JP4729411B2

JP4729411B2 - Traffic jam prediction information generation device, traffic jam prediction information generation method, and route search system

Info

Publication number: JP4729411B2
Application number: JP2006042248A
Authority: JP
Inventors: 修増谷; 弘利岩崎; 宏佐々木; 真一本位田; 靖志安藤
Original assignee: Inter University Research Institute Corp Research Organization of Information and Systems; Denso IT Laboratory Inc
Current assignee: Inter University Research Institute Corp Research Organization of Information and Systems; Denso IT Laboratory Inc
Priority date: 2006-02-20
Filing date: 2006-02-20
Publication date: 2011-07-20
Anticipated expiration: 2026-02-20
Also published as: JP2007219990A

Description

本発明は、車両が目的地へ行く過程における道路の渋滞予測を行う渋滞予測情報生成装置、渋滞予測情報生成方法、及び経路探索システムに関する。 The present invention relates to a traffic jam prediction information generation device, a traffic jam prediction information generation method, and a route search system that perform road traffic jam prediction in the process of a vehicle going to a destination.

従来から、最新かつ新鮮な信頼度の高い情報に基づいて、目的地までスムーズに快適に車両が走行できるように道路の渋滞予測を行う技術が存在している。このような技術は下記の特許文献１に記載されている。特許文献１に記載された技術は、時間の経過に伴って減退する情報（発生フェロモン）を用いて最新かつ新鮮な信頼度の高い渋滞予測を行うように構成されている。また、下記の非特許文献１及び２では、特許文献１に記載された技術の他にフェロモンの概念などが記載されている。
特開２００５−２３５１９４号公報（要約） “Traffic Prediction using Pheromone Model”、Osamu MASUTANI,Hiroshi SASAKI,Hirotoshi IWASAKI,Yasushi ANDO,Yoshiaki FUKAZAWA,Shinichi HONIDEN 12th World Congress on ITS,6-10 November 2005,San Francisco “Return From The Ant”（http://dochost.rz.huberlin.de/dissertationen/brueckner-sven-2000-06-21/PDF/Brueckner.pdf） 2. Description of the Related Art Conventionally, there is a technology for predicting traffic congestion on a road so that a vehicle can smoothly and comfortably travel to a destination based on the latest and fresh information with high reliability. Such a technique is described in Patent Document 1 below. The technique described in Patent Document 1 is configured to perform the latest and fresh, highly reliable traffic jam prediction using information (generated pheromone) that decreases with the passage of time. Non-Patent Documents 1 and 2 below describe the concept of pheromones in addition to the technique described in Patent Document 1.
JP 2005-235194 A (summary) “Traffic Prediction using Pheromone Model”, Osamu MASUTANI, Hiroshi SASAKI, Hirotoshi IWASAKI, Yasushi ANDO, Yoshiaki FUKAZAWA, Shinichi HONIDEN 12th World Congress on ITS, 6-10 November 2005, San Francisco “Return From The Ant” (http://dochost.rz.huberlin.de/dissertationen/brueckner-sven-2000-06-21/PDF/Brueckner.pdf)

しかしながら、上述した特許文献１に記載された技術では、生成される渋滞予測に用いる情報が発生フェロモンのみに依存しているため、高い精度の情報を生成することが困難である。また、渋滞予測では中長期的な傾向を捕らえるため、フェロモンを算出する際のパラメータの動的な対応が必要であるが、従来の技術では適切に対応することができないという問題があった。 However, in the technique described in Patent Document 1 described above, it is difficult to generate highly accurate information because the information used for predicting the traffic jam depends only on the generated pheromone. Moreover, in order to capture medium- to long-term trends in traffic jam prediction, it is necessary to dynamically cope with parameters when calculating pheromones, but there is a problem that conventional techniques cannot cope with them appropriately.

本発明は、上記問題を解決するためのものであり、パラメータの動的な対応ができ、より精度の高い渋滞予測のための情報を生成することができる渋滞予測情報生成装置、渋滞予測情報生成方法、及び経路探索システムを提供することを目的とする。 The present invention is intended to solve the above-described problem, and is capable of dynamically responding to parameters, and can generate information for predicting traffic congestion with higher accuracy. It is an object to provide a method and a route search system.

上記目的を達成するために、本発明によれば、車両が走行する走行路の所定の区間ごとの混雑程度を示す交通混雑程度情報を生成する渋滞予測情報生成装置であって、前記所定の区間ごとの前記走行路及び前記走行路周辺の交通量の情報及び前記所定の区間ごとの前記走行路に関する情報を含む地図情報を受信する受信手段と、受信された前記走行路及び前記走行路周辺の交通量の情報及び前記地図情報に基づいて、前記所定の区間ごとに車両のその時点の存在量を取得し、取得された前記存在量を含む時間関数による時間関数存在量を前記所定の区間ごとに算出し、算出された前記時間関数存在量を含む時間関数による時間の経過に伴って隣接する区間へ伝達する時間関数伝達量を前記所定の区間ごとに算出し、算出された前記時間関数存在量及び前記時間関数伝達量と、現時点における以前に算出された前記交通混雑程度情報と、隣接する区間のうち下流側に位置する区間での前記時間関数存在量がある閾値を超えた場合に生じる、前記時間の経過に伴って隣接する区間へ伝達する前記時間関数伝達量の前記車両の走行方向と逆方向の前記下流側に位置する区間からの時間関数伝達量、及び前記走行路に設置された信号機の信号サイクルの赤信号の長さに応じて生じる時間関数切替量の２つのうち少なくとも一方とに基づいて、所定の時間経過後の前記所定の区間ごとの混雑程度を示す前記交通混雑程度情報を渋滞予測情報として生成する生成手段とを備える渋滞予測情報生成装置が提供される。この構成により、入力データに対する要求を抑え、また予測をする際の計算量を抑え、経路探索のための信頼度のより高い渋滞予測の情報を生成することができる。 In order to achieve the above object, according to the present invention, there is provided a traffic jam prediction information generating device for generating traffic congestion degree information indicating a degree of congestion for each predetermined section of a traveling path on which a vehicle travels. Receiving means for receiving map information including information on the travel route for each and the traffic volume around the travel route and information on the travel route for each predetermined section; and the received travel route and the vicinity of the travel route Based on the traffic volume information and the map information, the vehicle's current abundance is acquired for each predetermined section, and the time function abundance based on the time function including the acquired abundance is determined for each predetermined section. A time function transfer amount to be transmitted to an adjacent section as time passes by the time function including the calculated time function existence amount for each predetermined section, and the calculated time function existence is calculated. Occurs when it exceeds the amount and the time function transfer amount, and the traffic congestion degree information calculated previously at the present time, a certain threshold the time function abundance of a section located downstream of the adjacent section A time function transmission amount from a section located on the downstream side in a direction opposite to the traveling direction of the vehicle, and a time function transmission amount transmitted to an adjacent section with the passage of time, and installed in the traveling path. The traffic congestion degree indicating the degree of congestion for each predetermined section after a predetermined time has elapsed based on at least one of two time function switching amounts generated according to the length of the red signal of the signal cycle of the traffic light There is provided a traffic jam prediction information generation device comprising a generation means for generating information as traffic jam prediction information. With this configuration, it is possible to reduce the demand for input data, reduce the amount of calculation for prediction, and generate more reliable traffic jam prediction information for route search.

また、本発明によれば、車両が走行する走行路の所定の区間ごとの混雑程度を示す交通混雑程度情報を生成する渋滞予測情報生成装置であって、前記所定の区間ごとの前記走行路及び前記走行路周辺の交通量の情報及び前記所定の区間ごとの前記走行路に関する情報を含む地図情報を受信する受信手段と、受信された前記走行路及び前記走行路周辺の交通量の情報及び前記地図情報に基づいて、前記所定の区間ごとに車両のその時点の存在量を取得し、取得された前記存在量を含む時間関数による時間関数存在量を前記所定の区間ごとに算出し、算出された前記時間関数存在量を含む時間関数による時間の経過に伴って隣接する区間へ伝達する時間関数伝達量を前記所定の区間ごとに算出し、算出された前記時間関数存在量及び前記時間関数伝達量と、現時点における以前に算出された前記交通混雑程度情報と、隣接する区間のうち下流側に位置する区間での前記時間関数存在量がある閾値を超えた場合に生じる、前記時間の経過に伴って隣接する区間へ伝達する前記時間関数伝達量の前記車両の走行方向と逆方向の前記下流側に位置する区間からの時間関数伝達量とに基づいて、所定の時間経過後の前記所定の区間ごとの混雑程度を示す前記交通混雑程度情報を渋滞予測情報として生成する生成手段とを備える渋滞予測情報生成装置が提供される。この構成により、入力データに対する要求を抑え、また予測をする際の計算量を抑え、経路探索のための信頼度のより高い渋滞予測の情報を生成することができる。 According to the present invention, there is also provided a traffic jam prediction information generating device that generates traffic congestion degree information indicating a degree of congestion for each predetermined section of a travel path on which the vehicle travels, the traffic path for each predetermined section, Receiving means for receiving map information including information on the traffic volume around the travel path and information on the travel path for each of the predetermined sections; the received information on the travel path and the traffic volume around the travel path; and Based on the map information, the abundance at that time of the vehicle is acquired for each predetermined section, and the time function abundance by the time function including the acquired abundance is calculated for each predetermined section. In addition, the time function transmission amount to be transmitted to adjacent sections as time passes by the time function including the time function existence amount is calculated for each of the predetermined sections, and the calculated time function existence amount and the time function are calculated. And Itaruryou, and the traffic congestion degree information calculated previously at the present time, caused when exceeding a certain threshold the time function abundance of a section located downstream of the adjacent sections, the course of the time The time function transmission amount transmitted to the adjacent section along with the time function transmission amount from the section located on the downstream side in the direction opposite to the traveling direction of the vehicle, the predetermined function after a predetermined time has elapsed. There is provided a traffic jam prediction information generating apparatus including a generation unit that generates the traffic congestion level information indicating the traffic level of each section as traffic jam prediction information. With this configuration, it is possible to reduce the demand for input data, reduce the amount of calculation for prediction, and generate more reliable traffic jam prediction information for route search.

また、本発明によれば、車両が走行する走行路の所定の区間ごとの混雑程度を示す交通混雑程度情報を生成する渋滞予測情報生成方法であって、前記所定の区間ごとの前記走行路及び前記走行路周辺の交通量の情報及び前記所定の区間ごとの前記走行路に関する情報を含む地図情報を受信するステップと、受信された前記走行路及び前記走行路周辺の交通量の情報及び前記地図情報に基づいて、前記所定の区間ごとに車両のその時点の存在量を取得し、取得された前記存在量を含む時間関数による時間関数存在量を前記所定の区間ごとに算出し、算出された前記時間関数存在量を含む時間関数による時間の経過に伴って隣接する区間へ伝達する時間関数伝達量を前記所定の区間ごとに算出し、算出された前記時間関数存在量及び前記時間関数伝達量と、現時点における以前に算出された前記交通混雑程度情報と、隣接する区間のうち下流側に位置する区間での前記時間関数存在量がある閾値を超えた場合に生じる、前記時間の経過に伴って隣接する区間へ伝達する前記時間関数伝達量の前記車両の走行方向と逆方向の前記下流側に位置する区間からの時間関数伝達量、及び前記走行路に設置された信号機の信号サイクルの赤信号の長さに応じて生じる時間関数切替量の２つのうち少なくとも一方とに基づいて、所定の時間経過後の前記所定の区間ごとの混雑程度を示す前記交通混雑程度情報を渋滞予測情報として生成するステップとを有する渋滞予測情報生成方法が提供される。この構成により、入力データに対する要求を抑え、また予測をする際の計算量を抑え、経路探索のための信頼度のより高い渋滞予測の情報を生成することができる。 According to the present invention, there is also provided a traffic jam prediction information generation method for generating traffic congestion degree information indicating a degree of congestion for each predetermined section of a travel path on which the vehicle travels, wherein the travel path for each predetermined section and Receiving map information including information on the traffic volume around the travel path and information on the travel path for each of the predetermined sections; the received information on the travel path and the traffic volume around the travel path; and the map Based on the information, the amount of the vehicle present at that time is obtained for each predetermined section, and the time function existence amount by the time function including the obtained amount is calculated for each of the predetermined sections. A time function transmission amount that is transmitted to an adjacent section as time passes by the time function including the time function existence amount is calculated for each predetermined section, and the calculated time function existence amount and the time function are calculated. And Itaruryou, and the traffic congestion degree information calculated previously at the present time, caused when exceeding a certain threshold the time function abundance of a section located downstream of the adjacent sections, the course of the time The amount of time function transmitted from the section located on the downstream side in the direction opposite to the traveling direction of the vehicle, and the signal cycle of the traffic signal installed on the traveling path. The traffic congestion degree information indicating the degree of congestion for each of the predetermined sections after the elapse of a predetermined time based on at least one of the two time function switching amounts generated according to the length of the red signal of the traffic congestion prediction information A method for generating traffic jam prediction information is provided. With this configuration, it is possible to reduce the demand for input data, reduce the amount of calculation for prediction, and generate more reliable traffic jam prediction information for route search.

また、本発明によれば、車両が走行する走行路の所定の区間ごとの混雑程度を示す交通混雑程度情報を生成する渋滞予測情報生成方法であって、前記所定の区間ごとの前記走行路及び前記走行路周辺の交通量の情報及び前記所定の区間ごとの前記走行路に関する情報を含む地図情報を受信するステップと、受信された前記走行路及び前記走行路周辺の交通量の情報及び前記地図情報に基づいて、前記所定の区間ごとに車両のその時点の存在量を取得し、取得された前記存在量を含む時間関数による時間関数存在量を前記所定の区間ごとに算出し、算出された前記時間関数存在量を含む時間関数による時間の経過に伴って隣接する区間へ伝達する時間関数伝達量を前記所定の区間ごとに算出し、算出された前記時間関数存在量及び前記時間関数伝達量と、現時点における以前に算出された前記交通混雑程度情報と、隣接する区間のうち下流側に位置する区間での前記時間関数存在量がある閾値を超えた場合に生じる、前記時間の経過に伴って隣接する区間へ伝達する前記時間関数伝達量の前記車両の走行方向と逆方向の前記下流側に位置する区間からの時間関数伝達量とに基づいて、所定の時間経過後の前記所定の区間ごとの混雑程度を示す前記交通混雑程度情報を渋滞予測情報として生成するステップとを有する渋滞予測情報生成方法が提供される。この構成により、入力データに対する要求を抑え、また予測をする際の計算量を抑え、経路探索のための信頼度のより高い渋滞予測の情報を生成することができる。 According to the present invention, there is also provided a traffic jam prediction information generation method for generating traffic congestion degree information indicating a degree of congestion for each predetermined section of a travel path on which the vehicle travels, wherein the travel path for each predetermined section and Receiving map information including information on the traffic volume around the travel path and information on the travel path for each of the predetermined sections; the received information on the travel path and the traffic volume around the travel path; and the map Based on the information, the amount of the vehicle present at that time is obtained for each predetermined section, and the time function existence amount by the time function including the obtained amount is calculated for each of the predetermined sections. A time function transmission amount that is transmitted to an adjacent section as time passes by the time function including the time function existence amount is calculated for each predetermined section, and the calculated time function existence amount and the time function are calculated. And Itaruryou, and the traffic congestion degree information calculated previously at the present time, caused when exceeding a certain threshold the time function abundance of a section located downstream of the adjacent sections, the course of the time The time function transmission amount transmitted to the adjacent section along with the time function transmission amount from the section located on the downstream side in the direction opposite to the traveling direction of the vehicle, the predetermined function after a predetermined time has elapsed. A traffic congestion prediction information generating method including the step of generating traffic congestion degree information indicating the degree of congestion for each section as traffic congestion prediction information. With this configuration, it is possible to reduce the demand for input data, reduce the amount of calculation for prediction, and generate more reliable traffic jam prediction information for route search.

また、本発明によれば、車両が走行する走行路の所定の区間ごとの混雑程度を示す交通混雑程度情報を生成する渋滞予測情報生成装置と、前記渋滞予測情報生成装置によって生成された前記交通混雑程度情報に基づいて、車両の目的地までの経路探索を行う経路探索装置とを備える経路探索システムであって、前記渋滞予測情報生成装置が、前記所定の区間ごとの前記走行路及び前記走行路周辺の交通量の情報及び前記所定の区間ごとの前記走行路に関する情報を含む地図情報を受信する受信手段と、受信された前記走行路及び前記走行路周辺の交通量の情報及び前記地図情報に基づいて、前記所定の区間ごとに車両のその時点の存在量を取得し、取得された前記存在量を含む時間関数による時間関数存在量を前記所定の区間ごとに算出し、算出された前記時間関数存在量を含む時間関数による時間の経過に伴って隣接する区間へ伝達する時間関数伝達量を前記所定の区間ごとに算出し、算出された前記時間関数存在量及び前記時間関数伝達量と、現時点における以前に算出された前記交通混雑程度情報と、隣接する区間のうち下流側に位置する区間での前記時間関数存在量がある閾値を超えた場合に生じる、前記時間の経過に伴って隣接する区間へ伝達する前記時間関数伝達量の前記車両の走行方向と逆方向の前記下流側に位置する区間からの時間関数伝達量、及び前記走行路に設置された信号機の信号サイクルの赤信号の長さに応じて生じる時間関数切替量の２つのうち少なくとも一方とに基づいて、所定の時間経過後の前記所定の区間ごとの混雑程度を示す前記交通混雑程度情報を渋滞予測情報として生成する生成手段とを備える経路探索システムが提供される。この構成により、入力データに対する要求を抑え、また予測をする際の計算量を抑え、経路探索のための信頼度のより高い渋滞予測の情報を生成することができる。 In addition, according to the present invention, the traffic jam prediction information generating device that generates traffic congestion degree information indicating the degree of congestion for each predetermined section of the travel route on which the vehicle travels, and the traffic generated by the traffic jam prediction information generating device. A route search system including a route search device that performs a route search to a vehicle destination based on congestion degree information, wherein the traffic jam prediction information generation device includes the travel route and the travel for each of the predetermined sections. Receiving means for receiving map information including information on the traffic volume around the road and information on the travel path for each predetermined section, the received information on the travel path and the traffic volume around the travel path, and the map information Based on the above, obtain the abundance at that time of the vehicle for each of the predetermined section, calculate a time function abundance by a time function including the acquired abundance for each of the predetermined sections, A time function transmission amount to be transmitted to an adjacent section with the passage of time by the time function including the time function existence amount that has been issued is calculated for each predetermined section, and the calculated time function existence amount and the time The amount of function transmission, the traffic congestion degree information calculated previously at the present time, and the time function existing in the section located downstream of the adjacent sections, when the amount of time function exceeds a certain threshold , The amount of time function transmitted from the section located on the downstream side in the direction opposite to the traveling direction of the vehicle, and the signal of the traffic light installed on the traveling path based on at least one of the two time function switching amount generated depending on the length of the cycle of red light, the traffic congestion degree indicating the degree congestion for each of the predetermined section after a predetermined time has elapsed Route search system comprising generation means for generating a broadcast as traffic jam prediction information is provided. With this configuration, it is possible to reduce the demand for input data, reduce the amount of calculation for prediction, and generate more reliable traffic jam prediction information for route search.

また、本発明によれば、車両が走行する走行路の所定の区間ごとの混雑程度を示す交通混雑程度情報を生成する渋滞予測情報生成装置と、前記渋滞予測情報生成装置によって生成された前記交通混雑程度情報に基づいて、車両の目的地までの経路探索を行う経路探索装置とを備える経路探索システムであって、前記渋滞予測情報生成装置が、前記所定の区間ごとの前記走行路及び前記走行路周辺の交通量の情報及び前記所定の区間ごとの前記走行路に関する情報を含む地図情報を受信する受信手段と、受信された前記走行路及び前記走行路周辺の交通量の情報及び前記地図情報に基づいて、前記所定の区間ごとに車両のその時点の存在量を取得し、取得された前記存在量を含む時間関数による時間関数存在量を前記所定の区間ごとに算出し、算出された前記時間関数存在量を含む時間関数による時間の経過に伴って隣接する区間へ伝達する時間関数伝達量を前記所定の区間ごとに算出し、算出された前記時間関数存在量及び前記時間関数伝達量と、現時点における以前に算出された前記交通混雑程度情報と、隣接する区間のうち下流側に位置する区間での前記時間関数存在量がある閾値を超えた場合に生じる、前記時間の経過に伴って隣接する区間へ伝達する前記時間関数伝達量の前記車両の走行方向と逆方向の前記下流側に位置する区間からの時間関数伝達量とに基づいて、所定の時間経過後の前記所定の区間ごとの混雑程度を示す前記交通混雑程度情報を渋滞予測情報として生成する生成手段とを備える経路探索システムが提供される。この構成により、入力データに対する要求を抑え、また予測をする際の計算量を抑え、経路探索のための信頼度のより高い渋滞予測の情報を生成することができる。 In addition, according to the present invention, the traffic jam prediction information generating device that generates traffic congestion degree information indicating the degree of congestion for each predetermined section of the travel route on which the vehicle travels, and the traffic generated by the traffic jam prediction information generating device. A route search system including a route search device that performs a route search to a vehicle destination based on congestion degree information, wherein the traffic jam prediction information generation device includes the travel route and the travel for each of the predetermined sections. Receiving means for receiving map information including information on the traffic volume around the road and information on the travel path for each predetermined section, the received information on the travel path and the traffic volume around the travel path, and the map information Based on the above, obtain the abundance at that time of the vehicle for each of the predetermined section, calculate a time function abundance by a time function including the acquired abundance for each of the predetermined sections, A time function transmission amount to be transmitted to an adjacent section with the passage of time by the time function including the time function existence amount that has been issued is calculated for each predetermined section, and the calculated time function existence amount and the time The amount of function transmission, the traffic congestion degree information calculated previously at the present time, and the time function existing in the section located downstream of the adjacent sections, when the amount of time function exceeds a certain threshold , Based on a time function transmission amount from a section located on the downstream side in a direction opposite to the traveling direction of the vehicle, the time function transmission amount transmitted to an adjacent section as time passes, after the predetermined time has elapsed A route search system is provided that includes generation means for generating the traffic congestion degree information indicating the degree of congestion for each predetermined section as traffic jam prediction information. With this configuration, it is possible to reduce the demand for input data, reduce the amount of calculation for prediction, and generate more reliable traffic jam prediction information for route search.

本発明の渋滞予測情報生成装置、渋滞予測情報生成方法、及び経路探索システムは、上記構成を有し、入力データに対する要求を抑え、また予測をする際の計算量を抑え、経路探索のための信頼度のより高い渋滞予測の情報を生成することができる。 The traffic jam prediction information generation apparatus, traffic jam prediction information generation method, and route search system of the present invention have the above-described configuration, suppress requests for input data, reduce the amount of calculation for prediction, and It is possible to generate congestion prediction information with higher reliability.

以下、本発明の実施の形態について、図１から図１０を用いて説明する。図１は本発明の実施の形態に係る経路探索システムの構成の一例を示す構成図である。図２は本発明の実施の形態に係る渋滞予測情報生成装置の構成の一例を示す構成図である。図３は本発明の実施の形態に係る渋滞予測情報生成装置におけるフェロモンを数値化した一例について説明するための図である。図４Ａは本発明の実施の形態に係る渋滞予測情報生成装置における発生フェロモンを数値化した一例を示す図である。図４Ｂは本発明の実施の形態に係る渋滞予測情報生成装置における図４Ａの発生フェロモンが時間ｔ１を経過したときの発生フェロモンを示す図である。図４Ｃは本発明の実施の形態に係る渋滞予測情報生成装置における図４Ｂの発生フェロモンが時間ｔ２を経過したときの発生フェロモンを示す図である。 Hereinafter, embodiments of the present invention will be described with reference to FIGS. FIG. 1 is a block diagram showing an example of the configuration of a route search system according to an embodiment of the present invention. FIG. 2 is a block diagram showing an example of the configuration of the traffic jam prediction information generating apparatus according to the embodiment of the present invention. FIG. 3 is a diagram for explaining an example in which pheromones are digitized in the traffic jam prediction information generating apparatus according to the embodiment of the present invention. FIG. 4A is a diagram showing an example in which generated pheromones are digitized in the traffic jam prediction information generating apparatus according to the embodiment of the present invention. FIG. 4B is a diagram showing a generated pheromone when the generated pheromone in FIG. 4A has passed the time t1 in the traffic jam prediction information generating apparatus according to the embodiment of the present invention. FIG. 4C is a diagram illustrating the generated pheromone when the generated pheromone in FIG. 4B has passed the time t2 in the traffic jam prediction information generation device according to the embodiment of the present invention.

図５Ａは本発明の実施の形態に係る渋滞予測情報生成装置における発生フェロモンを数値化した一例を示す図である。図５Ｂは本発明の実施の形態に係る渋滞予測情報生成装置における図５Ａの発生フェロモンが時間ｔ３経過したときに伝達される様子を示す図である。図５Ｃは本発明の実施の形態に係る渋滞予測情報生成装置における図５Ｂの発生フェロモンが時間ｔ４経過したときに伝達される様子を示す図である。図６は本発明の実施の形態に係る渋滞予測情報生成装置におけるショックウェーブフェロモンについて説明するための図である。図７は本発明の実施の形態に係る渋滞予測情報生成装置における信号フェロモンについて説明するための図である。図８は本発明の実施の形態に係る渋滞予測情報生成装置における信号フェロモンを算出する際の信号サイクルのオフセット、信号間隔、赤信号の期間を示す図である。図９は本発明の実施の形態に係る渋滞予測情報生成装置の動作フローについて説明するためのフローチャートである。 FIG. 5A is a diagram showing an example in which generated pheromones are digitized in the traffic jam prediction information generating apparatus according to the embodiment of the present invention. FIG. 5B is a diagram illustrating a state in which the generated pheromone of FIG. 5A is transmitted when time t3 has elapsed in the traffic jam prediction information generation device according to the exemplary embodiment of the present invention. FIG. 5C is a diagram illustrating a state in which the generated pheromone of FIG. 5B is transmitted when time t4 has elapsed in the traffic jam prediction information generation device according to the exemplary embodiment of the present invention. FIG. 6 is a diagram for explaining a shock wave pheromone in the traffic jam prediction information generating apparatus according to the embodiment of the present invention. FIG. 7 is a diagram for explaining a signal pheromone in the traffic jam prediction information generating apparatus according to the embodiment of the present invention. FIG. 8 is a diagram showing signal cycle offsets, signal intervals, and red signal periods when calculating a signal pheromone in the traffic jam prediction information generating apparatus according to the embodiment of the present invention. FIG. 9 is a flowchart for explaining an operation flow of the traffic jam prediction information generating apparatus according to the embodiment of the present invention.

まず、本発明の実施の形態に係る経路探索システムの構成の一例について図１を用いて説明する。図１に示すように、経路探索システム１００は、車両が走行する道路（走行路）において所定の道路区間ごとの交通量を後述する渋滞予測情報生成装置１０３に送信する情報提供装置１０１、車両が走行する道路を含む地図の情報を後述する渋滞予測情報生成装置１０３に送信する地図情報提供装置１０２、情報提供装置１０１から受信した所定の道路区間ごとの交通量の情報と、地図情報提供装置１０２から受信した所定の道路区間を含む地図情報とに基づいて、車両が走行する所定の道路区間ごとの混雑程度（所定の道路区間ごとにおける車両の密度など）を示す交通混雑程度情報（交通フェロモン量）を生成する渋滞予測情報生成装置１０３、渋滞予測情報生成装置１０３によって生成された交通混雑程度情報に基づいて、車両の最適な経路探索を行う経路探索装置１０４から構成されている。ここで、車両とは、例えば自動四輪車、自動二輪車などを言う。 First, an example of the configuration of the route search system according to the embodiment of the present invention will be described with reference to FIG. As shown in FIG. 1, the route search system 100 includes an information providing apparatus 101 that transmits a traffic volume for each predetermined road section to a traffic jam prediction information generating apparatus 103 described later on a road (traveling road) on which the vehicle travels. Map information providing apparatus 102 that transmits information on a map including a road to travel to a traffic jam prediction information generating apparatus 103 to be described later, information on traffic volume for each predetermined road section received from the information providing apparatus 101, and map information providing apparatus 102 Traffic congestion degree information (traffic pheromone amount) indicating the degree of congestion for each predetermined road section on which the vehicle travels (such as the density of the vehicle for each predetermined road section) based on the map information including the predetermined road section received from Based on the traffic congestion degree information generated by the traffic jam prediction information generating device 103 and the traffic jam prediction information generating device 103. And a route search device 104 that performs a route search. Here, the vehicle refers to, for example, an automobile and a motorcycle.

所定の道路区間とは、例えば交差点間の道路区間（リンクとも言う）を言う。ただし、所定の道路区間はリンクに限られるものではなく、リンクをさらに細かく分けた区間（セル）であっても、複数のリンクを含む区間（エリア）であってもよい。以下では所定の道路区間をリンクとして説明する。また、情報提供装置１０１として、例えばＶＩＣＳ（Vehicle Information and Communication System）やプローブカーシステムなど考えられる。プローブカーシステムとは、走行している車をセンサーとして活用し、道路上での様々な情報（データ）を収集し、収集されたデータに所定の処理をするシステムを言う。また、情報提供装置１０１は、リンクごとに構成されたリンクの統計情報（当該リンクにおける過去の交通状況などの情報）を渋滞予測情報生成装置１０３に送信するようにしてもよい。 The predetermined road section is, for example, a road section (also called a link) between intersections. However, the predetermined road section is not limited to the link, and may be a section (cell) obtained by further dividing the link or a section (area) including a plurality of links. Hereinafter, a predetermined road section will be described as a link. Further, as the information providing apparatus 101, for example, a VICS (Vehicle Information and Communication System) or a probe car system can be considered. The probe car system refers to a system that uses a traveling car as a sensor, collects various information (data) on the road, and performs predetermined processing on the collected data. Further, the information providing apparatus 101 may transmit link statistical information (information such as past traffic conditions in the link) configured for each link to the traffic jam prediction information generating apparatus 103.

地図情報提供装置１０２は、車両が走行する道路を含む地図の情報を渋滞予測情報生成装置１０３に送信するが、特に道路のリンクの構造に関する情報を送信する。この他に、地図情報提供装置１０２は、道路のリンクの特性、周辺施設の位置や属性、信号機の位置などの情報を渋滞予測情報生成装置１０３に送信するようにしてもよい。経路探索装置１０４は、例えば各車両に搭載されたカーナビゲーションシステムを言う。なお、経路探索装置１０４を各車両に搭載されたカーナビゲーションシステムとするのではなく、探索された経路情報を各車両に提供するサーバとして機能させるようにしてもよい。また、経路探索装置１０４は、カーナビゲーションシステムに限られるものでなく、携帯電話端末やＰＤＡ端末などの携帯情報端末であっても実施可能である。この場合、携帯電話端末やＰＤＡ端末などの携帯情報端末は、カーナビゲーションシステムが有する機能を備えている。 The map information providing apparatus 102 transmits information on a map including a road on which the vehicle is traveling to the traffic jam prediction information generating apparatus 103, and particularly transmits information on the structure of the road link. In addition, the map information providing apparatus 102 may transmit information such as road link characteristics, positions and attributes of surrounding facilities, and traffic signal positions to the traffic jam prediction information generating apparatus 103. The route search device 104 refers to a car navigation system mounted on each vehicle, for example. In addition, the route search device 104 may not be a car navigation system mounted on each vehicle, but may function as a server that provides the searched route information to each vehicle. The route search apparatus 104 is not limited to a car navigation system, and can be implemented even with a portable information terminal such as a mobile phone terminal or a PDA terminal. In this case, a portable information terminal such as a mobile phone terminal or a PDA terminal has a function that the car navigation system has.

次に、本発明の実施の形態に係る渋滞予測情報生成装置の構成の一例について図２を用いて説明する。図２に示すように、渋滞予測情報生成装置１０３は、受信部２００、生成部２０１、送信部２０２、記憶部２０３から構成されている。受信部２００は、例えば情報提供装置１０１から所定の道路区間（リンク）ごとの交通量の情報（リアルタイムの交通量の情報）や、地図情報提供装置１０２から所定の道路区間（リンク）を含む地図情報（リンクの構造を示す情報を含む）を受信するものである。 Next, an example of the configuration of the traffic jam prediction information generation apparatus according to the embodiment of the present invention will be described with reference to FIG. As illustrated in FIG. 2, the traffic jam prediction information generation device 103 includes a reception unit 200, a generation unit 201, a transmission unit 202, and a storage unit 203. The receiving unit 200 includes, for example, traffic information (information on real-time traffic volume) for each predetermined road section (link) from the information providing apparatus 101 and a map including a predetermined road section (link) from the map information providing apparatus 102. Information (including information indicating the structure of the link) is received.

生成部２０１は、受信された交通量の情報及び地図情報に基づいて、リンクごとに車両の存在量を取得し、取得された存在量を含む時間関数による時間関数存在量（発生フェロモンとも言う）をリンクごとに算出し、算出された時間関数存在量を含む時間関数による時間の経過に伴って隣接するリンクへ伝達する時間関数伝達量（伝達フェロモンとも言う）をリンクごとに算出し、算出された時間関数存在量及び時間関数伝達量と、現時点における以前に算出された交通混雑程度情報（残存フェロモンとも言う）と、所定の条件を満たした場合に生じる、時間の経過に伴って隣接するリンクへ伝達する方向と逆方向の時間関数伝達量、及び道路に設置された信号の切り替わりサイクルによって生じる時間関数切替量のうち少なくとも一方とに基づいて、所定の時間経過後の交通混雑程度情報（交通フェロモンともいう）を生成するものである。送信部２０２は、生成された交通混雑程度情報を経路探索装置１０４に送信するものである。 The generation unit 201 acquires the vehicle abundance for each link based on the received traffic information and map information, and a time function abundance (also referred to as a generated pheromone) by a time function including the obtained abundance. Is calculated for each link, and the time function transfer amount (also referred to as transfer pheromone) to be transmitted to the adjacent link with the passage of time by the time function including the calculated time function abundance is calculated for each link. Time function existence amount and time function transmission amount, traffic congestion degree information calculated at the present time (also called residual pheromone), and adjacent links over time that occur when predetermined conditions are met At least one of the amount of time function transmission in the direction opposite to the direction of transmission to the time and the amount of time function switching caused by the switching cycle of signals installed on the road Zui it, and generates a traffic congestion degree information after lapse of a predetermined time period (also referred to as traffic pheromones). The transmission unit 202 transmits the generated traffic congestion degree information to the route search device 104.

時間関数とは、例えば時間ｔの関数である。ここで、時間関数はあらかじめ設定されたものであっても動的に設定されるものであっても構わない。また、フェロモンは、例えば「車両が５台存在する場合に１０である」という基準によって決められる。フェロモンの範囲は０から１０であっても０から１００であっても構わず、これらに限られるものではない。数値化されたフェロモンの一例について図３を用いて説明する。図３に示すように、地図は第１のエリア３００及び第２のエリア３０１に分けられており、地図上には道路３０２が示されている。以下、これらの分けられた各エリアを情報伝達フィールドとも言う。地図は情報伝達フィールド３００及び３０１のみに分けられるものではなく、複数の情報伝達フィールドに分けられるとしても構わない。また、情報伝達フィールドがさらに交差点間の道路ごとに分けられた領域を情報伝達プログラムフィールド（上述したリンクに相当）とも言う。図３では、情報伝達プログラムフィールド（リンク）３０２に相当する場所において発生する発生フェロモンが１０とされた場合を示している。 The time function is a function of time t, for example. Here, the time function may be set in advance or dynamically. Further, the pheromone is determined based on, for example, a criterion of “10 when there are five vehicles”. The range of the pheromone may be 0 to 10 or 0 to 100, and is not limited thereto. An example of a digitized pheromone will be described with reference to FIG. As shown in FIG. 3, the map is divided into a first area 300 and a second area 301, and a road 302 is shown on the map. Hereinafter, these divided areas are also referred to as information transmission fields. The map is not divided into only the information transmission fields 300 and 301, but may be divided into a plurality of information transmission fields. An area where the information transmission field is further divided for each road between intersections is also referred to as an information transmission program field (corresponding to the link described above). FIG. 3 shows a case where the number of generated pheromones generated at a place corresponding to the information transmission program field (link) 302 is 10.

発生フェロモンは、時間の経過に伴って減衰していくものであり、例えば１秒間に１割減衰していくものである。減衰させる時間は１秒間隔に限られず、何秒かの間隔で減衰させても構わない。また、減衰させる割合も１割に限られず何割であっても構わないが１割から２割の割合で減衰させることが好ましい。ここで、時間の経過に伴って減衰していく発生フェロモンについて図４Ａ〜４Ｃを用いて説明する。図４Ａ〜４Ｃに示すように、図４Ａに示すリンク３０２の発生フェロモンは、ｔ１時間経過すると図４Ｂのようになり、さらにｔ２時間経過すると図４Ｃのようになる。このように、時間の経過に伴って発生フェロモンは減衰していく。なお、所定の時間が経過した場合には、発生フェロモンを無効とするように設定されていてもよい。 The generated pheromone is attenuated with the passage of time, for example, is attenuated by 10% per second. The time for attenuation is not limited to 1 second intervals, and may be attenuated at intervals of several seconds. Further, the rate of attenuation is not limited to 10% and may be any number, but it is preferable to attenuate at a rate of 10% to 20%. Here, the generated pheromone that decays with time will be described with reference to FIGS. As shown in FIGS. 4A to 4C, the generated pheromone of the link 302 shown in FIG. 4A becomes as shown in FIG. 4B when t1 time elapses, and becomes as shown in FIG. 4C when t2 time elapses. In this way, the generated pheromone decays with time. In addition, when predetermined time passes, you may set so that the generation | occurrence | production pheromone may be invalidated.

また、発生フェロモンは、時間の経過に伴って隣接するリンク（情報伝達プログラムフィールド）などに伝達されていくものであり、例えば１秒間に１割伝達されていくものである。伝達させる時間は１秒間隔に限られず、何秒かの間隔で伝達させてもよい。また、伝達させる割合も１割に限られず何割であってもよい。 Further, the generated pheromone is transmitted to an adjacent link (information transmission program field) or the like with the passage of time, for example, 10% is transmitted per second. The transmission time is not limited to the one second interval, and may be transmitted at intervals of several seconds. Further, the rate of transmission is not limited to 10% and may be any number.

また、隣接するリンク（情報伝達プログラムフィールド）などにすでに存在する発生フェロモンが、所定の数値を超えている場合には伝達する割合は小さく、所定の数値を超えていない場合には伝達する割合は大きくなる。例えば、隣接するリンク（情報伝達プログラムフィールド）の発生フェロモンが５を超えた場合には１秒間に１割伝達し、５を超えない場合であれば１秒間に２割伝達するようにしてもよい。ここで、時間の経過に伴って伝達していく発生フェロモンについて図５Ａ〜５Ｃを用いて説明する。図５Ａ〜５Ｃに示すように、図５Ａに示すリンク３０２は、ｔ３時間経過すると図５Ｂのようになり、さらにｔ４時間経過すると図５Ｃのようになる。このように、時間の経過に伴って発生フェロモンは伝達されていく。なお、所定の時間が経過した場合には、発生フェロモンを無効とするように設定されていてもよい。 In addition, if the generated pheromone already existing in the adjacent link (information transmission program field) exceeds a predetermined value, the transmission rate is small, and if it does not exceed the predetermined value, the transmission rate is growing. For example, when the number of generated pheromones of an adjacent link (information transmission program field) exceeds 5, 10% may be transmitted per second, and when it does not exceed 5, 20% may be transmitted per second. . Here, the generated pheromone that is transmitted with the passage of time will be described with reference to FIGS. As shown in FIGS. 5A to 5C, the link 302 shown in FIG. 5A becomes as shown in FIG. 5B when t3 time elapses, and becomes as shown in FIG. 5C when t4 time elapses. Thus, the generated pheromone is transmitted with the passage of time. In addition, when predetermined time passes, you may set so that the generation | occurrence | production pheromone may be invalidated.

ここで、交通フェロモン量、発生フェロモン、伝達フェロモンの算出式を式（１）から式（３）にそれぞれ示す。なお、これらの式は漸化式である。 Here, the calculation formulas of the traffic pheromone amount, the generated pheromone, and the transmission pheromone are shown in formulas (1) to (3), respectively. Note that these equations are recurrence equations.

なお、式（１）におけるＥ_pはリンクｐにおける蒸発率（０＜Ｅ_p＜１）を示し、α_pはリンクｐにおける発生率（０＜α_p＜１）を示している。ｒ（ｔ、ｐ）は時刻ｔにおけるリンクｐで発生する発生フェロモンの強さを示している。ｑ（ｔ、ｐ）は時刻ｔにおけるリンクｐでの伝達フェロモンの強さを示している。また、式（２）において、Ｃ(ｔ、ｐ)は時刻ｔにおけるリンクｐでの車両通過台数を示しており、ｖ（ｔ、ｐ）は時刻ｔにおけるリンクｐでの車両の速度を示している。また、式（３）において、Ｆ_p'はリンクｐ´における拡散率（０＜Ｆ_p'＜１）を示し、Ｎ_upstream（ｐ）はリンクｐの上流に位置するリンクの数を示しており、例えばあるリンクｐに発生フェロモンが拡散されてくる拡散元（上流とも言う）のリンクの数である。 In Equation (1), E _p represents the evaporation rate at link p (0 <E _p <1), and α _p represents the occurrence rate at link p (0 <α _p <1). r (t, p) indicates the strength of the generated pheromone generated at the link p at time t. q (t, p) indicates the strength of the transmission pheromone on the link p at time t. In the equation (2), C (t, p) indicates the number of vehicles passing through the link p at time t, and v (t, p) indicates the speed of the vehicle at the link p at time t. Yes. In Expression (3), F _{p ′} represents the spreading factor (0 <F _{p ′} <1) in the link p ′, and N _upstream (p) represents the number of links located upstream of the link p. For example, it is the number of diffusion source links (also referred to as upstream) in which the generated pheromone is diffused to a certain link p.

なお、蒸発率、発生率、拡散率、後述するショックウェーブフェロモンにおける閾値などのパラメータの調整には、ＧＡ（Genetic Algorithm：遺伝的アルゴリズム）などによる統計データによる最適化、学習の静的調整、ＧＡなどによる統計データによる最適化、学習の繰り返し実行や強化学習による逐次的パラメータの調整の動的調整がある。なお、強化学習に関しては公知の一般的な技術であるため説明は省略する。また、蒸発率、発生率、拡散率などのパラメータを観測可能な特徴と対応付けて算出してもよい。蒸発率Ｅは、例えばあるリンクにおける密度（車両の密度など）に所定の係数を掛けたもの（時間的変動のあるもの）、拡散率Ｆは、例えばあるリンクの下流のリンクにおける車両の平均速度を車両の速度の最大値で割ったもの（時間的変動のあるもの）、あるリンクの下流のリンク数をリンク数の最大値で割ったもの（リンクに固定で時間的変動のないもの）などとしてもよい。なお、時間的変動のある他の要素として、例えば平均旅行（所要）時間、通過台数、停止線待ちの平均車両台数などの情報がある。また、時間的変動のない他の要素として、例えばリンク長、車線数、車幅、バス路線であるか否かなどの情報がある。 For adjustment of parameters such as evaporation rate, incidence rate, diffusivity, threshold value in shock wave pheromone described later, optimization by statistical data such as GA (Genetic Algorithm), static adjustment of learning, GA, etc. There are optimization by statistical data, dynamic adjustment of sequential parameter adjustment by repeated execution of learning and reinforcement learning. Since reinforcement learning is a known general technique, description thereof is omitted. Further, parameters such as evaporation rate, occurrence rate, and diffusion rate may be calculated in association with observable features. The evaporation rate E is, for example, a density (a vehicle density or the like) at a certain link multiplied by a predetermined coefficient (with time fluctuation), and the spreading factor F is, for example, an average vehicle speed at a link downstream of a certain link. Divided by the maximum value of the vehicle speed (with time fluctuation), the number of links downstream of a link divided by the maximum number of links (fixed to the link and without time fluctuation), etc. It is good. In addition, as other elements having temporal variation, there are information such as average travel (required) time, number of passing vehicles, average number of vehicles waiting for a stop line, and the like. Further, as other elements having no temporal variation, for example, there is information such as link length, number of lanes, vehicle width, and whether or not a bus route.

さらに、本発明の実施の形態では、後述するショックウェーブフェロモン及び信号フェロモンをも考慮した交通フェロモンを算出することを考える。まず、ショックウェーブフェロモンについて図６を用いて説明する。図６に示すように、リンク１の下流のリンクに相当する下流リンク（図６ではリンク５〜７）で発生する、例えば発生フェロモンの量が所定の閾値ｃを超えた場合に、上流に位置するリンク１に逆流してくるフェロモンをショックウェーブフェロモンと言う。ショックウェーブフェロモンは、下記の式（４）によって算出される。Ｎ_downstream（ｐ）はリンクｐの下流に位置するリンクの数を示しており、図６ではリンク５〜７の３となる。この式（４）から、あるリンクｐ´における発生フェロモンが所定の閾値ｃを超えない場合には０となり、ショックウェーブフェロモンは発生しない。 Furthermore, in the embodiment of the present invention, it is considered to calculate a traffic pheromone that also considers a shock wave pheromone and a signal pheromone described later. First, the shock wave pheromone will be described with reference to FIG. As shown in FIG. 6, when the amount of generated pheromone exceeds a predetermined threshold value c, for example, generated in a downstream link (links 5 to 7 in FIG. 6) corresponding to the downstream link of link 1, it is located upstream. The pheromone that flows back to link 1 is called the shock wave pheromone. The shock wave pheromone is calculated by the following equation (4). N _downstream (p) indicates the number of links located downstream of the link p. In FIG. From this equation (4), when the generated pheromone in a certain link p ′ does not exceed the predetermined threshold c, it becomes 0, and no shock wave pheromone is generated.

一方、信号フェロモンについて図７、８を用いて説明する。信号フェロモンとは、図７に示すような信号機の信号サイクルの赤信号の長さに比例して発生されるフェロモンを言う。なお、信号フェロモンは発散、拡散はしない。信号フェロモンは下記の式（５）によって算出され、この式（５）は時刻ｔのリンクｐにおける赤信号時間を示している。なお、Ｌ_redは実際の信号サイクルから求めたり、図８に示すように信号サイクルのオフセット、信号間隔、赤信号の期間をパラメータ化してＧＡなどにより推定してもよい。 On the other hand, the signal pheromone will be described with reference to FIGS. The signal pheromone is a pheromone generated in proportion to the length of the red signal in the signal cycle of the traffic light as shown in FIG. The signal pheromone does not diverge or spread. The signal pheromone is calculated by the following equation (5), and this equation (5) indicates the red signal time in the link p at time t. Note that L _red may be obtained from an actual signal cycle, or may be estimated by GA by parameterizing the signal cycle offset, signal interval, and red signal period as shown in FIG.

上述したショックウェーブフェロモン及び信号フェロモンを考慮した場合には、下記の式（６）をも加味して交通フェロモンを算出する。なお、βは所定の係数である。このように、複数のフェロモンを用いることにより、より精度の高い渋滞予測をすることができる。 When the above-described shock wave pheromone and signal pheromone are considered, the traffic pheromone is calculated in consideration of the following equation (6). Note that β is a predetermined coefficient. As described above, by using a plurality of pheromones, it is possible to predict traffic congestion with higher accuracy.

ここで、上述した渋滞予測情報生成装置１０３によって生成されたリンクごとの交通フェロモンである予測結果に基づいて経路探索をする経路探索装置１０４の経路探索への応用について説明する。１つに、経路探索装置１０４は、予測結果を利用した経路探索にダイクストラ法の改良手法（http://mmi.tudelft.nl/pub/Patrick/Ehlert.P.A.M-TRAIL Workshop2001.pdfを参照）などを適用して動的な環境での経路探索を行う。 Here, an application of the route search device 104 that performs a route search based on a prediction result that is a traffic pheromone for each link generated by the traffic jam prediction information generation device 103 described above to a route search will be described. For example, the route search apparatus 104 improves the Dijkstra method for route search using the prediction result (see http://mmi.tudelft.nl/pub/Patrick/Ehlert.PAM-TRAIL Workshop2001.pdf), etc. To search for routes in a dynamic environment.

次に、本発明の実施の形態に係る渋滞予測情報生成装置の動作フローについて図９を用いて説明する。図９に示すように、渋滞予測情報生成装置１０３は、情報提供装置１０１からリンクごとの交通量の情報（リアルタイムの交通量の情報）や、地図情報提供装置１０２からリンクを含む地図情報（リンクの構造を示す情報を含む）を受信する（ステップＳ９０１）。渋滞予測情報生成装置１０３は、受信された交通量の情報及び地図情報に基づいて、リンクごとに車両の存在量を取得し、取得された存在量を含む時間関数による時間関数存在量（発生フェロモン）をリンクごとに算出し、算出された時間関数存在量を含む時間関数による時間の経過に伴って隣接するリンクへ伝達する時間関数伝達量（伝達フェロモン）をリンクごとに算出し、算出された時間関数存在量及び時間関数伝達量と、現時点における以前に算出された交通混雑程度情報（残存フェロモン）と、所定の条件を満たした場合に生じる、時間の経過に伴って隣接するリンクへ伝達する方向（車両の走行方向）と逆方向の時間関数伝達量（ショックウェーブフェロモン）、及び道路に設置された信号の切り替わりサイクルによって生じる時間関数切替量（信号フェロモン）の２つのうち少なくとも一方とに基づいて、所定の時間経過後の交通混雑程度情報（交通フェロモン）を生成する（ステップＳ９０２）。渋滞予測情報生成装置１０３は、生成された交通混雑程度情報を経路探索装置１０４に送信する（ステップＳ９０３）。なお、信号フェロモンを除いたケースの渋滞予測情報生成装置の動作フローについても同様に考えられる。 Next, the operation flow of the traffic jam prediction information generation apparatus according to the embodiment of the present invention will be described with reference to FIG. As shown in FIG. 9, the traffic jam prediction information generating device 103 receives traffic information (link information in real time) from the information providing device 101 and map information (link) including a link from the map information providing device 102. Is received) (step S901). The traffic jam prediction information generation device 103 acquires the vehicle abundance for each link based on the received traffic information and map information, and a time function abundance (occurrence pheromone based on the time function including the obtained abundance). ) Is calculated for each link, and the time function transfer amount (transfer pheromone) transmitted to the adjacent link as time passes by the time function including the calculated time function abundance is calculated for each link. Time function existence amount and time function transmission amount, traffic congestion degree information calculated at the present time (remaining pheromone), and transmission to adjacent links as time elapses when a predetermined condition is met It is caused by the time function transmission amount (shock wave pheromone) in the opposite direction to the direction (vehicle traveling direction) and the switching cycle of signals installed on the road During the function switching amount based on at least one of the two (signal pheromones), and generates a traffic congestion degree information after lapse of a predetermined time (traffic pheromone) (step S902). The traffic jam prediction information generation device 103 transmits the generated traffic congestion degree information to the route search device 104 (step S903). The operation flow of the traffic jam prediction information generation apparatus in the case excluding the signal pheromone can be considered similarly.

また、上述した態様以外にさらにより高い精度を目指すため、以下に示すフェロモンを利用する。具体的には、渋滞を緩和させる要因のフェロモン（Attractive フェロモン）と、渋滞を大きくする要因のフェロモン（Repulsive フェロモン）とを混ぜて合わせて利用する。Attractive フェロモンの例としては車間距離フェロモンであり、Repulsive フェロモンの例としては上述した交通フェロモン、ブレーキフェロモンである。 Moreover, in order to aim at a still higher precision besides the aspect mentioned above, the following pheromones are used. Specifically, a pheromone that reduces traffic congestion (Attractive pheromone) and a pheromone that increases traffic congestion (Repulsive pheromone) are used in combination. Examples of attractive pheromones are inter-vehicle distance pheromones, and examples of repulsive pheromones include the traffic pheromones and brake pheromones described above.

まず、車間距離フェロモンについて説明する。この車間距離フェロモンは車間距離が所定の閾値を超えた場合に発生するものである。車間距離フェロモンの算出式を下記の式（７）に示す。これは、時刻ｔ、位置（リンク）ｐにおける車間距離フェロモンである。ここで、ｄｉｓ_i（ｔ、ｐ）は車両ＩＤがｉである車両における車間距離であり、Ｃ（ｔ、ｐ）は位置ｐにおける区間（ｔ−１、ｔ）内の車両ＩＤの集合である。 First, the inter-vehicle distance pheromone will be described. This inter-vehicle distance pheromone is generated when the inter-vehicle distance exceeds a predetermined threshold. The formula for calculating the inter-vehicle distance pheromone is shown in the following formula (7). This is an inter-vehicle distance pheromone at time t and position (link) p. Here, dis _i (t, p) is an inter-vehicle distance in a vehicle with the vehicle ID i, and C (t, p) is a set of vehicle IDs in the section (t−1, t) at the position p. .

一方、ブレーキフェロモンはブレーキを踏む回数に比例して一定量のフェロモンを発生するものである。ブレーキフェロモンの算出式を下記の式（８）に示す。これは時刻ｔ、位置（リンク）ｐにおけるブレーキフェロモンである。ここで、braking_i（ｔ、ｐ）は車両ＩＤがｉの車両におけるブレーキを踏んだ回数である。Ｃ（ｔ、ｐ）は位置ｐにおける区間（ｔ−１、ｔ）内の車両ＩＤの集合である。 On the other hand, the brake pheromone generates a certain amount of pheromone in proportion to the number of times the brake is depressed. The formula for calculating the brake pheromone is shown in the following formula (8). This is a brake pheromone at time t and position (link) p. Here, braking _i (t, p) is the number of times the brake is stepped on the vehicle whose vehicle ID is i. C (t, p) is a set of vehicle IDs in the section (t−1, t) at the position p.

ここでの基本のモデル式を以下の式（９）、（１０）に示す。このモデル式には、フェロモンの種類を識別する変数ｆが追加されている。これにより、蒸発率や拡散率をフェロモンの種類ごとに変化させることが可能となる。また、多種のフェロモンを合成する場合、蒸発率や拡散率をフェロモンの種類ごとに調整できるため、混合比をパラメータ化する必要はない。 The basic model formulas here are shown in the following formulas (9) and (10). A variable f for identifying the type of pheromone is added to this model formula. This makes it possible to change the evaporation rate and diffusivity for each type of pheromone. In addition, when synthesizing various pheromones, the evaporation rate and diffusivity can be adjusted for each type of pheromone, so there is no need to parameterize the mixing ratio.

これにより、より高い精度の情報を生成することができ、また、上述した式（１）や式（３）で示した蒸発率や拡散率などのパラメータについて、車両速度などの観測量と関連付けを行うことで予測精度を高めることができる。ここで、パラメータのモデルの一例を以下に示す。例えば、蒸発パラメータを走行速度に応じて決定する場合のパラメータを式（１１）に示す。これは速度が速いほど蒸発する割合が大きくなる。 As a result, information with higher accuracy can be generated, and the parameters such as the evaporation rate and the diffusion rate shown in the above formulas (1) and (3) are associated with the observation amount such as the vehicle speed. By doing so, the prediction accuracy can be increased. Here, an example of a parameter model is shown below. For example, the parameter in the case where the evaporation parameter is determined according to the traveling speed is shown in Expression (11). As the speed increases, the rate of evaporation increases.

また、例えば、拡散パラメータを走行速度に応じて決定する場合のパラメータを式（１２）に示す。 Further, for example, a parameter in the case where the diffusion parameter is determined according to the traveling speed is shown in Expression (12).

これは速度が速いほど拡散する割合が大きくなる。なお、ここでは拡散パラメータをＧ_fで示している。 This is because the rate of diffusion increases as the speed increases. Here, the diffusion parameter is indicated by G _f .

本発明に係る渋滞予測情報生成装置、渋滞予測情報生成方法、及び経路探索システムは、上記構成を有し、パラメータの動的な対応ができ、より精度の高い渋滞予測のための情報を生成することができるため、車両が目的地へ行く過程における道路の渋滞予測を行う渋滞予測情報生成装置、渋滞予測情報生成方法、及び経路探索システムなどに有用である。 The traffic jam prediction information generating apparatus, the traffic jam prediction information generating method, and the route search system according to the present invention have the above-described configuration, can dynamically handle parameters, and generate information for traffic jam prediction with higher accuracy. Therefore, the present invention is useful for a traffic jam prediction information generation device, a traffic jam prediction information generation method, a route search system, and the like that perform road traffic jam prediction in the process of a vehicle going to a destination.

本発明の実施の形態に係る経路探索システムの構成の一例を示す構成図である。It is a block diagram which shows an example of a structure of the route search system which concerns on embodiment of this invention. 本発明の実施の形態に係る渋滞予測情報生成装置の構成の一例を示す構成図である。It is a block diagram which shows an example of a structure of the traffic congestion prediction information generation apparatus which concerns on embodiment of this invention. 本発明の実施の形態に係る渋滞予測情報生成装置におけるフェロモンを数値化した一例について説明するための図である。It is a figure for demonstrating the example which digitized the pheromone in the traffic congestion prediction information generation apparatus which concerns on embodiment of this invention. 本発明の実施の形態に係る渋滞予測情報生成装置における発生フェロモンを数値化した一例を示す図である。It is a figure which shows an example which digitized the generation | occurrence | production pheromone in the traffic congestion prediction information generation apparatus which concerns on embodiment of this invention. 本発明の実施の形態に係る渋滞予測情報生成装置における図４Ａの発生フェロモンが時間ｔ１を経過したときの発生フェロモンを示す図である。It is a figure which shows the generation | occurrence | production pheromone when the generation | occurrence | production pheromone of FIG. 4A in the traffic congestion prediction information generation apparatus which concerns on embodiment of this invention passes time t1. 本発明の実施の形態に係る渋滞予測情報生成装置における図４Ｂの発生フェロモンが時間ｔ２を経過したときの発生フェロモンを示す図である。It is a figure which shows the generation | occurrence | production pheromone when the generation | occurrence | production pheromone of FIG. 4B in the traffic congestion prediction information generation apparatus which concerns on embodiment of this invention has passed time t2. 本発明の実施の形態に係る渋滞予測情報生成装置における発生フェロモンを数値化した一例を示す図である。It is a figure which shows an example which digitized the generation | occurrence | production pheromone in the traffic congestion prediction information generation apparatus which concerns on embodiment of this invention. 本発明の実施の形態に係る渋滞予測情報生成装置における図５Ａの発生フェロモンが時間ｔ３経過したときに伝達される様子を示す図である。It is a figure which shows a mode that the generation | occurrence | production pheromone of FIG. 5A is transmitted when time t3 passes in the traffic jam prediction information generation apparatus which concerns on embodiment of this invention. 本発明の実施の形態に係る渋滞予測情報生成装置における図５Ｂの発生フェロモンが時間ｔ４経過したときに伝達される様子を示す図である。It is a figure which shows a mode that the generation | occurrence | production pheromone of FIG. 5B in the congestion prediction information generation apparatus which concerns on embodiment of this invention is transmitted when time t4 passes. 本発明の実施の形態に係る渋滞予測情報生成装置におけるショックウェーブフェロモンについて説明するための図である。It is a figure for demonstrating the shock wave pheromone in the traffic congestion prediction information generation apparatus which concerns on embodiment of this invention. 本発明の実施の形態に係る渋滞予測情報生成装置における信号フェロモンについて説明するための図である。It is a figure for demonstrating the signal pheromone in the traffic congestion prediction information generation apparatus which concerns on embodiment of this invention. 本発明の実施の形態に係る渋滞予測情報生成装置における信号フェロモンを算出する際の信号サイクルのオフセット、信号間隔、赤信号の期間を示す図である。It is a figure which shows the offset of a signal cycle, the signal interval, and the period of a red signal at the time of calculating the signal pheromone in the traffic congestion prediction information generation apparatus which concerns on embodiment of this invention. 本発明の実施の形態に係る渋滞予測情報生成装置の動作フローについて説明するためのフローチャートである。It is a flowchart for demonstrating the operation | movement flow of the traffic congestion prediction information generation apparatus which concerns on embodiment of this invention.

Explanation of symbols

１、２、３、４、５、６、７リンク
１００経路探索システム
１０１情報提供装置
１０２地図情報提供装置
１０３渋滞予測情報生成装置
１０４経路探索装置
２００受信部（受信手段）
２０１生成部（生成手段）
２０２送信部
２０３記憶部
３００第１のエリア（情報伝達フィールド）
３０１第２のエリア（情報伝達フィールド）
３０２道路（情報伝達プログラムフィールド（リンク））
1, 2, 3, 4, 5, 6, 7 Link 100 Route search system 101 Information providing device 102 Map information providing device 103 Congestion prediction information generating device 104 Route searching device 200 Receiving unit (receiving means)
201 generator (generator)
202 Transmission unit 203 Storage unit 300 First area (information transmission field)
301 2nd area (information transmission field)
302 Road (Information transmission program field (link))

Claims

A traffic jam prediction information generating device that generates traffic congestion degree information indicating a degree of congestion for each predetermined section of a travel path on which a vehicle travels,
Receiving means for receiving map information including information on the travel path for each predetermined section and traffic volume around the travel path and information on the travel path for each predetermined section;
Based on the received information on the travel route, the traffic volume around the travel route, and the map information, the present amount of the vehicle for each predetermined section is obtained, and the time including the obtained existence amount A time function transmission amount that is transmitted to an adjacent interval as time elapses due to the time function including the calculated time function existence amount is calculated for each predetermined interval. Calculated for each time, the calculated time function existing amount and the time function transmission amount, the traffic congestion degree information calculated previously at the present time, and the time in the section located downstream of the adjacent sections caused when exceeding a certain threshold function abundance, or section positioned in the downstream side of the traveling direction opposite to the direction of the vehicle of the time function transmission amount to transmit to the sections adjacent with the lapse of the time Time function transfer quantity, and on the basis of at least one of the two time function switching amount generated depending on the length of the red signal of the signal cycle of the installed traffic signal to the travel path, wherein after a predetermined time has elapsed Generating means for generating the traffic congestion degree information indicating the degree of congestion for each predetermined section as traffic jam prediction information;
A traffic jam prediction information generation device provided.

A traffic jam prediction information generating device that generates traffic congestion degree information indicating a degree of congestion for each predetermined section of a travel path on which a vehicle travels,
Receiving means for receiving map information including information on the travel path for each predetermined section and traffic volume around the travel path and information on the travel path for each predetermined section;
Based on the received information on the travel route, the traffic volume around the travel route, and the map information, the present amount of the vehicle for each predetermined section is obtained, and the time including the obtained existence amount A time function transmission amount that is transmitted to an adjacent interval as time elapses due to the time function including the calculated time function existence amount is calculated for each predetermined interval. Calculated for each time, the calculated time function existing amount and the time function transmission amount, the traffic congestion degree information calculated previously at the present time, and the time in the section located downstream of the adjacent sections caused when exceeding a certain threshold function abundance, or section positioned in the downstream side of the traveling direction opposite to the direction of the vehicle of the time function transmission amount to transmit to the sections adjacent with the lapse of the time Based on the time function transfer quantity, and generating means for generating the traffic congestion degree information indicating the degree of congestion of each of the predetermined section after elapse of a predetermined time as traffic jam prediction information,
A traffic jam prediction information generation device provided.

A traffic jam prediction information generating method for generating traffic congestion degree information indicating a degree of congestion for each predetermined section of a travel route on which a vehicle travels,
Receiving map information including information on the travel route for each predetermined section and traffic volume around the travel path and information on the travel path for each predetermined section;
Based on the received information on the travel route, the traffic volume around the travel route, and the map information, the present amount of the vehicle for each predetermined section is obtained, and the time including the obtained existence amount A time function transmission amount that is transmitted to an adjacent interval as time elapses due to the time function including the calculated time function existence amount is calculated for each predetermined interval. Calculated for each time, the calculated time function existing amount and the time function transmission amount, the traffic congestion degree information calculated previously at the present time, and the time in the section located downstream of the adjacent sections caused when exceeding a certain threshold function abundance, or section positioned in the downstream side of the traveling direction opposite to the direction of the vehicle of the time function transmission amount to transmit to the sections adjacent with the lapse of the time Time function transfer quantity, and on the basis of at least one of the two time function switching amount generated depending on the length of the red signal of the signal cycle of the installed traffic signal to the travel path, wherein after a predetermined time has elapsed Generating the traffic congestion degree information indicating the degree of congestion for each predetermined section as traffic jam prediction information,
A method for generating traffic jam prediction information.

A traffic jam prediction information generating method for generating traffic congestion degree information indicating a degree of congestion for each predetermined section of a travel route on which a vehicle travels,
Receiving map information including information on the travel route for each predetermined section and traffic volume around the travel path and information on the travel path for each predetermined section;
Based on the received information on the travel route, the traffic volume around the travel route, and the map information, the present amount of the vehicle for each predetermined section is obtained, and the time including the obtained existence amount A time function transmission amount that is transmitted to an adjacent interval as time elapses due to the time function including the calculated time function existence amount is calculated for each predetermined interval. Calculated for each time, the calculated time function existing amount and the time function transmission amount, the traffic congestion degree information calculated previously at the present time, and the time in the section located downstream of the adjacent sections caused when exceeding a certain threshold function abundance, or section positioned in the downstream side of the traveling direction opposite to the direction of the vehicle of the time function transmission amount to transmit to the sections adjacent with the lapse of the time Of on the basis of the time function transfer amount, and generating a traffic jam prediction information the traffic congestion degree information indicating the degree of congestion of each of the predetermined section after elapse of a predetermined time,
A method for generating traffic jam prediction information.

A traffic jam prediction information generating device that generates traffic congestion degree information indicating the degree of congestion for each predetermined section of a travel path on which a vehicle travels;
A route search system comprising a route search device for searching for a route to a destination of a vehicle based on the traffic congestion degree information generated by the congestion prediction information generation device,
The traffic jam prediction information generating device
Receiving means for receiving map information including information on the travel path for each predetermined section and traffic volume around the travel path and information on the travel path for each predetermined section;
Based on the received information on the travel route, the traffic volume around the travel route, and the map information, the present amount of the vehicle for each predetermined section is obtained, and the time including the obtained existence amount A time function transmission amount that is transmitted to an adjacent interval as time elapses due to the time function including the calculated time function existence amount is calculated for each predetermined interval. Calculated for each time, the calculated time function existing amount and the time function transmission amount, the traffic congestion degree information calculated previously at the present time, and the time in the section located downstream of the adjacent sections caused when exceeding a certain threshold function abundance, or section positioned in the downstream side of the traveling direction opposite to the direction of the vehicle of the time function transmission amount to transmit to the sections adjacent with the lapse of the time Time function transfer quantity, and on the basis of at least one of the two time function switching amount generated depending on the length of the red signal of the signal cycle of the installed traffic signal to the travel path, wherein after a predetermined time has elapsed A route search system comprising: generation means for generating the traffic congestion degree information indicating the degree of congestion for each predetermined section as congestion prediction information.

A traffic jam prediction information generating device that generates traffic congestion degree information indicating the degree of congestion for each predetermined section of a travel path on which a vehicle travels;
A route search system comprising a route search device for searching for a route to a destination of a vehicle based on the traffic congestion degree information generated by the congestion prediction information generation device,
The traffic jam prediction information generating device
Receiving means for receiving map information including information on the travel path for each predetermined section and traffic volume around the travel path and information on the travel path for each predetermined section;
Based on the received information on the travel route, the traffic volume around the travel route, and the map information, the present amount of the vehicle for each predetermined section is obtained, and the time including the obtained existence amount A time function transmission amount that is transmitted to an adjacent interval as time elapses due to the time function including the calculated time function existence amount is calculated for each predetermined interval. Calculated for each time, the calculated time function existing amount and the time function transmission amount, the traffic congestion degree information calculated previously at the present time, and the time in the section located downstream of the adjacent sections caused when exceeding a certain threshold function abundance, or section positioned in the downstream side of the traveling direction opposite to the direction of the vehicle of the time function transmission amount to transmit to the sections adjacent with the lapse of the time The route search system comprising of, based on the time function transfer amount, and generating means for generating the traffic congestion degree information indicating the degree of congestion of each of the predetermined section after a predetermined time has elapsed as congestion prediction information.