the fuel-saving driving diagnosis method of the present invention and the relationship with the user.
In FIG. 1, 1 is a carrier, 2 is a data collection company, and 3 is a safe driving guidance company. The carrier 1 mounts the in-vehicle terminal 11 shown in FIG. In addition, the data collection company 2 collects physical data of each operating vehicle 10 via the in-vehicle terminal 11, and the driver operates a control panel described later, or a labor manager or a safe driving manager Labor data input by operating the terminal keyboard is captured.
[0012]
That is, the physical data automatically collected by the in-vehicle terminal 11 installed in the operating vehicle 10 of the carrier 1 and the labor data input by the driver are installed in the data collection company 2 via the communication network 4. Collected by the data collection server 20. Further, labor data and fuel consumption data that cannot be input by the driver are data collection servers installed in the data collection company 2 via the communication line 5 by the labor manager operating the installed management terminal 12. 20 is transmitted.
Here, the data collection company 2 automatically creates a daily report indicating the daily labor situation and operation status of each driver, sends it to the management terminal 12 via the communication line 5, and presents it to the labor manager. On the other hand, the physical data, labor data, and fuel consumption data accumulated in the data collection server 20 are directly or partially processed, and installed in the safe driving guidance operator 3 via the communication network 6. Transfer to the fuel-saving driving diagnosis server 30.
[0013]
Based on the collected physical data and labor data, the safe driving / fuel-saving driving diagnosis server 30 performs statistical processing for each diagnosis item to generate safe driving diagnosis content, or generates fuel-saving driving diagnosis content.
Then, the safety instructor receives the access and outputs it as a form. Based on the form, the safety instructor provides guidance to the driver who is enrolled in the transporter 1 with the aim of enlightening safe driving and improving fuel efficiency.
[0014]
In this embodiment, the data collection company 2 and the safe driving guidance company 3 exist independently, and the data collection server 20 and the safe driving / fuel-saving driving diagnosis server 30 are installed respectively. It is not essential to exist independently, and may be integrated as a safe driving and fuel-saving driving diagnosis service provider. In this case, the data collection server 20 and the safe driving / fuel-saving driving diagnosis server 30 are connected to a LAN (Local Area).
Network).
The communication networks 4, 5, and 6 have been described as existing separately, but this is not essential, and the communication network may be shared using a network based on the Internet.
[0015]
FIG. 10 shows an operation sequence for daily form output, and FIG. 11 shows an operation sequence for driver comparison form output. Here, all show the relationship among the data collection server 20, the safe driving / fuel-saving driving diagnosis server 30, and the safe driving manager.
In FIG. 10, first, the safe driving / fuel-saving driving diagnosis server 30 outputs a list of contract corporations (transporters 1) according to the access of the safe driving manager. When the safe driving manager selects a corporation from a corporation list (corporate list display) displayed on his / her terminal, the safe driving / fuel-saving driving diagnosis server 30 can receive a list of drivers of the corporation from the data collection server 20. Information (driver list information) is acquired and transferred to the safe driving manager (S1 to S15), and driver selection information is acquired from the safe driving manager (S16). The safe driving / fuel-saving driving diagnosis server 30 transmits information of the selected driver and the affiliated company to the data collection server 20 (S17).
[0016]
Next, the safe driving / fuel-saving driving diagnosis server 30 acquires the travel date list information of the driver from the data collection server 20 (S18), processes it into a driver travel date calendar, and transfers it to the safe driving manager. The driver travel date calendar is displayed on the terminal held by the safe driving manager (S19). FIG. 12 shows an example of the driver travel date calendar displayed on the terminal held by the safe driving manager. FIG. 12 is a diagram showing an information input GUI (Graphical User Interface) screen for daily form output. In addition, an input parameter, a shipping company, a list of drivers with attributes, and a form type to be output are shown. Yes. In FIG. 12, “driver daily form” in the form type column is a display corresponding to the operation sequence of the daily form output process of FIG. Similarly, “Driver comparison ...” in the form column is a display corresponding to the operation sequence of FIG.
[0017]
Returning to FIG. The safe driving manager selects the departure date using the GUI screen described above (S20). As a result, the safe driving / fuel-saving driving diagnosis server 30 checks the existence of relevant data necessary for daily form (display) creation and creates daily forms (S101, S102).
On the other hand, if the corresponding data does not exist, the departure date is specified with respect to the data collection server 20 to obtain the daily data of the corresponding driver, and the data is accumulated (S103), and physical data TBL (to be described later) Table), labor data TBL and fuel consumption data TBL are constructed (S104), and a daily form display is created. The created daily form can be viewed by the safe driving manager on his / her terminal, and can also be printed by a printer.
[0018]
FIG. 11 shows the output process of the driver comparison form. The difference from the daily form output process shown in FIG. 10 is that a plurality of drivers are designated (S26) and the travel date calendar is displayed, It is to designate the period before and after the instruction (S30; period selection, S31; late period selection). The subsequent processing (S111 to S114) after checking the existence of the corresponding data based on the data is the same as the daily form output processing (S101 to S104) shown in FIG.
An example of a daily form is shown in FIGS. 20 and 21, and an example of a driver comparison form is shown in FIGS.
[0019]
Here, the outline of the equipment installed in the operating vehicle will be described with reference to FIGS. 2 and 3.
The traveling vehicle 10 shown in FIG. 2 has a GPS unit (receiver) 13/114 (Global Positioning System) and a vehicle speed from which the traveling position, speed, and direction of the traveling vehicle 10 can be obtained with the vehicle boarding terminal 11 as a core. A vehicle speed pulse sensor (vehicle sensor 14 for obtaining physical data) is connected. Furthermore, an operation panel 113 to which labor data is input by pressing a start button, an end button, an unloading button, a loading button, a break button, an expressway travel button, and an expressway release button are assigned. Physical data (position information by GPS receiver (unit) 13, vehicle speed information by vehicle speed pulse sensor), packet communication unit 115 for wireless communication for transmitting labor data to data collection server 20, CF memory card, etc. are connected Has been. Here, an acceleration sensor or the like that can obtain the acceleration of the vehicle may be connected as the vehicle sensor 14. Incidentally, in the present embodiment, the acceleration is obtained by calculation from the vehicle speed information.
FIG. 2 shows an example in which the in-vehicle terminal 11 is mounted, but a car navigation device having a wireless communication function can be substituted.
[0020]
FIG. 3 shows the internal configuration of the in-vehicle terminal 11 in a block expanded manner. In FIG. 3, blocks denoted by the same reference numerals as those shown in FIG. 2 are the same as those shown in FIG.
The in-vehicle terminal 11 has a CPU 111 as a control center, and is connected to a memory 112, an operation panel 113, a packet communication unit 115, and an I / O circuit 116. The packet communication unit 115 collects the physical data captured by the GPS receiver 13 and the vehicle sensor 14 and the labor data input and captured via the operation panel 113 through the wireless communication network 4 under the control of the CPU 111. It is transmitted to the server 20.
[0021]
Connected to the I / O circuit 116 is a display 117 composed of a GPS (unit) receiver 114, a vehicle sensor 14, a liquid crystal monitor, and the like.
The I / O circuit 116 takes in the current position (positional information) of the operating vehicle measured by the GPS sensor 13 and the GPS receiver 114 and supplies it to the CPU 111, and also measures the vehicle speed of the operating vehicle measured via the vehicle sensor 14. Information is supplied to the CPU 111 (an acceleration sensor or the like may be provided as the vehicle sensor 14). The result processed by the CPU 111 is displayed on the display 117 via the I / O circuit 116.
The operation panel 113 is assigned a start button, end button, unloading, loading button, break button, highway travel button, and highway release button, and the driver inputs these as labor data by pressing these buttons as appropriate. Shall be.
[0022]
FIG. 4 is a block diagram showing a functional development of the internal configuration of the safe driving / fuel-saving driving diagnosis server 30 in which the safe driving diagnosis device and the fuel-saving driving diagnosis device of the present invention are mounted.
The safe driving / fuel-saving driving diagnosis server 30 includes a communication control unit 31, a safe driving diagnosis content generation unit 32, a form output unit 33, a fuel-saving driving diagnosis content generation unit 34, and a vehicle DB (Data Base) 35. Composed.
[0023]
The communication control unit 31 manages a connection interface between the data collection server 20 installed by the data collection company 2 and the management terminal 12 installed in the carrier 1, and these are temporarily connected using a network such as the Internet. In this case, TCP / IP (Transmission Control Protocol / Internet Protocol) is implemented and the protocol is put into practice.
The safe driving diagnosis content generation unit 32 includes physical data obtained when the vehicle travels acquired from the in-vehicle terminal 11 via the communication network 4, and labor data input and captured by the driver or the operation manager or the safe driving manager. It has a function of generating a safe driving diagnosis content by performing statistical processing for each diagnosis item using at least one, and here, a speed violation information TBL creation unit 321, an acceleration statistical information TBL creation unit 322, A maximum steering wheel acceleration calculation unit 323, an acceleration standard deviation / distortion calculation unit 324, and a break law violation information processing unit 325 are included. The acceleration is calculated from the difference (ΔV) and the measurement time difference (Δt) between the initial speed (previous value, m−1th data) and the final speed (current value, mth data) of the vehicle speed information.
[0024]
The speed violation information TBL creating unit 321 uses the vehicle position information and the vehicle speed information acquired from the in-vehicle terminal 11 via the communication network 4 at regular time intervals, maps the position information on the map, and speed regulation (not shown). The speed regulation information at the current position is fetched from the information database, and a value obtained by calculating a percentage of the time traveled exceeding the speed regulation within the running time by comparing the speed regulation information with the vehicle speed information is stored in the physical data TBL354. The speed violation information TBL (3541, see FIG. 5) is written.
In addition, the acceleration statistical information TBL creation unit 322 performs rapid acceleration / deceleration (determined according to the difference between the initial speed and the final speed of the vehicle speed information acquired from the in-vehicle terminal 11 via the communication network 4 at regular time intervals ( (According to the same calculation method as acceleration) is counted, and the average value of the number of sudden accelerations / decelerations per driving unit time and the maximum acceleration are calculated to calculate acceleration statistical information in the physical data TBL354. It has a function of writing to TBL (3542).
The number of decelerations where the initial speed is a predetermined value or more (for example, 30 km / h, 40 km / h,...) Is defined as the number of near-misses, and the maximum near-miss acceleration at that time is also used as one of the diagnostic items.
[0025]
The number of sudden handles / maximum handle acceleration calculation unit 323 obtains the distribution of centrifugal force of the vehicle based on the sudden handle state acquired from the in-vehicle terminal 11 via the communication network 4, and calculates the standard deviation of the centrifugal force based on the centrifugal force distribution. It has a function to do.
The acceleration standard deviation / distortion calculation unit 324 obtains the acceleration distribution of the traveling vehicle 10 based on the acceleration information acquired from the in-vehicle terminal 11 via the communication network 4 at regular time intervals, and the standard deviation of the acceleration based on the acceleration distribution. And the function of calculating the bias of the acceleration / deceleration distribution. Details will be described later.
[0026]
The break law violation information processing unit 325 captures the travel interruption time and the travel start time after the start of travel in the operating vehicle acquired from the in-vehicle terminal 11 via the communication network 4, calculates the continuous operation time based on the difference, and calculates the continuous operation time. A violation of the law that is recognized by calculating the break time from the difference between the work start time and the break time within the operation time, and comparing the break time within the continuous operation time with the break time specified by the break law. It has a function to calculate the presence or absence and the number of times.
In each of the speed violation information TBL creation unit 321, the acceleration statistical information TBL creation unit 322, the number of sudden handles / maximum handle acceleration calculation unit 323, the acceleration standard deviation / distortion calculation unit 324, and the break law violation information processing unit 325 described above. The calculated values are supplied as respective diagnostic items to the safe driving diagnosis content generation unit 32, statistically processed by the safe driving diagnosis content generation unit 32, and output as a safe driving diagnosis content by the form output unit 33.
[0027]
The safe driving diagnosis content generation unit 32 also adds evaluation data related to a driving tendency of a certain driver to the generated content for each diagnostic item based on the access of the safe driving instructor, and performs safe driving for the driver. It has a function of outputting a safety diagnosis content necessary for providing guidance via a form output unit 33. At this time, the safe driving diagnosis content generation unit 32 comprehensively determines each diagnosis item, scores the degree of safe driving, and reflects it in the form.
It also has a function for outputting evaluation data for relative comparison indicating the driving tendency of a plurality of drivers for each diagnosis item.
[0028]
On the other hand, the fuel-saving driving diagnosis content generation unit 34 performs statistical processing on the physical data in the vehicle travel acquired from the in-vehicle terminal 11 via the communication network 4 and the labor data input and captured by the driver or the operation manager. And has a function of generating fuel-saving driving diagnosis content, and includes an idling total processing unit 341.
The idling total processing unit 341 obtains the idling histogram (3561, .3) when the idling state acquired from the in-vehicle terminal 11 via the communication network 4 and the engine is ON and the position information does not change has passed for a predetermined time. This function is used as a diagnostic item for generating fuel-saving driving diagnostic content. In addition, the fuel-saving driving diagnosis content generation unit 34 takes in the average acceleration value, average speed, mileage, and actual fuel consumption rate input by the labor manager of the carrier 1 and calculates it by the idling tabulation processing unit 341. The fuel-saving driving diagnosis content is generated by statistical processing together with the number of times and the time data.
[0029]
An example of the table (TBL) structure of the vehicle DB 35 is shown in FIGS. The table used in the embodiment of the present invention is composed of a travel locus time series TBL352, acceleration histogram TBL353, physical data TBL354, labor data TBL355, and fuel consumption data TBL356 with driver data TBL351 as a core.
[0030]
The driver data TBL 351 includes, in addition to bibliographic items such as the driver, vehicle type, vehicle weight, vehicle number, driving departure date, driving departure time, etc., the travel locus time series TBL 352, acceleration histogram TBL 353, physical data TBL 354, labor data TBL 355, fuel consumption data It has each data field of TBL356.
The travel locus time series TBL 352 includes data fields of number, date, time, latitude, longitude, type, maximum speed, accumulated distance, initial speed, final speed, average speed, acceleration variance, and lateral G. Here, the fixed-cycle running in the type data field is data at intervals of n minutes (n = 1, 2, 15), the maximum speed is the maximum speed in the past n minutes (unit is km / h), The average speed is the average speed in the past n minutes (unit is km / h), and the acceleration standard dispersion is the acceleration dispersion in the past n minutes (unit is (km / h / sec) ² ), Lateral G is the acceleration in the lateral direction when there is a sudden lateral G (+ is in the right direction, the unit is km / h), the initial speed and the final speed are recorded during sudden acceleration / deceleration. The initial speed and final speed of deceleration (unit: km / h). In this embodiment, since the lateral G sensor is not provided as the vehicle sensor 14, the lateral G column is blank in the table. Incidentally, the lateral G can be obtained by calculation from the basic specification data of the vehicle, the travel locus and the vehicle speed information (for example, it can be calculated as a lateral force acting in a direction perpendicular to the traveling direction of the vehicle).
[0031]
In addition, in FIG. 6, the code of the type of the traveling locus time series TBL352 (start, end work, break, high speed, high speed release, travel, loading, unloading / automatic operation detection, power OFF, idling start, idling end, fixed cycle Travel, sudden acceleration, sudden deceleration, sudden steering) and recorded information (maximum speed, accumulated distance, initial speed, final speed, average speed, acceleration standard deviation, lateral G) at that time are expressed in a matrix.
Here, items marked with ○ are data items that are automatically collected by pressing a button on the operation panel 113 or are automatically collected, and the date, time, latitude, longitude, and type data are essential regardless of the type code. And In FIG. 6, for example, when traveling at a constant cycle (when traveling stably), the maximum speed, average speed, and acceleration standard deviation are recorded in the travel locus time series TBL 352, and during rapid acceleration and rapid deceleration The accumulated distance, the initial speed, and the final speed are recorded in the travel locus time series TBL352. The accumulated travel distance is recorded at the start of idling. The start of idling is determined after it is determined by idling determination that idling is being performed. In FIG. 6, the recording of the lateral G is performed in an embodiment using a vehicle including a lateral G sensor or an embodiment in which the lateral G is obtained by calculation.
[0032]
The acceleration histogram TBL353 has data fields of acceleration, number of times of vertical acceleration (number of times of acceleration in the front-rear direction), and number of times of lateral acceleration. Here, each of the vertical distributions when the acceleration distribution is divided into 20 (−10 to +10) is shown. The number of accelerations and the number of lateral accelerations are shown. Here, the threshold value for rapid acceleration / deceleration is 11 km / h / sec (0.3 G), and a value exceeding the threshold value is stored. The number of times here is the number of times when 1 second is taken as the measurement time (Δt = 1 sec).
The physical data TBL 354 includes data fields for travel time, travel time, travel distance, average speed, highway travel time, general road travel time, highway travel distance, general road travel distance, highway average speed, and general road average speed. The secondary data of the speed violation information TBL (3541) and the acceleration statistical information TBL (3542) is created and assigned from the primary data written to each. An example of data items of the speed violation information TBL (3541) is shown in FIG. 7, and an example of data items of the acceleration statistical information TBL (3542) is shown in FIG. Details of these data items will be described later together with the respective table creation logic.
[0033]
The labor data TBLL 355 includes data items such as operation time, maximum continuous operation time, and break law violation information. How to create break law violation information will be described later.
The fuel consumption data TBL 356 includes primary data of idling time, number of idlings, average value of vertical acceleration, average speed, travel distance, actual fuel consumption rate, and idling histogram TBL (3561) of secondary data created from these primary data. An example of the data item of the idling histogram TBL (3561) is shown in FIG. Details of the data items will be described later together with the table creation logic.
[0034]
Hereinafter, although detailed description will be made for each safe driving diagnosis item, examples of physical data, labor data, and fuel consumption data will be described in advance.
Physical data includes vehicle speed, position, longitudinal acceleration, lateral acceleration time series, idling state, and mileage. The acceleration is blank, and in the embodiment in which the lateral acceleration is obtained, the time series of the lateral acceleration is written as shown in the table). Labor data includes the departure of a car, the end of traveling, a resting state, loading, unloading conditions, and traveling state on a highway. Further, the fuel consumption data includes the amount of fuel supplied to the vehicle, the travel distance when the fuel is supplied, date, and time.
Hereinafter, secondary data generated from the physical, labor, and fuel consumption data by the safe driving diagnosis program and the fuel saving driving diagnosis program (hereinafter simply referred to as a program) of the present invention, that is, speed violation information and acceleration statistics. Information, sudden handle count / maximum handle acceleration, acceleration standard deviation / distortion, break law violation information, and idling histogram will be described in detail.
[0035]
FIG. 13 is a PAD (Problem Analysis Diagram) representation of the program structure for creating the speed violation information TBL (3541) shown in FIG. 5, and FIG. 7 shows the data structure of the speed violation information TBL (3541). Has been. Hereinafter, the program structure will be described with reference to FIG.
The speed violation information TBL creation unit 321 first sets 0 to the “speed violation number counter (highway / general road)” assigned to the program and 0 to the “speed violation distance” (D131). Next, the vehicle position information and the vehicle speed information are taken from the travel locus time series TBL352 and the position information is mapped (map matching) on the map (D132).
[0036]
Then, road speed regulation information corresponding to the current position obtained by map matching is acquired only for the type code “constant period” of the travel locus time series TBL352 (D134). Here, the speed restriction information is acquired from a map database (not shown), but the speed regulation information may be fixed as a general road 40 km and a highway 80 km. On the other hand, the speed regulation speed is set to 80 km / h when the data is “fixed cycle” and the vehicle is traveling on a highway (D136). That is, the speed regulation speed is changed according to the road (according to the speed regulation).
Further, the obtained speed regulation information (α) is compared with the average speed (β) in the data of the fixed period (D137), and when α> β and driving on the highway ( In step S138, the highway speed violation number counter is incremented by 1, and the average speed × n minutes / 60 is added to the highway speed violation distance (D139). In the case of α> β and when traveling on a general road, the average speed × n minutes / 60 is added to the general road speed violation distance (D140). The above procedure is repeated for all “fixed period” data (D133).
[0037]
Then, the maximum speed of the general road and the maximum speed of the highway in the speed violation information TBL (3541) are obtained, and the percentage of time that the vehicle travels exceeding the speed regulation within the travel time is calculated. The speed violation information TBL (3541) is created as the distance ratio, the general road speed violation time ratio, the highway speed violation distance ratio, the highway speed violation time ratio, the speed violation distance ratio, and the speed violation time ratio, and written to the physical data TBL354. (D142).
[0038]
FIGS. 14 and 15 show a PAD representation of the program structure for creating the acceleration statistical information TBL (3542), and FIG. 8 shows the data structure of the acceleration statistical information TBL (3542).
Here, the number of occurrences of sudden acceleration / deceleration determined according to the difference between the initial speed and the final speed of the vehicle speed information acquired at regular time intervals is counted, and the sudden acceleration / deceleration that occurs per unit travel time is counted. The average number of times and the maximum sudden acceleration / deceleration are calculated.
[0039]
Specifically, in FIG. 14A, the acceleration statistical information TBL creation unit 322 sets the total acceleration counter (fG to hG) assigned to the program to 0 (times) and the maximum sudden acceleration to 0 (G). (D151). It should be noted that f <g <h (that is, f is moderate and h is abrupt for both acceleration and deceleration).
Next, the initial speed and the final speed are taken from the travel locus time series TBL352, and the difference (final speed-initial speed) is calculated, and the difference is within the range of level 1 (akm / h / sec to bkm / h / sec). (D153), the rapid acceleration number counter (fG to gG) is incremented by +1 (D154). If not, and if the difference is in the range of level 2 (ckm / h / sec to dkm / h / sec) (D155), the sudden acceleration counter (gG to hG) is incremented by one. (D156) If NO, if the difference is level 3 (within ekm / h / sec) (D157), the number of sudden accelerations (over hG) is incremented by +1 (D158). The above operations (D153 to D158) are repeated for all data whose type code is “rapid acceleration” (D152).
When the difference (final speed−initial speed) is greater than or equal to the maximum sudden acceleration × 3.6 × 9.8 (D159), the maximum sudden acceleration is expressed as (initial speed−final speed) / (3.6 × 9 .8) (D160).
[0040]
A method for calculating the number of times the acceleration standard deviation threshold is exceeded will be described using the PAD expression of FIG.
First, the acceleration statistical information TBL creation unit 322 sets an acceleration standard deviation threshold excess number counter assigned to the program to 0 (D161). When the value obtained by calculating the square root of acceleration variance is equal to or greater than the acceleration standard deviation threshold value (D163), the acceleration standard threshold count counter is incremented by 1 (D164). The above operations (D163, D164) are repeated for all “constant period” data in the type code of the travel locus time series TBL352 (D162).
The acceleration statistical information calculated as described above is written in the acceleration statistical information TBL (3542) in the physical data TBL 354 as the number of rapid accelerations for each level as shown in FIG. 8A. Further, as shown in FIG. 8B, it is possible to create a time frequency or distance frequency number TBL (3543) as time frequency or distance frequency as necessary.
[0041]
FIG. 15 is a PAD representation of the procedure for creating the number of sudden decelerations, the number of near-misses, the maximum sudden acceleration, and the maximum near-miss acceleration. Here, the number of decelerations at which the initial speed is equal to or greater than a predetermined value such as 30 km / h (= j) is defined as the number of near-misses, and the maximum near-miss acceleration at that time is calculated.
First, the acceleration statistical information TBL creation unit 322 sets the total deceleration number counter (fG to hG) and the near-miss number counter assigned to the program to 0, and sets the maximum sudden deceleration and the maximum near-miss acceleration to 0 (D171).
[0042]
Next, the initial speed (m-1st data) and the final speed (mth data) are taken from the travel locus time series TBL352, the difference (final speed-initial speed) is calculated, and the difference is level 1 (akm / When it is in the range of h / sec to bkm / h / sec) (D174), the rapid acceleration number counter (fG to gG) is incremented by 1 (D175). If not, and if the difference is in the range of level 2 (ckm / h / sec to dkm / h / sec) (D176), the sudden acceleration counter (gG to hG) is incremented by one. (D177) If NO, if the difference is level 3 (within ekm / h / sec) (D178), the number of sudden accelerations (over hG) is incremented by +1 (D179). When the difference (initial speed−final speed) is greater than or equal to the maximum sudden acceleration × 3.6 × 9.8 (D180), the maximum sudden acceleration is expressed as (initial speed−final speed) / (3.6 × 9 .8) (D181). The above operations (D174 to D181) are repeated when the initial speed is jkm / h or less (D173) for all data for which the type code is “rapid deceleration” (D172).
[0043]
The acceleration statistical information calculated as described above is written in the acceleration statistical information TBL (3542) in the physical data TBL 354 as the number of rapid decelerations for each level as shown in FIG. Further, as shown in FIG. 8B, it is possible to create a time frequency or distance frequency TBL (3543) (not shown) as time frequency or distance frequency as required.
[0044]
On the other hand, in D173, when the initial speed is equal to or less than jkm / h, the difference between the initial speed and the final speed is in the range of level 1 (akm / h / sec to bkm / h / sec) (D182). fG to gG) are incremented by 1 (D183). If not, and if the difference is in the range of level 2 (ckm / h / sec to dkm / h / sec) (D184), the near miss number counter (gG to hG) is incremented by 1 ( D185), if NO, if the difference is level 3 (within ekm / h / sec) (D186), the number of near-misses (more than hG) is incremented by +1 (D187).
When the difference (initial speed−final speed) is equal to or greater than the maximum near-miss acceleration × 3.6 × 9.8 (D188), the maximum near-miss acceleration is set to (initial speed−final speed) / (3.6 × 9 .8) (D189). The above operations (D182 to D188) are repeated (D173) when the initial speed is equal to or less than jkm / h (D173) for all data whose type code is “rapid deceleration”.
[0045]
As shown in FIG. 8A, the acceleration statistical information calculated as described above is written in the acceleration statistical information TBL (3542) in the physical data TBL 354 as the number of near misses for each level. Further, as shown in FIG. 8B, it is possible to create a time frequency or distance frequency number TBL (3543) as time frequency or distance frequency as necessary.
[0046]
FIG. 16 is a PAD representation of a method for calculating the number of sudden handles and the maximum handle acceleration. The embodiment of the steep handle or the like is executed by, for example, a vehicle including a rudder angle sensor and a lateral G sensor (a vehicle having a function of calculating a rudder angle [steering speed] and a lateral G [lateral force]). .
First, the number of sudden handles / maximum handle acceleration calculation unit 323 sets the total sudden handles count counter assigned to the program to 0, the maximum right handle acceleration to 0, and the maximum left handle acceleration to 0 (D191).
Next, when the lateral G is taken from the travel locus time series TBL 352 and the lateral G is in the range of level 1 (akm / h / sec to bkm / h / sec) (D194), the steep right hand wheel counter (fG ˜gG) is incremented by 1 (D195). If not, and if the lateral G is in the range of level 2 (ckm / h / sec to dkm / h / sec) (D196), the steep right hand wheel counter (gG to hG) is incremented by +1. Count up (D197). If NO, if the lateral G is at level 3 (within ekm / h / sec) (D198), the steep right hand wheel counter (above hG) is incremented by +1 ( D199). If the lateral G is greater than or equal to maximum steep right handle acceleration × 3.6 × 9.8 (D200), the maximum steep right handle acceleration is set to lateral G / (3.6 × 9.8). (D201).
The above operations (D194 to D201) are repeated (D192) when the lateral G is 0 or more (D193) for all the data whose type code is “steep handle”.
[0047]
As shown in FIG. 8A, the acceleration statistical information calculated as described above is acceleration statistical information in the physical data TBL 354 as the maximum number of steep right handles and the maximum steep right handle acceleration for each level. It is written in TBL (3542). Further, as shown in FIG. 8B, it is possible to create a time frequency or distance frequency number TBL (3543) as time frequency or distance frequency as necessary.
[0048]
On the other hand, in D193, when the lateral G is 0 or less (G ≦ 0), when the lateral G is in the range of level 1 (akm / h / sec to bkm / h / sec) (D202), the steep left handle The number counter (fG to gG) is incremented by 1 (D203). If not, and if the lateral G is in the range of level 2 (ckm / h / sec to dkm / h / sec) (D204), the steep left steering wheel counter (gG to hG) is incremented by +1. Count up (D205), if not, and if the lateral G is at level 3 (within ekm / h / sec) (D206), the steep left handle number counter (more than hG) is incremented by +1 ( D207). When the lateral G is greater than or equal to maximum steep left handle acceleration × 3.6 × 9.8 (D208), the maximum steep left handle acceleration is set to lateral G / (3.6 × 9.8). (D209).
The above operations (D202 to D209) are repeated when the horizontal G is 0 or more (D193) for all data for which the type code is “steep handle” (D192).
[0049]
As shown in FIG. 8A, the acceleration statistical information calculated as described above is acceleration statistical information in the physical data TBL 354 as the maximum steep left handle count and the maximum steep left handle acceleration for each level. It is written in TBL (3542). Further, as shown in FIG. 8B, it is possible to create a time frequency or distance frequency number TBL (3543) as time frequency or distance frequency as necessary.
[0050]
FIG. 17A is a diagram showing the calculation method of the acceleration standard deviation, acceleration skewness, and acceleration average value in PAD expression.
Here, the acceleration distribution of the vehicle is obtained, and the standard deviation of the acceleration due to the acceleration distribution and the bias (acceleration distortion) of the acceleration / deceleration distribution are calculated.
[0051]
Specifically, the acceleration standard deviation / distortion calculation unit 324 reads data from the acceleration histogram TBL353 and the travel locus time series TBL352, and sets an acceleration count memory (assigned to a program) for each acceleration (D211). Then, the longitudinal acceleration average value Ap is calculated from the acceleration count memory.
The longitudinal acceleration average value Ap is obtained by calculating the following arithmetic expression (1) (D212). However, a> 0 here.
Ap = ΣaA (a) / (N × 3.6 × 9.8) (1)
Next, the vertical acceleration standard deviation σ is obtained from the acceleration count memory by calculating the following arithmetic expression (2) (D213).
σ = √ (Σ (a−Aa) ² A (a) / (N × 3.6 × 9.8)) (2)
[0052]
Then, the vertical acceleration skewness S is obtained from the acceleration count memory by calculating the following arithmetic expression (3) (D214).
S = Σ (a−Aa) ^Three A (a) / (N × 3.6 × 9.8 × σ ^Three ) ... (3)
Further, the lateral acceleration standard deviation σv is obtained from the acceleration count memory by calculating the following arithmetic expression (4) (D215).
σv = √ (Σ (a−Av) ² C (a) / (M × 3.6 × 9.6) (4)
Here, A (a) is the acceleration frequency memory of the vertical acceleration a, C (a) is the acceleration frequency memory of the lateral acceleration a, and a is a value from the maximum vertical deceleration to the maximum vertical acceleration (the unit is km / h). / sec), N is the total number of longitudinal accelerations ΣA (a), M is the total number of lateral accelerations ΣC (a), Aa is the longitudinal acceleration average value ΣaA (a) / N, and Av is the lateral acceleration average value ΣaC (a) / M.
The number of sudden steering / maximum steering acceleration calculation unit 323 obtains the distribution of centrifugal force of the vehicle according to the sudden steering state acquired from the in-vehicle terminal 11 via the communication network 4, and the standard deviation of the centrifugal force due to the centrifugal force distribution. The standard deviation of the centrifugal force calculated here uses the above-described arithmetic expression (2). By the way, from the standard deviation of the centrifugal force, it is possible to read the driving characteristics of the driver, such as the driver easily turning the steering wheel, so the driver can be instructed in safe driving along the standard deviation of the centrifugal force. .
[0053]
FIG. 17B shows how to set the acceleration count memory for each acceleration by PAD expression.
Here, first, the maximum value and the minimum value of the acceleration a are obtained (D216: unit is km / h / sec). Then, the memory of A (a) is cleared over the range from the maximum value to the minimum value (D217). Next, the number of vertical accelerations is taken from A (a) and C (a) from the acceleration histogram TBL353 and set (D219). This is because, when the rapid acceleration / deceleration state code is set in the type code with reference to the travel locus time series TBL352, the acceleration at that time is checked, and the acceleration is divided into 20 (rapid acceleration threshold value-11 km / h). / Sec to sudden deceleration threshold +11 km / h / sec). In the program, over the range of sudden deceleration threshold + 1 to sudden acceleration threshold minus 1, it is repeatedly performed while incrementing A (a) by 1 for all accelerations whose type code is “rapid acceleration” and “rapid deceleration”. (D220, D221).
Similarly, when “steep steering wheel” is set in the type code of the travel locus time series TBL352, the C (lateral G) is incremented by +1 for all the accelerations (D223) and repeated (D222).
[0054]
Next, another calculation method for the standard deviation of acceleration will be described.
[0055]
[Expression 1]

[0056]
In addition, a description will be given of how to obtain the standard deviation of acceleration when acceleration acceleration / deceleration (sudden acceleration / deceleration) equal to or greater than a predetermined threshold for acceleration / deceleration is stored instead of the acceleration time series. An example of the calculation method in this case is shown in Formula 2.
[0057]
[Expression 2]

[0058]
Expression 2 is an approximate expression obtained empirically by least squares. The weighting factors A (1), A (2), A (3), and A (4) in the equation here are obtained from the accelerations taken every second (calculated from the initial speed and the final speed). (The same applies below). Specifically, the acceleration every 1 second is sampled (calculated) in advance, and the true value of the acceleration dispersion obtained by Equation 1 based on the value, and a value equal to or greater than the threshold value determined for acceleration / deceleration from the acceleration time-series data. Take out the data as sudden acceleration data and sudden deceleration data, and calculate the values of sudden acceleration / deceleration frequency, sudden acceleration average, sudden deceleration average, sudden acceleration square average, and sudden deceleration square average from these data. Weighting factors A (1), A (2), A (3), A (4) are obtained in advance. The weighting factors A (1), A (2), A (3), and A (4) may be obtained by performing function approximation using a neural network.
[0059]
Next, another calculation method for the degree of acceleration distortion will be described. This calculation is shown in Equation 3.
[0060]
[Equation 3]

[0061]
Here, the acceleration standard deviation is a value calculated by Equation 1. In addition, a description will be given of how to obtain the degree of distortion of acceleration in the case of storing acceleration that is not a time series of acceleration but is equal to or greater than a predetermined acceleration / deceleration threshold.
[0062]
An example of the calculation method in this case is shown in Formula 4.
[0063]
[Expression 4]

[0064]
Equation 4 is an approximate expression obtained empirically by least squares. The weighting factors B (0), B (1), B (2), B (3), B (4), and B (5) in the formula here are obtained from the accelerations collected every second in advance. Keep it. Specifically, the acceleration every 1 second is sampled in advance, and the true value of the acceleration dispersion obtained by Equation 3 and the value above the threshold are extracted from the acceleration time series based on the value, and the rapid acceleration / deceleration frequency, rapid acceleration average, sudden acceleration Average values of deceleration average, rapid acceleration square average, rapid deceleration square average, sudden acceleration cube average, and sudden deceleration cube average are calculated, and weight coefficients B (0), B (1), B ( 2), B (3), B (4), B (5) are obtained in advance. The weighting factors B (0), B (1), B (2), B (3), B (4), and B (5) may be obtained by performing function approximation using a neural network.
[0065]
FIG. 18 is a PAD representation of the break law violation information processing. The law requires a 30-minute break every 4 hours of continuous operation.
Therefore, the travel interruption time and the travel start time after the travel start in the traveling vehicle 10 are taken in, the continuous operation time is calculated from the difference, and the break time is calculated from the difference between the work start time and the break time within the continuous operation time. Thus, it is determined whether or not there is a violation of the law recognized by comparing the break time within the continuous operation time with the break time specified by the break law. If there is a violation of the law, the number of days (number of times) may be counted and used as a diagnostic item.
[0066]
Specifically, FIG. 18A shows a method for calculating the continuous operation time. The break law violation information processing unit 325 first sets the maximum continuous operation time to 0 (D231). Then, when the start code recorded in the travel trajectory time series TBL352 is searched and the start of travel of the operating vehicle is detected, loading, unloading, resting, closing work, etc., after the start of the next travel Until the travel is interrupted (D234), the travel suspension time after the start of travel is detected (D235).
Then, the continuous operation time is set as a difference between the travel interruption time after the start of travel and the travel start time (D236), and is set in the memory as the maximum continuous operation time (D237). The above operations (D233 to D237) are repeated for all data for which the start code is set in the travel locus time series TBL352.
[0067]
FIG. 18B shows a process for setting a break law violation. Here, the break law violation information processing unit 325 calculates T = t + 4 hours and 30 minutes from the start to the end, and if T exceeds the end, sets the end time (D239). Then, the rest time in T from time t is totaled (D240), and when the total value of the rest time is 30 minutes or less, it is set as a violation of law (D241, D242).
The total processing of the break time is shown in FIG. That is, the break law violation information processing unit 325 searches for a break code recorded in the travel trajectory time series TBL352 within the range of time t to T (D243) (D244), and sets the break time as the travel start time. The difference between the break times is accumulated from the break to the next run (running, automatic driving detection code) (D245, D246).
[0068]
FIG. 19 is a PAD representation of the total idling time, the number of idlings, and the idling histogram aggregation method. FIG. 9 shows the data structure of the idling histogram TBL (3561).
Here, when the engine is ON and the idling state in which the position information does not change has passed for a predetermined time, the number of times and the time are measured and set.
[0069]
Specifically, the idling total processing unit 341 first sets the total idling histogram number memory allocated to the program to 0, and sets the total idling time value to 0 (D251). When the start of idling is detected by referring to the travel locus time series TBL352 (D253), the idling time is obtained as the difference between the idling end time and the idling start time until the next idling end (D254) (D255). And the idling histogram count memory is incremented by +1 from the idling time (D256).
Then, the idling total processing unit 341 sets the idling count as the sum of the idling histogram count memory (D257).
In addition, the fuel-saving driving diagnosis by the fuel-saving driving diagnosis content generation unit 34 includes the longitudinal acceleration average value recorded in the acceleration statistical information TBL creation unit 322, the average speed recorded in the physical data TBL354, and the travel distance. This is done by statistically processing the actual fuel consumption rate input by operating the management terminal 12 by the labor manager of the carrier 1. Incidentally, the in-vehicle terminal 11 of the present embodiment functions in a state where the engine is ON, and transmits vehicle speed information and position information. For this reason, if the driver stops the engine with the ignition switch, the idling time is not measured.
[0070]
Next, the evaluation of the safe driving diagnosis and the calculation method of the degree will be described. The degree of safe driving diagnosis is set to be inversely proportional to the probability of causing a traffic accident. If the driver's accident history (how many accidents have occurred per year, hereinafter referred to as accident frequency) is obtained in advance, this information and information on acceleration standard deviation and acceleration skewness pairs are collected. Then, a relational expression for explaining the accident frequency with the acceleration standard deviation and the value of the acceleration distortion is obtained.
Thereby, if the acceleration standard deviation and the acceleration skewness are measured, the frequency of the accident can be estimated. Here, the acceleration standard deviation is a physical quantity that is easily affected by the traveling area, but also includes the physical quantity because it reflects the driver's personality to some extent. An example of this relationship is the following method according to Equation 5.
[0071]
[Equation 5]

[0072]
Here, the weighting factors C (0), C (1), and C (2) in this equation are calculated based on the information on the driver's accident history, acceleration standard deviation, and acceleration skewness pair collected in advance. The historical data is regarded as the accident frequency, and is obtained by the least square. Further, the reciprocal of the accident frequency in Formula 5 is defined as the degree of safe driving. Further, here, instead of the least squares, it is also possible to use a neural network to obtain a function approximation using an accident frequency as an output, an acceleration standard deviation, and an acceleration distortion as an input. Further, when it is difficult to obtain an accident history, the above-described (Equation 5) is used based on the acceleration standard deviation and acceleration skewness, using sudden deceleration, which is likely to lead to an accident, as data instead of the accident frequency. Similarly, the accident frequency is estimated.
[0073]
Next, a method for improving the accuracy of the degree of safe driving will be described. This accuracy is determined by learning using the least squares of Equation 5 or a neural network. For this reason, the accuracy improves as the number of data of the subject increases. Therefore, relearning may be performed by using a group of acceleration skewness and standard deviation values stored in the safe driving / fuel-saving driving diagnosis server 30 of FIG.
[0074]
In addition, when there is no learning data at all, as a simple method, the following is set as an example using threshold values obtained by analyzing acceleration time-series data of traveling data of a plurality of subjects in advance. . When the acceleration standard deviation is xkm / h / sec or more and the absolute value of the acceleration distortion is y or more, the degree of safe driving is set to zero. In other cases, the degree of safe driving is simplified to 1.
[0075]
Finally, the warning screen and the message contents are described. Based on the distribution of driver's driving acceleration, a warning is given when the degree of safe driving evaluated by evaluating the driver's personality is equal to or greater than a preset threshold. Here, the threshold value may be one or more. In the present embodiment, five levels are prepared, and four threshold values serving as boundaries between the levels are prepared. The names A, B, C, D, and E are assigned in order from the safe level.
In addition, in order to determine the bias of the acceleration acceleration / deceleration distribution, the acceleration skewness is calculated, and if the acceleration skewness has a large value on the deceleration side, the driver is accelerated. If it is determined that the deceleration is sudden and the acceleration side has a large value, the driver determines that the acceleration is rapid and the deceleration is slow. For example, when the acceleration skewness is −y or less, the acceleration skewness is divided into three cases when y or more. When the acceleration skewness is −y or less, when it is y or more, a diagnostic message is prepared by combining the three combinations in other cases and a safe level combination.
[0076]
Here, when the acceleration skewness is −y, it means that the base of the deceleration of the acceleration distribution is wider than the acceleration, and therefore the message “Brake is tight” is inserted. When the acceleration skewness is greater than or equal to y, it means that the acceleration base of the acceleration distribution is wider than the deceleration, so the message “Accel is tight” is inserted.
Further, when the acceleration distortion exceeds -y and is less than y, it means that the distribution of acceleration is nearly symmetrical, so that a message that the accelerator is tight and the brake is tight cannot be entered. Also, when the safety level is A, “I am driving safely”, when I is B, “I am safe driving”, and when I is “normal driving” In the case of D, the message “I am driving somewhat rough” is entered, and in the case of E, the message “I am driving rough” is entered. The message when only two safety levels are set is displayed only for levels A and E.
[0077]
That is, when the degree of safe driving is on the safe side (when the value is larger) than the threshold value, a message is issued from the level A message group according to the skewness value. On the other hand, if it is on the danger side (when the value is smaller) than the threshold, a message is issued from the level E message group according to the skewness value.
This diagnostic message can be changed by the administrator of the safe driving service center.
[0078]
Next, how to determine the threshold when the safety level is divided into five stages A to E will be described. This threshold value is obtained by collecting driver subjects in advance and classifying subjective safety levels. Then, an average value of the degree of safe driving at each safety level is obtained. An intermediate value of the average value for each safety level is set as a threshold value for each safety level. For example, assume that the level A level of safe driving is 100 and the level B level of safe driving is 80. In this case, the threshold values of levels A and B are 90.
[0079]
A method for determining the threshold when there are two safety levels will be described. Here, when the degree of safety is calculated based on the frequency of traffic accidents, it is set to the reciprocal of the frequency of traffic accidents of the average driver. When the degree of safety is calculated based on the number of sudden decelerations, the reciprocal of the average driver's sudden deceleration frequency is used. The average driver here uses the average of drivers in Japan or the average value of drivers stored in the safe driving / fuel-saving driving diagnosis server 30.
[0080]
Here, an example of a form created according to the present invention will be described with reference to FIGS.
According to the daily form shown in FIG. 20, each diagnosis item is comprehensively judged and a comment for guidance is attached, and the degree of safety and fuel-saving driving is scored and expressed (A). Moreover, each diagnostic item value of a driving day is described and the improvement degree of driving | operation can be grasped | ascertained by comparing with the average value of the last month (B).
Further, a travel route for one operation (starting to closing) is shown. Here, by marking a place where dangerous driving was performed in addition to the work place, it is possible to identify an area where these frequently occur, and to use it for guidance (C). A speed change graph is also displayed. By marking the time zone of driving with speed violation and speed irregularity, it is possible to grasp the time zone in which these are likely to occur and to use for guidance (D).
[0081]
The form shown in FIG. 21 is a monthly form (monthly average evaluation of the driver for one month and driving information history for one week), and a form is created for a particularly important item among the diagnosis items, and the month of the driver is prepared. Used to report average driving safety and fuel economy trends.
Here, the change in the value of each diagnostic item during one month is displayed in a graph. Here, when it seems that it is an unusual driving | operation, a detail can be confirmed with the daily report shown in FIG. 20 (E). Moreover, the average value of each diagnostic item value for one month is recorded. Moreover, the tendency can be clearly shown by displaying the value of the previous month (F).
[0082]
The diagnosis items are: driver, travel date and time, operation time [h], handle time [h], maximum continuous handle time [h], travel time [h], idling time [h], break time [h] , Work time [h], mileage [km], safe driving points, fuel-saving driving points, instruction comment, speed [km / h], average speed [km / h], instantaneous maximum speed [km / h], speed Violation time ratio [%], acceleration, number of sudden acceleration / deceleration [times / h], number of near-misses [times / h], maximum acceleration / deceleration [G], acceleration / deceleration distribution [times / h], acceleration / deceleration tendency distribution.
[0083]
The driver comparison data shown in FIG. 22 shows the driving distance and driving time in comparison with the drivers A, B, C, and D, and the average operating time and driving distance per operation are as follows: It is almost the same before and after the safe driving diagnosis. The driver comparison data shown in FIG. 23 shows the evaluation regarding the speed, and Mr. B and Mr. D can confirm that the rate of speed violations decreases before and after the instruction.
[0084]
The driver comparison data shown in FIG. 24 shows an evaluation regarding sudden acceleration / deceleration before and after safety guidance. Here, Mr. A and Mr. B have a slight rapid acceleration / deceleration tendency, and Mr. C and Mr. D can confirm that the tendency of sudden deceleration has decreased as a result of the instruction. Further, the driver comparison data shown in FIG. 25 shows an evaluation regarding driving tendency before and after safety guidance. Here, the closer to the origin, the more stable the driver (constant speed), and no unnecessary acceleration (accelerator work is stable). Mr. D is recognized to improve both acceleration and accelerator work through safety guidance. Supplementally, in FIG. 25, the driver tends to accelerate more rapidly as the data is higher, and the driver tends to decelerate more rapidly as the data is lower. In addition, the driver comparison data shown in FIG. 26 shows the evaluation (idling time, fuel consumption) regarding the fuel consumption before and after the guidance. Here, since the refrigerated vehicle is exemplified as the operating vehicle, there is no improvement in the idling frequency due to the special circumstances (the idling stop is not possible due to the movement of the refrigerator), but the improvement by guidance is not possible in normal vehicles It was seen. In addition, all of the fuel efficiency has been improved.
[0085]
The driver comparison data shown in FIGS. 22 to 26 can be used as corporate data. In other words, by comparing Mr. A as the in-house A group and Mr. B as the in-house B group, it is possible to have a sense of competition in the company regarding safe driving and fuel-saving driving, and to activate the activities. Of course, it is possible to create a form for comparing a group and individuals within the group side by side. As a result, the safe driving manager can obtain various forms with a minimum of labor, and can efficiently provide guidance. Of course, the company (company) may be regarded as one group so as to handle the data so that the companies can be compared.
[0086]
As described above, the present embodiment automatically creates a rating and a form that match a driver by centrally managing physical data and labor data acquired from a vehicle and analyzing its statistical properties. In addition, when a safety instructor performs safe driving guidance or fuel-saving driving guidance with reference to the above-described centrally managed data, the content can be easily created to reduce the burden on the safety instructor.
In addition, the safe driving diagnosis content generation unit 32, the form output unit 33, the fuel-saving driving diagnosis content generation unit 34, the speed violation information TBL generation unit 321, the acceleration statistical information TBL generation unit 322, the number of sudden handles and the maximum handle shown in FIG. The procedure executed by each of the acceleration calculation unit 323, the acceleration standard deviation / distortion calculation unit 324, the break law violation information processing unit 325, and the idling aggregation processing unit 341 is recorded on a computer-readable recording medium. It is assumed that the safe driving and fuel-saving driving diagnosis device of the present invention is realized by causing the computer system to read and execute the recorded program. The computer system here includes an OS and hardware such as peripheral devices.
[0087]
In implementing the present invention, the acceleration sensor is not an essential sensor, and the acceleration may be calculated by calculation based on vehicle speed information or vehicle position information (position information). For lateral G (lateral acceleration), safe driving diagnosis and fuel-saving driving diagnosis can be performed without using lateral G. Of course, the lateral G may be used to perform safe driving diagnosis and fuel-saving driving diagnosis. Further, the lateral G may be measured using a dedicated sensor, or may be obtained by calculation from position information, vehicle speed information, and the like, for example. Further, the vehicle speed pulse sensor that measures the vehicle speed is an example, and the present invention is not limited to the sensor that generates the vehicle speed pulse. Further, in-house speed regulation may be added to the speed regulation information.
Furthermore, the communication unit 115 mounted on the operating vehicle 10 is not limited to a packet, and the same effect can be obtained even if wireless communication is performed using a normal modem using a mobile phone. In the present embodiment, the GPS sensor 13 and the vehicle sensor 14 (vehicle speed) are provided in the vehicle, and the current position and the vehicle speed of the operating vehicle 10 are measured. However, depending on the performance of the GPS sensor 13, the vehicle sensor 14 may be used. It can be unnecessary. Thus, the safe driving diagnosis and the fuel-saving driving diagnosis are realized while the hardware (including the in-vehicle terminal 11) mounted on the operating vehicle 10 has a simple configuration. Incidentally, since the vehicle speed pulse sensor and the GPS sensor 13 for measuring the vehicle speed are widely used, the financial burden on the user and the service provider is reduced when performing such a safe driving diagnosis and the like.
[0088]
【The invention's effect】
As described above, according to the present invention, a rating or a form that matches a driver can be automatically created. In addition, when a safety instructor refers to data to be managed (unified management) and provides safe driving guidance or fuel consumption guidance, the content can be easily created, so the burden on the safety instructor for that purpose should be reduced. Can do. In addition, the result of the safe driving diagnosis is appropriately notified or the safety guidance leads to enlightenment of the driver's safe driving, and a reduction in traffic accidents is expected.
[Brief description of the drawings]
FIG. 1 is a diagram quoted for clarifying a relationship between a business model for realizing a safe driving diagnosis method and a fuel-saving driving diagnosis method of the present invention, and its user.
FIG. 2 is a diagram cited for explaining the configuration of an in-vehicle terminal mounted on a running vehicle used in the present invention.
3 is a block diagram showing an internal configuration of the in-vehicle terminal shown in FIGS. 1 and 2. FIG.
4 is a block diagram showing an expanded function of the internal configuration of the safe driving / fuel-saving driving diagnosis server shown in FIG. 1. FIG.
FIG. 5 is a diagram cited for explaining data structures of various tables of the vehicle DB used in the embodiment of the present invention.
FIG. 6 is a matrix representation of the relationship between the type code of the travel locus time series TBL used in the embodiment of the present invention and recorded information.
FIG. 7 is a diagram cited for explaining the data structure of speed violation information TBL used in the embodiment of the present invention.
FIG. 8 is a diagram cited for explaining the data structure of acceleration statistical information TBL used in the embodiment of the present invention.
FIG. 9 is a diagram cited for explaining a data structure of an idling histogram TBL used in the embodiment of the present invention.
FIG. 10 is an operation sequence diagram cited for explaining the operation of the embodiment of the present invention.
FIG. 11 is an operation sequence diagram cited for explaining the operation of the embodiment of the present invention;
FIG. 12 is a diagram showing an example of a GUI screen for form output used in the embodiment of the present invention.
FIG. 13 is a diagram showing a PAD representation of a partial structure of the safe driving diagnosis program of the present invention.
FIG. 14 is a diagram showing a partial structure of the safe driving diagnosis program of the present invention in PAD expression.
FIG. 15 is a diagram showing a PAD representation of a partial structure of the safe driving diagnosis program of the present invention.
FIG. 16 is a diagram showing a PAD representation of a partial structure of the safe driving diagnosis program of the present invention.
FIG. 17 is a diagram showing a PAD representation of a partial structure of the safe driving diagnosis program of the present invention.
FIG. 18 is a diagram showing a partial structure of the safe driving diagnosis program of the present invention in PAD expression.
FIG. 19 is a diagram showing a partial structure of the fuel-saving driving diagnosis program of the present invention in PAD expression.
FIG. 20 is a diagram showing an example of a format of a form output by the safe driving diagnosis apparatus of the present invention.
FIG. 21 is a diagram showing an example of a format of a form output by the safe driving diagnosis apparatus of the present invention.
FIG. 22 is a diagram showing an example of a format of a form output by the safe driving diagnosis apparatus of the present invention.
FIG. 23 is a diagram showing an example of a format of a form output by the safe driving diagnosis apparatus of the present invention.
FIG. 24 is a diagram showing an example of a format of a form output by the safe driving diagnosis apparatus of the present invention.
FIG. 25 is a diagram showing an example of a format of a form output by the safe driving diagnosis apparatus of the present invention.
FIG. 26 is a diagram showing an example of a format of a form output by the fuel-saving driving diagnosis device of the present invention.
[Explanation of symbols]
DESCRIPTION OF SYMBOLS 1 ... Transporter, 2 ... Data collection company, 3 ... Safe driving guidance business operator, 4, 5, 6 ... Communication network, 10 ... Operation vehicle, 11 ... In-vehicle terminal, 12 ... Management terminal, 20 ... Data collection server, DESCRIPTION OF SYMBOLS 30 ... Safe driving / fuel-saving driving diagnosis server (safe driving diagnostic device, fuel-saving driving diagnostic device), 31 ... communication control unit, 32 ... safe driving diagnosis content generation unit, 33 ... form output unit, 34 ... fuel-saving driving diagnosis Content generation unit, 35 ... vehicle DB

Claims (6)

A safe driving diagnosis method using a safe driving diagnosis device that performs a safe driving diagnosis on a vehicle driving situation of a driver,
The safe driving diagnosis device is:
Predetermined recording time (T) of vehicle acceleration α (t) calculated according to the difference between the initial speed and the final speed of the vehicle speed information acquired at predetermined time intervals from the in-vehicle terminal via the communication network was stored in the storage unit, the acceleration alpha (t) of the vehicle read from the storage unit to calculate an acceleration skewness indicating the deviation of the deceleration distribution by applying the equation (1),
Or
Predetermined recording time (T) of vehicle acceleration α (t) calculated according to the difference obtained by subtracting the initial speed from the final speed of the vehicle speed information acquired at predetermined time intervals from the in-vehicle terminal via the communication network Minutes are stored in the storage unit, and when the vehicle acceleration α (t) read from the storage unit is equal to or greater than the first predetermined acceleration value, the number of sudden accelerations is increased by 1 within a predetermined recording time (T). The number of sudden accelerations is calculated and the calculated number of sudden accelerations is divided by T to obtain the sudden acceleration frequency. When the acceleration α (t) is equal to or less than a second predetermined acceleration value, the number of sudden decelerations is increased by one and predetermined The number of sudden decelerations within the recording time (T) is calculated and the number of sudden decelerations calculated is divided by T to obtain the sudden deceleration frequency, and an average value for acceleration α (t) equal to or greater than a first predetermined acceleration value, The average value of the squared value and the average value of the squared value are the sudden acceleration average and the sudden acceleration squared respectively. The absolute value of the average value for the acceleration α (t) below the second predetermined acceleration, the average value of the squared value, the absolute value of the average value of the squared value, and the absolute value of the average value of the squared power , And calculate the acceleration skewness indicating the bias of the acceleration / deceleration distribution by applying to the equation (2), respectively,
The calculated acceleration distortion is compared with a threshold value indicating a preset upper limit value and lower limit value, and a message indicating the driving tendency of the driver corresponding to the magnitude comparison result is read from the storage unit and output. thing,
A safe driving diagnosis method characterized by the above.
However,
Equation (1) is

Equation (2) is

The safe driving diagnosis device is:
As a message indicating the driving tendency of the driver corresponding to the magnitude comparison result, when the acceleration distortion degree is equal to or greater than the upper limit value, a message relating to rapid acceleration is read out from the storage unit and output, and the acceleration distortion is If the lower limit value or less, read and output a message related to rapid deceleration from the storage unit,
The safe driving diagnosis method according to claim 1. The safe driving diagnosis device is:
Furthermore, the standard deviation of the acceleration distribution is calculated,
The acceleration strain and the standard deviation of the acceleration distribution are respectively multiplied by weighting factors and added, and a predetermined value is added to calculate the accident frequency, and the reciprocal of the accident frequency is calculated as the degree of safe driving. ,
Outputting the magnitude of the degree of safe driving in points,
The safe driving diagnosis method according to claim 1, wherein: A safe driving diagnosis device for performing a safe driving diagnosis on a vehicle driving situation of a driver
Predetermined recording time (T) of vehicle acceleration α (t) calculated according to the difference between the initial speed and the final speed of the vehicle speed information acquired at predetermined time intervals from the in-vehicle terminal via the communication network was stored in the storage unit, the acceleration alpha (t) of the vehicle read from the storage unit to calculate an acceleration skewness indicating the deviation of the deceleration distribution by applying the equation (1),
Or
Predetermined recording time (T) of vehicle acceleration α (t) calculated according to the difference obtained by subtracting the initial speed from the final speed of the vehicle speed information acquired at predetermined time intervals from the in-vehicle terminal via the communication network Minutes are stored in the storage unit, and when the vehicle acceleration α (t) read from the storage unit is equal to or greater than the first predetermined acceleration value, the number of sudden accelerations is increased by 1 within a predetermined recording time (T). The number of sudden accelerations is calculated and the calculated number of sudden accelerations is divided by T to obtain the sudden acceleration frequency. When the acceleration α (t) is equal to or less than a second predetermined acceleration value, the number of sudden decelerations is increased by one and predetermined The number of sudden decelerations within the recording time (T) is calculated and the number of sudden decelerations calculated is divided by T to obtain the sudden deceleration frequency, and an average value for acceleration α (t) equal to or greater than a first predetermined acceleration value, The average value of the squared value and the average value of the squared value are the sudden acceleration average and the sudden acceleration squared respectively. The absolute value of the average value for the acceleration α (t) below the second predetermined acceleration, the average value of the squared value, the absolute value of the average value of the squared value, and the absolute value of the average value of the squared power , And calculate the acceleration skewness indicating the bias of the acceleration / deceleration distribution by applying to the equation (2), respectively,
The calculated acceleration distortion is compared with a threshold value indicating a preset upper limit value and lower limit value, and a message indicating the driving tendency of the driver corresponding to the magnitude comparison result is read from the storage unit and output. thing,
A safe driving diagnostic device.
However,
Equation (1) is

Equation (2) is

As a message showing the driving tendency of the driver corresponding to the previous SL magnitude comparison result, if the acceleration skewness is equal to or greater than the upper limit value, and outputs read a message relating to rapid acceleration from the storage unit, the acceleration strain Is less than or equal to the lower limit value, a message relating to sudden deceleration is read out from the storage unit and output,
The safe driving diagnostic apparatus according to claim 4. Furthermore, the standard deviation of the acceleration distribution is calculated,
The acceleration strain and the standard deviation of the acceleration distribution are respectively multiplied by weighting factors and added, and a predetermined value is added to calculate the accident frequency, and the reciprocal of the accident frequency is calculated as the degree of safe driving. ,
Outputting the magnitude of the degree of safe driving in points,
6. The safe driving diagnosis apparatus according to claim 4 or 5, wherein:

Images