JP4041886B2 - Method for autonomous movement control of artificial object by recognition of human intention, communication system and program thereof - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
  The present invention is particularly relevant for the externalized intent (appropriateness)YichengArtifacts (such as wheelchairs, umbrella stands, and book transporters) that are selected according to the human “gesture” including the voice are automatically displaced in position and posture to artificially move to human positions. The present invention relates to an autonomous movement control method for an artificial object by recognizing a human intention, and a communication system and program for the autonomous movement of the object.
[0002]
[Prior art]
Conventionally, control of an artificial object, for example, a wired remote controller or a wireless remote controller using weak radio waves or infrared rays has been widely used. In addition to such passive control, an example in which an artificial object moves autonomously (actively) by the control is also known.
[0003]
  For example, a human externalized intention at a remote locationofRecognitionIntellectCorrespondingly, there is an example in which an artificial object autonomously displaces its position / posture and moves to a position (predetermined position) near the target human being (for example, see Non-Patent Document 1).
[0004]
  This kindIntentionalRecognition recognizes the necessary transmission between an active artifact and a human being as an implicit physical transmission. This implicit physical communication is related to physical and / or environmental attributes such as “body movement, interpersonal contact, interpersonal distance, surrounding language, physical characteristics, clothing, ornaments, silence”.
[0005]
An example of a two-legged robot is known as a similar proposal. This example of a two-legged robot understands human “gestures” and the environment (recognition of obstacles to travel and its detour travel) through image analysis that links multiple recognition technologies, and To improve dialogue skills. That is, the artificial object moves autonomously by understanding the correct traveling direction (see, for example, Non-Patent Document 2).
[0006]
[Non-Patent Document 1]
Kazunori Terada and Toyoaki Nishida, "Harmonic Artifacts (HARMONICAR)" National Conference of the Japanese Society for Artificial Intelligence (16th) Proceedings (CD-ROM), 1C4-04 (2002)
[Non-Patent Document 2]
Issued by Nikkei BP, "Nikkei Electronics Magazine" 2003-1-6 "New ASIMO that understands human behavior and leads to the Internet" p24-p25
[0007]
[Problems to be solved by the invention]
Such a conventional example has the following disadvantages (1) and (2).
(1) In the example of the former non-patent document 1, in order to correspond one artificial object to an externalized intention of one human, the vicinity position of the target human being that the artificial object autonomously aims at Like to move on. Therefore, it is not possible to automatically move from a plurality of types of artifacts to a target position near a human being by an appropriate selection by inference and search based on the human's external intention. In other words, the convenience as an easy-to-use system for adapting various artifacts to various people could not be obtained.
(2) In the example of the latter non-patent document 2, the robot of the artificial object automatically matches the target position (for example, for example, so as to correspond to one person's externalized intention). (Guide position for visitors). In this case as well, various robots can be adapted (selected) to various people and cannot be moved. That is, convenience as a system that is more convenient to use cannot be obtained.
[0008]
In the latter example of Non-Patent Document 2, an obstacle is recognized when traveling to a target position, and it is possible to travel around the obstacle, and the ability to interact through the exchange of information through the network. Although improvement is aimed at, the information in this case cannot be autonomously traveled to various target positions using, for example, the structure in the building as map information. That is, convenience as a more advanced system cannot be obtained.
[0009]
  In view of the above circumstances, the present invention intends that a plurality of types of artifacts are externalized by humans (here, “human gestures”).)ofRecognitionIntellectSelect one or more by inference and search based onseedArtifact) andAutonomousBy controlling the position and posture, it becomes possible to move autonomously to the position near the target human being, and as a result, it can adapt various artifacts to various people, and it is easy to use It is an object of the present invention to provide an autonomous movement control method of an artifact, a communication system thereof, and a program by inference and search with respect to an externalized intention of a human who improve convenience as a system.
[0010]
[Means for Solving the Problems]
  An autonomous movement control method for an artifact by recognizing a human intention according to the present invention that achieves the above object is as follows.Implemented in communication system with multiple types of artifacts and autonomous mobility control system, Because multiple types of artifacts move selectively and autonomously,The autonomous mobile control system performs the following steps.
(A)At least one of the multiple types of artifactsCaptured, Human at a distanceOne or both of the imaging data and the audio data are received.
(B) Using one or both of the received imaging data and audio data,
When using imaging data, by comparing the degree of similarity between the pre-stored standard pattern for image recognition including physical movements and physical features and the imaging data, or determining the degree of similarity, or
When using voice data, by comparing the voice data with a standard pattern for voice recognition with respect to the voice of the target person stored in advance, the voice data is recognized,
HumanRecognize the externalized intention ofRecognizeRu.
(C)recognitiondid, Automatically select at least one of multiple types of artifacts in response to human externalization.
(D)ChoiceShiArtifactofPosition and postureAutonomouslyDisplaceShown by recognized intentEyeTargetMoveThe
[0011]
  In the autonomous movement control method for artifacts by recognizing human intentions according to the present invention, a plurality of types of artifacts are selectively (one or plural) by inference / retrieval based on human externalized intentions.seedArtifacts), andAutonomousPerform position / posture control and autonomouslyTargetIt will be possible to move it. As a result, the convenience as an easy-to-use system for adapting various artifacts to various people is improved.
  The recognition of intention is performed by one or both of inference and search control execution and / or information search for one or both of imaging data and voice data of a target destination human being. By performing inference / retrieval based on the externalized intention, the selectivity when selecting (one or more kinds of artifacts) is improved.
[0012]
  The artifact mobile communication system of the present invention that achieves the above object is as follows.Including an autonomous movement control system and multiple types of artifacts,Multiple species selected by recognition including inference and search for human externalized intentionsofArtifacts move autonomouslyTheAt least one of the plurality of types of artifacts is a spaced positionPerson ofExternalized intentions betweenofrecognitionofImaging means for outputting imaging data forIntentionalOne or both of sound collecting means for outputting sound data for recognitionAnd control / drive means for moving to the target position indicated by the recognized intention based on the movement instruction data received from the autonomous movement control system;Withing. The autonomous mobile control system received from an artifactImaging dataas well asFrom one or both of the audio dataHuman intentionsRecognitionHand to recognizeStep and frontRecognizingIntentionIn the figureSelect the corresponding artifact from multiple typesHandStep and frontSelectionSelected artifactsofDisplace position / postureCalculate and send the movement instruction dataMeans.
[0013]
In the artifact mobile communication system according to the present invention, as in the above-described “autonomous movement control method for artifacts by recognizing human intention”, a system that is easy to use and adapts various artifacts to various people. As a result, convenience will be improved.
[0014]
  The program of the present invention that achieves the above object is as follows.A communication system with multiple types of artifacts and an autonomous movement control system.Autonomous Artifacts by Inferring and Retrieving Human Externalized Intentions for Multiple Kinds of Artifacts to Move selectively and autonomously by Recognition including Inference and Search for Human Externalized Intentions The following steps for causing the computer to execute various movement controls.
[0015]
(A)At least one of the multiple types of artifactsCaptured, Human at a distanceReceives either or both of imaging data and audio data forStep to do,
(B) Using one or both of the received imaging data and audio data,
When using the imaging data, by comparing the degree of similarity between the pre-stored standard pattern for image recognition including physical movements and physical features and the imaging data, the similarity is determined, Or
When using the voice data, by comparing the voice data with a standard pattern for voice recognition with respect to the voice of the target person stored in advance, the voice data is recognized,
Recognizing the human intention;
(C) RecognizedFrom multiple types of artifacts in response to human externalized intentionsAt least oneStep to choose,
(D)This choiceShiArtifactofPosition and postureAutonomouslyDisplaceShown by recognized intentEyeTargetStep to move to the table.
[0016]
According to the program of the present invention, the present invention can be provided through download / installation from an information recording medium (package software or the like) or a communication network. Therefore, the present invention can be easily implemented with a microcomputer or the like mounted on various devices, and its versatility is improved.
[0017]
The following is a preferable aspect in the technical idea of the above-described “autonomous movement control method of an artificial object by recognizing human intention”.
[0018]
ThroughSupport information including map information including information that displays passages in facilities provided on the trust network, traffic information to consider when artifacts move, and weather informationBased on the captured map information, move at least one of a plurality of types of artifacts, and identify the destination person from the imaging data received from the moved artifacts, or the map information and the traffic information And automatic selection of the artifact based on either weather information. As a result, the eyesTargetIt will be possible to move it.
[0019]
  The artifactofPosition and postureAutonomouslyDisplace the eyesTargetMoveLetWhen this artifact isDetectedObstacles on the run ahead of the moveObstacle detection data aboutTheReceiveYouStepBypassing the floor and this detected obstacle,TargetArtificial objectsTheMove autonomouslyStageAnd a floor.
[0020]
  In this case, the resultTargetIt can be moved automatically and accurately to the position.
[0023]
  The artificial object displaces its position and posture autonomously,TargetWhen moving to a device, the control stage stores the movement path data based on the displacement of the movement position / posture, and the original movement start position by moving the movement path based on the movement position / posture displacement stored in reverse. And a control stage for returning to the position.
[0024]
In this case, convenience as a system that is more convenient to use is improved.
[0027]
The following is a preferable aspect in the technical idea of the above-mentioned “artificial object mobile communication system”.
[0028]
  Autonomous mobility control systemSupport information including map information with information to display passages in the facility provided on the communication network, traffic information to be taken into account when artifacts move, and weather informationTakeGainDomeansThePrepare for them.
  Further, the means for recognizing a human intention moves at least one of the plurality of types of artifacts based on the map information, and identifies the destination person from the imaging data received from the moved artifact. The means for identifying and selecting the artifact automatically selects the artifact based on any of the map information, the traffic information and the weather information.
[0029]
  Further, the control / driving means autonomously displaces the position / posture of the selected artifact,Shown by recognized intentEyeTargetWhen moving to a place, this artifact detects obstacles on the road ahead of the movement.And send obstacle detection dataDetection meansWith,The means for calculating and transmitting the movement instruction data uses the obstacle detection data,By bypassing the obstacle detected by the detection means,TargetArtifacts move autonomously on the tableCalculate movement instruction dataYouThe
[0034]
  Further, the control / drive meansBut, ArtifactofPosition and postureAutonomouslyDisplace the eyesTargetAnd at least two wheels for moving the device and rotational drive means for individually rotating the two wheels.
[0035]
  Said detection meansBut,It is configured with a photographic device.ObstacleImaging data as dataGenerateThe
[0036]
  Said detection meansIs configured to include a laser detection device, and the laser detection device detects a reflection intensity by laser transmission and reception, and detects the reflection intensity and a threshold value.ObstacledataGenerateThe
[0037]
In this case, the freedom degree of a structure improves.
[0041]
The following is a preferable aspect in the technical idea of the “program” described above.
[0042]
  SaidIntentionalIn the step of recognizing, processing for capturing support information including at least one of map information having information for displaying a passage in the facility provided from the communication network, traffic information, and weather information;Based on the map information, move at least one of the plurality of types of artifacts and identify the person to move from the imaging data received from the moved artifacts, or the map information, the traffic Automatic selection of the artifact based on either information or weather informationProcessing is executed by a computer.
[0043]
  The artifactofPosition and postureAutonomouslyDisplaceShown by recognized intentEyeTargetMoveLetWhen this artifact isDetectedObstacles on the run ahead of the moveReceive obstacle detection data forAnd bypassing the detected obstacleTargetThe computer further executes the step of autonomously moving the artifact on the table.
[0045]
  The artifactofPosition and postureAutonomouslyDisplace the eyesTargetThe step of storing the movement path data based on the displacement of the movement position / posture when moving to the device and the movement path based on the displacement of the stored movement position / posture reversely moved to return to the original position. And further causing the computer to execute the steps.
[0047]
According to the program of the present invention, the present invention can be provided through download / installation from an information recording medium (package software or the like) or a communication network. Therefore, the present invention can be easily implemented with a microcomputer or the like mounted on various devices, and its versatility is improved.
[0050]
In such a preferable aspect of the “program” described above, versatility is further improved.
[0051]
DETAILED DESCRIPTION OF THE INVENTION
Hereinafter, embodiments of the present invention will be described with reference to the drawings. In this description, the configuration and the arrangement relationship are schematically shown to such an extent that the present invention can be understood. Therefore, the present invention is not limited to the following embodiments, and can be applied to various forms based on the description of the scope of claims.
[0052]
(Basic functions of the first embodiment)
FIG. 1 is a block diagram for explaining functions in the first embodiment of an autonomous movement control method of an artifact by recognizing a human intention, a communication system thereof, and a program.
[0053]
  Referring to FIG. 1, in the basic function of the first embodiment, a plurality of types of artifacts are externalized by humans at remote positions by wireless control of an autonomous movement control system and a position / posture action unit. Vs intentRespondThen, the position / posture is selectively and autonomously displaced to move to a position (predetermined position) in the vicinity of the target moving person.
[0054]
  For this reason, the artificial object performs displacement (movement) of the position and posture in the autonomous movement to the position near the target human being. othersTheCaptured images and audio data from multiple types of artifactsTTransfer to the autonomous mobile control system by bidirectional communication. In addition, artificialThing, Due to displacement of position / postureTheNeighboring position of the target destination humanMove toMovementTo do. This movefor,Drive partOf the artifactBy detecting the position and orientation of the data by bidirectional communication,Position / posture action sectionForward to.
[0055]
  Autonomous movement control systemPhotographed image and soundThe inference engine and / or the search engine in the database (D / B) based on the data and the position / orientation of the drive unit detected, and the processing by the search engine are executed through the control system, and the human requested A moving artifact is selected and determined from a plurality of types to a position in the vicinity of the person to be moved.
[0056]
Furthermore, the autonomous movement control system transfers wireless control data to the position / posture action unit to the position near the target human being for the selected and decided artifact, and the position / posture control is continuously performed wirelessly. To run.
[0057]
  In this way, a plurality of types of artifacts are selected by inference based on the human's externalized intention (including search, and appropriately omits the expression of “search”), andAutonomousConvenient as an easy-to-use system that performs position / posture control and automatically moves to the position near the target human being, resulting in adapting various artifacts to various human beings. Improves.
[0058]
  Here, the autonomous mobile control system 4 and the autonomous mobile artifacts 1, 2, and 3 are configured separately.WhenHowever, it is also possible to adopt a configuration in which the two are integrated into an artifact.
[0059]
Similarly, in the second embodiment to the fourth embodiment, in the functional configuration and realization means (specific example), the autonomous movement control system 4 and the autonomous moving artifacts 1, 2, and 3 are described as separate bodies. An integral configuration of the control system 4 and the autonomous mobile artifacts 1, 2, 3 is also a design matter for those skilled in the art and is included in the present invention.
[0060]
Hereinafter, means (specific example) for realizing the basic functions of the first embodiment will be described.
[0061]
(Configuration of the first embodiment and individual operations of main parts)
FIG. 2 is an external view / block diagram showing a specific configuration example of the first embodiment of the present invention.
[0062]
Referring to FIG. 2, the configuration example of the first embodiment includes an autonomous mobile artifact 1 that is a wheelchair (also referred to as wheelchair 1 as appropriate) and an autonomous mobile artifact 2 that is a book (publication) transporter ( And a book transporter 2 as appropriate) and an autonomous moving artifact 3 as an umbrella stand (also indicated as an umbrella stand 3 as appropriate).
[0063]
Furthermore, the configuration example of the first embodiment includes an autonomous movement control system 4 (corresponding to support information acquisition means and information addition processing means in the present invention (claims)) and a motion capture system (for example, manufactured by Ascension Technology). As a position / posture action control device 5.
[0064]
The wheelchair 1 described above includes an artifact (wheelchair) control system 1a (also appropriately described as wheelchair control system 1a) and an artifact (wheelchair) drive system 1b (also appropriately described as wheelchair drive system 1b). The transporter 2 includes an artifact (book transporter) control system 2a (also referred to as a book transporter control system 2a as appropriate) and an artifact (book transporter) drive system 2b (as appropriate as a book transporter drive system 2b). Notation).
[0065]
Further, the umbrella stand 3 is also referred to as an artifact (umbrella stand) control system 3a (also referred to as an umbrella stand control system 3a as appropriate) and an artifact (umbrella stand) drive system 3b (as appropriate, referred to as an umbrella stand drive system 3b). ).
[0066]
The artifact control systems 1a to 3a and the artifact drive systems 1b to 3b correspond to the control / drive means and the information provision processing means in the present invention (claims).
[0067]
3 is a block diagram showing an example of the internal electrical configuration of the artifact control systems 1a to 3a of the wheelchair 1, the book transporter 2 and the umbrella stand 3 in FIG.
[0068]
Referring to FIG. 3, the artifact control systems 1a to 3a execute control here, and an MPU (Microprocessing Unit) 31 comprising a CPU, a ROM storing a program and a working RAM, and an autonomous movement control system 4 and a weak radio wave transmission / reception unit 32 for performing data transmission through a wireless section (wireless communication line) with the position / posture action control device 5 and imaging of a target person m (corresponding to a person in the present invention (claim)) An imaging device (optical system) 33 that outputs data (corresponding to the imaging means in the present invention (claim)), a solid-state memory (main storage device) 35 such as a flash memory that stores processing data, and processing data And a D / B device (auxiliary storage device) 36 using a hard disk device (HDD) or the like.
[0069]
The artificial object control systems 1a to 3a correspond to the sound collecting means in the microphone 37a (the present invention (Claim)) to which the sound of a call (movement request) from the wheelchair 1 to the umbrella stand 3 is input from the subject person m. ) And a sound processing unit 37b (which corresponds to the sound collecting means in the present invention (claim)) for amplifying and digitally converting (A / D) the sound signal from the microphone 37a.
[0070]
FIG. 4 is a block diagram showing an example of an internal electrical configuration of each of the artifact drive systems 1b to 3b from the wheelchair 1 to the umbrella stand 3 in FIG.
[0071]
Referring to FIG. 4, the artifact drive systems 1 b to 3 b include an imaging device (optical system) 47 a that outputs imaging data of an obstacle in front of running (moving) (obstacle detection means in the present invention (claim)). , Corresponding to the imaging processing means), a laser detection device 47b (corresponding to the obstacle detection means and wireless detection means in the present invention (claim)) for detecting the subject m in front or an obstacle, and Driving portions 48a and 49a (corresponding to the rotational driving means in the present invention (claim)) for controlling the power supply from the battery E for traveling, and the position / posture of the subject m are changed. Motors (M) 48b and 49b (corresponding to the rotation driving means in the present invention (claims)) for rotating the wheels.
[0072]
FIG. 5 is a block diagram showing an example of an internal electrical configuration of the autonomous mobile control system 4 in FIG.
[0073]
  Referring to FIG. 5, this autonomous mobile control system 4 corresponds to the MPU 51 (inference / search means / selection processing means in the present invention (claim)) comprising a CPU, RAM, and ROM for executing the control here. ), And a solid-state memory (main storage device) 53 such as a flash memory for storing processing data;Of the intention of the subject mWireless section connection (communication) between a database (D / B) device 54 for recognition (corresponding to the inference / retrieval means / selection processing means in the present invention (claim)) and the autonomous mobile control system 4 To perform a wireless section connection (for example, the above-described Bluetooth method) between the weak radio wave transmission / reception unit 55a for performing line connection / full-duplex communication (for example, Bluetooth method) and the position / posture action control device 5 by the weak radio wave. , And a D / B device (auxiliary storage device) 56 such as a hard disk device (HDD) for storing processing data.Intention recognitionD / B device 54 forThe intention of the subject m by inference and searchRecognitionKnowledgeAn inference processing unit 54a and a search processing unit 54b are provided.
[0074]
(Overall operation of the first embodiment)
FIG. 6 is a sequence diagram of the operation of the first embodiment, and FIG. 7 is a flowchart showing a processing procedure of the operations of the artifact control systems 1a to 3a and the artifact drive systems 1b to 3b of the first embodiment. .
[0075]
Moreover, FIG. 8 is a flowchart which shows the process sequence of the operation | movement of the autonomous mobile control system 4 of 1st Embodiment.
[0076]
Referring to FIGS. 2 to 5, 6, 7, and 8, the wheelchair 1, the book transporter 2, and the umbrella stand 3 (see FIG. 2) include a wheelchair control system 1 a, a book transporter control system 2 a, and The photographing direction of the photographing device 33 in the umbrella stand control system 3a (see FIG. 2 and FIG. 3) is set (installed so as to be able to photograph) so as to photograph the subject m. The captured image data (frame) is transferred to the autonomous mobile control system 4 through the weak radio wave transmission / reception unit 32 under the control of the MPU 31 (see steps S71 and S72 / FIG. 7—hereinafter “FIG. 7” is not shown).
[0077]
  In the autonomous movement control system 4, the weak electric power from the wheelchair control system 1a, the book transporter control system 2a, and the umbrella stand control system 3a.WaveThe signal is received by the weak radio wave transmission / reception unit 55a (see step S81 / FIG. 8—hereinafter “FIG. 8” is not shown).
[0078]
  next,Imaging received as weak radio wavesThe image data of the subject m is cut out from the data (step S82). A flag F1 is set on the extracted image data of the target person m and temporarily stored in the RAM of the MPU 51 (step S83). For the received signal from the weak radio wave transceiver 55aIntentionalRecognition is performed. First, by the control of the MPU 51 and the inference processing unit 54a and the search processing unit 54b in the D / B device 54, the inference engine based on the D / B data and the search engine are activated.Of the intention of the subject mRecognition is performed.
[0079]
ThisHere, as an implicit transmission on the body, the subject mPerson ofExternalizationFigureRecognize this by inferring and searching.
[0080]
Implicit bodily transmissions are those relating to physical movement, interpersonal contact, interpersonal distance, surrounding language, physical characteristics, clothing, ornaments, silence body and / or environmental attributes.
[0081]
  In these, hereinafter, image recognition (for example, the above-described physical movements and physical features) and voice recognition (for example, peripheral languages) are performed.IntentionalThis will be explained as recognition.
[0082]
In addition, as the above-mentioned “body motion and physical characteristics”, when calling the wheelchair 1 (request for movement), the “invitation” and “sleeping on the bed” state with respect to the wheelchair 1 placement direction, In the case of calling 2 (movement request), there is a state in which the target person m beckons with the hand holding the book while standing.
[0083]
In addition, as the above-mentioned “peripheral language”, there are utterances such as “wheelchair movement” and “book transporter movement” by the subject m.
[0084]
  IntentionalAs a recognition, the following processing examples (1) and (2) are taken up here.
[0085]
(1) The wheelchair 1, the book transporter 2 and the umbrella stand 3 have their respective functions (for example, the wheelchair 1 is used by the target person m who has difficulty walking, and the data relating to the function is the D / B shown in FIG. It is stored in advance in the device 54 or stored as needed, and similarly, data relating to the functions of the book transporter 2 and the umbrella stand 3 are stored.
[0086]
Data relating to this function includes the following data (1-A) and (1-B) for image recognition and voice recognition.
[0087]
(1-A) As image recognition data, imaging of the subject m is stored as a standard pattern. For this reason, (a) the imaging area of the target person m is cut out, and (b) normalization is performed to correct the size and inclination of the cut out imaging. Then, (c) feature point extraction is performed to compare the input image pickup (target person m) with the stored standard pattern with improved ease and certainty. Well-known features include local features such as the location and orientation of line components and edge components, and features such as area, perimeter, maximum width, and minimum width.
[0088]
  (1-B) As voice recognition data,Human intentionRegister words for recognition. The utterance word of the target person m is registered in advance in the MPU 51 and the D / B device 54 through the microphone 37a and the voice processing unit 37b. This is a time-series frequency spectrum and a standard pattern.
[0089]
  (2) The D / B device 54 processes the inference / retrieval for the above-described image recognition data and voice recognition data by the inference processing unit 54a, and retrieves the search for the image recognition data and the voice recognition data. Processed by the unit 54b and described nextIntentionalRecognition inference / retrieval (2-A), (2-B) is executed.
[0090]
(2-A) The processing for the data for image recognition stores the above-described imaging of the target person m as a standard pattern, cuts out the imaging area of the target person m, and determines the size, inclination, etc. Perform normalization to correct Then, feature points are extracted in order to improve and compare the ease and certainty between the input imaging (target person m) and the stored standard pattern.
[0091]
Then, the degree of similarity is determined by comparing the degree of similarity between the input image pickup (target person m) and the stored standard pattern, which is processing for the image recognition data. In this determination, it is well known to use the similarity based on the distance between vectors and the angle of two vectors in a multidimensional vector of a pattern as an index.
[0092]
(2-B) In the process for the voice recognition data, the MPU 51 and the D / B device 54 process the voice recognition of the voice of the subject m through the microphone 37a and the voice processing unit 37b. In this speech recognition process, the generated speech is recognized by comparing the input pattern (speech speech of the target person m) with the registered standard pattern. As a speech recognition method in this case, a DP (Dynamic Programming) matching method for efficiently performing the distance between input patterns while performing normalization in the time axis direction is well known.
[0093]
  In this example, the above image recognition and voice recognitionIntentionalThe MPU 51 executes a selection determination as to which of the wheelchair 1 to the umbrella stand 3 is to be moved (called) to the own position from the “intention / externalization of human being” by recognition (step S84). .
[0094]
  For example, in the state of “invitation” and “sleeping on a bed” with respect to the wheelchair 1 placement direction or when calling the book transporter 2 (movement request), this selection decision is made with a hand holding a book. The state in which the target person m beckons while standingSuchDo by.
[0095]
"When the “humanized external intention” can be recognized (step S85: Yes), that is, when the selection decision for the wheelchair 1, the book transporter 2 and the umbrella stand 3 is made from the human externalized intention, A flag F2 is set in the selection determination data and temporarily stored in the RAM of the MPU 51 (step S86).
[0096]
  In step S85IntentionalIf the recognition is not possible (No), the process returns to step S84 to repeat the inference / search.
[0097]
Next, when the MPU 51 fetches the position / posture recognition data from the position / posture action control device 5 through the weak radio wave transmission / reception unit 55b (step S87: Yes), the flag F3 is temporarily set to the position / posture recognition data. Thus, it is stored in the RAM of the MPU 51 (step S88).
[0098]
The position / posture recognition data has, for example, a data structure including the following contents (1) to (3).
(1) Image data of target person m (flag F1)
(2) Selection decision data (flag F2) indicating one between the selected wheelchair 1 and the umbrella stand 3
(3) Current position / posture data from the selected wheelchair 1 to the umbrella stand 3, in other words, the umbrella from the selected wheelchair 1 in the rotational drive of the motors (M) 48b, 49b from the drive units 48a, 49a. Position / posture recognition data (flag F3) indicating the amount of movement of the stand 3 from the start of travel and the direction corresponding to the travel direction
[0099]
Next, the respective data of the flag F1 (image data of the subject m), the flag F2 (selection determination data), and the flag F3 (position / posture recognition data) are read out, and the subject m (flag in the image data) is read. In F1), movement instruction data (its position / posture) of the umbrella stand 3 (selection decision data / flag F2) is calculated from the selected wheelchair 1 (step S89).
[0100]
The calculation of the movement instruction data has the following contents (1) to (4), for example.
(1) Judgment (calculation) of whether the image data (flag F1) of the subject m in the photographing apparatus 33 has been enlarged as a result of movement and has become a predetermined size.
(2) When the image data of the subject m (corresponding to the flag F1) in the photographing device 33 has a predetermined size and the detection by the laser detection device 47b indicates a predetermined immediately preceding value.
(3) Selection decision data indicating the umbrella stand 3 from the selected wheelchair 1 (corresponding to the flag F2)
(4) Position / posture recognition data (corresponding to flag F3) for moving the umbrella stand 3 (corresponding to selection determination data / flag F2) from the selected wheelchair 1 toward the subject m
[0101]
  The movement instruction data calculated in this way (including selection result data from the wheelchair 1 to the umbrella stand 3)Autonomous movement control system 4Is transmitted from the weak radio wave transmission / reception unit 55b to the position / posture action control device 5 (step S90).The movement instruction data isPosition / posture action control device 5Et al.The MPU 31 that is transferred from the wheelchair 1 to the umbrella stand 3 and receives the movement instruction data through the weak radio wave transmission / reception unit 32 is a drive unit of any one of the wheelchair drive system 1b, the book transporter drive system 2b, and the umbrella stand drive system 3b. Drive control signals are output to 48a and 48b.
[0102]
The motors 48b and 49b operate by energization from the drive units 48a and 49a, and the umbrella stand 3 starts moving from the selected wheelchair 1 toward the subject m (step S74).
[0103]
The umbrella stand 3 from the selected wheelchair 1 at the start of the movement moves with a posture with a predetermined position in front. That is, it is the direction of use by the target person m, which differs depending on the functions of the respective artifacts from the wheelchair 1 to the umbrella stand 3. For example, since the subject person sits from the front, the wheelchair 1 moves in a direction in which the front faces the subject m at the time of arrival (movement stop).
[0104]
The book transporter 2 is limited to storing and taking out publications from one direction. That is, since the publication is stored and taken out (withdrawal) with the lid open, the loading / unloading portion moves in such a posture that the front faces the subject m at the time of arrival (movement stop).
[0105]
Moreover, since the umbrella stand 3 can be stored and taken out (in / out) of the umbrella from any peripheral direction, the umbrella stand 3 moves with a posture regardless of the direction at the time of arrival (movement stop).
[0106]
A drive system (wheelchair drive system 1b, book transporter drive system 2b, umbrella stand drive system sampled at regular intervals under the control of the MPU 31 while any one of the wheelchair 1 to the umbrella stand 3 is moved as described above. Data indicating the motion state (that is, position / posture) of any of 3b) is transmitted to the position / posture action control device 5 (step S75).
[0107]
This transmission corresponds to the closed loop control. When the motors 48b and 49b are, for example, stepping motors, the open loop control is executed. In other words, the above transmission is not performed.
[0108]
Furthermore, any one of the wheelchair 1 to the umbrella stand 3 moves, while moving, transmits imaging data of the subject m from the imaging device 33 sampled at regular intervals under the control of the MPU 31 to the position / posture action control device 5 (step) S76).
[0109]
Data indicating the movement state and imaging data of the subject m are transferred from the position / posture action control device 5 to the autonomous movement control system 4.
[0110]
When the autonomous movement control system 4 takes in the data indicating the movement state (position / posture data) and the imaging data of the subject m (steps S91, S92), the MPU 51 determines whether the imaging is at a predetermined position / size. Is determined (step S93). That is, it is determined whether any one of the wheelchair 1 and the umbrella stand 3 that has been selected and decided has been moved to the vicinity of the subject m.
[0111]
  Here, when any of the wheelchair 1 and the umbrella stand 3 that are determined and not determined to be in a predetermined position / size has not moved to a position near the subject m (step S93: No), movement instruction data (position / posture) ) Is repeated from step S89. In the case of a predetermined position / size, that is, when any of the wheelchair 1 that has been selected and decided has moved to a position near the subject m (step S93: Yes), movement stop data is generated,Autonomous movement control system 4FromPosition / attitude action control device 5(Step S94).
[0112]
  Position / attitude action control device 5The MPU 31 in any of the artifact control systems 1a to 3a from the wheelchair 1 to the umbrella stand 3 that has received the transfer of the movement stop data from the vehicle through the weak radio wave transmission / reception unit 32 includes a wheelchair drive system 1b, a book transporter drive system 2b, A drive control signal is output to any one of the drive units 48a and 49a of the umbrella stand drive system 3b, and the control of the position and posture is stopped (step S78).
[0113]
  (Advantages of the first embodiment)
  As described above, in the first embodiment, the intention of the target person m being externalized from the wheelchair 1, the book transporter 2, and the umbrella stand 3, which are a plurality of types of artifacts.In the figurePerforms selection and position / posture control based on inference / search based on the object, and the artificial object can move autonomously to a position near the target person m (the closest to the person m). As a result, convenience as a system that can adapt various artifacts to various people and is easy to use is improved.
[0114]
(Basic functions of the second embodiment)
FIG. 9 is a block diagram for explaining functions in the second embodiment.
[0115]
Referring to FIG. 9, in the function of the second embodiment, “support information providing system (map information providing system in facility, traffic information providing system, weather information providing” is newly added to the function of the first embodiment in FIG. System) "has been added.
[0116]
  Incorporating information from this support information provision system, for example, in the case of rain, an umbrella stand as an artifact is externalized by a human being at a position away from multiple types of artifacts.In the figureMake the corresponding selection. In this way, the accuracy of selection is improved. Other functions are the same as those in the first embodiment, and redundant description thereof is omitted.
[0117]
A specific example of the function of the second embodiment will be described below.
[0118]
(Configuration of Second Embodiment and Individual Operation of Main Part)
FIG. 10 is a block diagram illustrating a specific example of a main part of the second embodiment.
[0119]
Referring to FIG. 10, in the specific example of the second embodiment, the specific example of the first embodiment of FIG. 2 (autonomous mobile artifacts 1, 2, 3, autonomous mobile control system 4, position / posture action control device 5). ), A support information providing system 6 and a wireless communication network 7 are newly added. The wireless communication network 7 operates in the same manner in the wired communication network.
[0120]
The support information providing system 6 is a map information providing system, a traffic information providing system, or a weather information providing system as described with reference to FIG.
[0121]
The map information providing system includes, for example, map information known for traveling in GPS navigation. It also provides information that displays the passages within the facility (eg, building).
[0122]
The traffic information providing system provides traffic information (for example, road traffic stop information, railway seasonal operation information, building establishment / demolition information) (for example, road traffic information communication system / VICS: Vehicle Information and Communication System). .
[0123]
The meteorological information providing system provides weather information such as weather conditions and weather forecasts in each place (for example, weather data comprehensive processing system / CSMS: Computer System for Meteorological Services).
[0124]
Other configurations are the same as those in FIGS. 3 to 5 in the first embodiment. The duplicate description here is omitted.
[0125]
(Overall operation of the second embodiment)
FIG. 11 is a sequence diagram in the second embodiment, and FIG. 12 is a flowchart showing a main processing procedure of the operation of the autonomous mobile control system 4 in the second embodiment.
[0126]
Here, only different portions from the first embodiment will be described.
[0127]
  Referring to FIGS. 10 to 12, in the main processing in the operation of the second embodiment, the autonomous mobile control system 4 performs wheelchair control in the same manner as in the first embodiment (see FIGS. 6 and 8). Weak electricity from system 1a, book transporter control system 2a and umbrella stand control system 3aWaveNext, the image data of the subject m is cut out. The extracted image data of the target person m is stored.
[0128]
  After this,Of the intention of the subject mIn the inference processing unit 54a and the search processing unit 54b in the D / B device 54 for detection, the inference engine and the search engine are activated based on the D / B data.IntentionalRecognition is performed. like thisIntentionalBased on the recognition, a determination is made as to which of the wheelchair 1 to the umbrella stand 3 the target person m wants to move (call) to his / her own position (this process is the same as steps S81 to S85 of FIG. 8 in the first embodiment). ).
[0129]
  like thisIntentionalIn recognition, in the second embodiment, as shown in FIGS.The autonomous mobile control system 4Support information provision system 6FromCapture necessary support information.
[0130]
The incorporation of the necessary support information has the following contents (1) to (3).
[0131]
(1) From the map information providing system in the facility, for example, information for displaying the passage in the facility (in the nursing facility or in the library) is captured. The information in the care facility is used, for example, when the subject m calls the wheelchair 1 or the umbrella stand 3 placed in the shadow etc. only by speaking from a place where the subject cannot see.
[0132]
  In this case, the inference processing unit 54a and the search processing unit 54b in the D / B device 54 are based only on utterance information (when captured image data cannot be obtained).IntentionalRecognize. At this time, the inference processing unit 54a and the search processing unit 54b in the D / B device 54 recognize the position of the uttered target person m, refer to the map information in the taken care facility,At least one of multiple types of artifactsMove slightly forward / backward / left / rightLet,Moved artifact hasThe target person m registered in advance is identified from the imaging data obtained by the imaging device 33 (FIG. 3). The subsequent movement to the target position m of the target movement destination is the same as described above.
[0133]
The map information in the library is also used for moving the book transporter 2 and the umbrella stand 3 in the same manner as the map information in the care facility.
[0134]
(2) The above-mentioned traffic information (for example, road traffic stop information, railway seasonal operation information, new building construction / demolition information) is captured from the traffic information providing system.
[0135]
  For example, the “must” that the subject m calls at the entrance to the outside is performed, and the inference processing unit 54a and the search processing unit 54b in the D / B device 54IntentionalIn the inference / retrieval by recognition, when the traffic stop information of the road to the facility where the target person m exists is taken in, in this example, the wheelchair 1 is not taken out by the vehicle, and the selection excluding the wheelchair 1 is selected. I do. Furthermore, the selection which excluded the book transporter 2 is performed on the assumption that a book is not carried in and out. As a result, inference and search for calling the umbrella stand 3 are performed.
[0136]
(3) From the weather information provision system, the sky around the subject m
Capture weather information such as weather conditions and weather forecasts. The inference processing unit 54a and the search processing unit 54b in the D / B device 54 execute inference and search utilizing the weather information.
[0137]
  For example, in the case of rain, the umbrella stand 3 is intended to be externalized by the subject m at a remote location.In the figureSelect accordingly. On sunny days, umbrella stand 3IntentionalIn recognition, the selection is excluded. In other words, the wheelchair 1 is preferentially selected on a sunny day.
[0138]
  Support information providing system in this way6Based on support information fromIntentionalRecognition is performed.
[0139]
The operations of the other autonomous moving artifacts 1, 2, 3 (wheelchair 1 to umbrella stand 3), autonomous movement control system 4 and position / posture action control device 5 are the same as those in the first embodiment.
[0140]
(Advantages of the second embodiment)
As described above, in the second embodiment, in addition to the advantages of the first embodiment described above (an improvement in convenience that can adapt various artifacts to various people), a plurality of types of wheelchairs 1 and a book transporter 2 are used. And the accuracy of selection by inference / search from the umbrella stand 3 is improved.
[0141]
(Basic functions of the third embodiment)
FIG. 13 is a block diagram for explaining functions in the third embodiment.
[0142]
Referring to FIG. 13, in the basic function of the third embodiment, as in the first embodiment described with reference to FIG. 1, a plurality of types of artifacts are wirelessly connected to the autonomous movement control system and the position / posture action unit. By control, the man-made object selectively and autonomously displaces its position / posture to automatically move to the position near the target person m in response to the externalized intention of a human at a distant position. To do.
[0143]
In this automatic movement, the artificial object detects an obstacle in traveling ahead of the movement, and performs control for displacing the position and posture for moving around the obstacle. The subsequent description is the same as the basic functional description of the first embodiment.
[0144]
  In the basic function of this third embodiment,ImagingTransfer of data and detection data of the position / orientation of the driving unit, selection determination from a plurality of types of artifacts requested by the subject by the inference engine and / or search engine, and the purpose for the selection-decided artifact Position / posture control for the artifact up to the position near the human being to be moved is continuously executed.
[0145]
In this way, one of the wheelchair 1 and the umbrella stand 3 which are a plurality of kinds of artifacts is selected by inference and search based on the externalized intention of the subject m, and the position / posture control is performed. , Automatically moves to a position near the subject m. At this time, the artificial object detects obstacles on the road ahead of the movement, performs the detour, controls the position / posture to the position near the target human being, and autonomously Move appropriately.
[0146]
A specific example of the function of the third embodiment will be described below.
[0147]
(Configuration of Third Embodiment and Individual Operation of Main Part)
The specific configuration of the third embodiment is the same as the specific example of the first embodiment shown in FIG. 2 (autonomous movement artifacts 1, 2, 3, autonomous movement control system 4, position / posture action control device 5). is there. The transmission sequence, processing flow, etc. are different.
[0148]
Other configurations are the same as those described with reference to FIGS. 3 to 5 in the first embodiment. The duplicate description here is omitted.
[0149]
(Overall operation of the third embodiment)
FIG. 14 is a sequence diagram in the third embodiment. Moreover, FIG. 15 is a flowchart which shows the principal process procedure of operation | movement of the artifact control system 1a-3a of 3rd Embodiment, and the artifact drive system 1b-3b, FIG. 16 is the autonomous movement in 3rd Embodiment. 4 is a flowchart showing a main processing procedure of an operation of the control system 4.
[0150]
Here, the part (running by the above-mentioned obstacle detection) different from FIGS. 6 to 8 of the first embodiment will be mainly described.
[0151]
Referring to FIGS. 2 to 5, 14, and 15, the umbrella stand 3 from the selected wheelchair 1 transfers the captured image data to the autonomous movement control system 4, and the wheelchair is driven by the movement instruction data. A drive control signal is output to the drive units 48a and 49a of the system 1b, the book transporter drive system 2b, and the umbrella stand drive system 3b (step S71 in FIG. 7). The motors 48b and 49b operate by energization from the drive units 48a and 49a, and the umbrella stand 3 starts moving from the selected wheelchair 1 toward the subject m (steps S72, S73, and S74 in FIG. 7).
[0152]
Here, obstacle detection will be described with reference to FIGS. 14A, 14B, and 15. FIG.
[0153]
This obstacle detection is performed by (a) imaging data of the imaging device 47a shown in FIG. 4 and (b) transmission of laser waves by the laser detection device 47b and presence / absence / level of reflected waves.
[0154]
Note that (a) the imaging device 47a and (b) the laser detection device 47b are used in an environment where the artificial object mobile communication system is applied, for example, in a moving range within a facility, in a situation where obstacles occur during travel (human (A) a multiple-person room in a hospital or the like where movement is expected) employs both (a) the imaging device 47a and (b) the laser detection device 47b.
[0155]
In an environment where the artifact mobile communication system is applied, in a single room in a hospital where human movement is extremely small, one of (a) an imaging device 47a and (b) a laser detection device 47b is employed.
[0156]
The imaging device 47a detects an obstacle by, for example, image recognition on imaging data with a focus of 1 meter (m) ahead. In this case, pattern recognition using obstacles other than the imaging data of the subject m is executed.
[0157]
Note that the pattern recognition processing in this case is the same as the image recognition described in the first embodiment, and redundant description thereof is omitted here.
[0158]
The laser detection device 47b starts laser transmission and reception at the same time as any one selected from the wheelchair 1 in the umbrella stand 3 starts moving, detects the reflection intensity, and is detected first by the MPU 31 (FIG. 3). The target direction is calculated while being aware of the direction of the subject m to be moved.
[0159]
In this case, when the obstacle is detected, the obstacle may be a passerby who passes by directly, so the traveling is stopped for a certain period of time (the degree that the person passes) (step S101).
[0160]
In this obstacle detection process, when no obstacle is detected (step S102: No), the position / posture data transfer is processed as in the first embodiment shown in FIG. 7 (step S75). Further, when an obstacle is detected in the obstacle detection process in step S101 (step S102: Yes), the obstacle detection data (determined from the reflection level and threshold value at the laser detection device 47b) is autonomous movement control system. 4 (transmission) (step S103). Thereafter, steps S76 to S78 are processed in the same manner as in the first embodiment shown in FIG. 7, and the process proceeds to a return process described below (step S104).
[0161]
For the processing from the wheelchair 1 to the umbrella stand 3 selected in this way, the autonomous movement control system 4 performs steps S81 to S86 (see FIG. 8) in FIGS. 2 to 5, 14 and 16 (see). The routine (see the routine of the first embodiment) is processed to determine whether the obstacle detection data is captured (step S111). When the obstacle detection data is taken in (step S111: Yes), a flag F21 is set on the obstacle detection data and temporarily stored in the RAM of the MPU 51 (step S112).
[0162]
Thereafter, similarly to the first embodiment of FIG. 8, steps S87 and S88 are processed to generate movement instruction data (step S89A).
[0163]
In this step S89A, similarly to the first embodiment of FIG. 8 described above, the following data (1) to (4) are processed to generate movement instruction data.
(1) Image data / flag F1 of the subject m
(2) Selection decision data / flag F2 indicating the selected wheelchair 1, book transporter 2 and umbrella stand 3
(3) Current position / posture data / flag F3 from the selected wheelchair 1 to the umbrella stand 3
(4) Calculation of movement instruction data (position / posture) for traveling around obstacles (obstacle detection data / flag F21)
[0164]
The movement instruction data (including selection result data from the wheelchair 1 to the umbrella stand 3) calculated in this way is sent from the autonomous movement control system 4 to the position / posture action control device 5 as in the first embodiment shown in FIG. In the same manner as in the first embodiment, one of the umbrella stands 3 moves from the wheelchair 1 by selection determination to reach the subject m (see steps S90 to S94 in FIG. 8 of the first embodiment). ).
[0165]
(Advantages of the third embodiment)
Thus, in the third embodiment, any one of the wheelchair 1 to the umbrella stand 3 which is a plurality of types of artifacts is selected by inference / search based on the externalized intention of the subject m, and thereafter Position / posture control is performed to automatically move to the target position of the target person m. At this time, any one of the wheelchair 1 to the umbrella stand 3 detects an obstacle in traveling ahead of the movement, and controls the position / posture to a position near the target human being while performing the detour. Will be able to perform the appropriate autonomous movement.
[0166]
Furthermore, the artificial object can return to the original position by moving the movement path by the displacement of the position / posture of the movement to the original position. In other words, it is more convenient and the convenience is further improved.
[0167]
(Basic functions of the fourth embodiment)
FIG. 17 is a block diagram for explaining functions in the fourth embodiment.
[0168]
Referring to FIG. 17, in the function of the fourth embodiment, a new “support information providing system (map information providing system, traffic)” is added to the function of the third embodiment of FIG. 13 to FIG. Information provision system and weather information provision system) ”have been added.
[0169]
  Incorporating information from this support information provision system, for example, in the case of rain, an umbrella stand as an artifact is externalized by a human being at a position away from multiple types of artifacts.In the figureSelect correspondingly and travel around the obstacle.
[0170]
In this way, the accuracy of selection from a plurality of types of artifacts corresponding to the human's external intention is improved, and the vehicle travels to a position near the target human being while bypassing the obstacle. To do. Other functions are the same as those in the third embodiment, and a duplicate description thereof is omitted.
[0171]
Hereinafter, a specific example of the function of the fourth embodiment will be described.
[0172]
(Configuration of Fourth Embodiment and Individual Operation of Main Part)
The configuration of the fourth embodiment is the same as the configuration of the third embodiment described above (obstacle detection travel / specific example of the first embodiment of FIG. 1: autonomous artifacts 1 to 3, autonomous movement control system 4, position In addition to the attitude action control device 5), the support information providing system 6 and the wireless communication network 7 described in the second embodiment are added.
[0173]
The support information providing system 6 is a map information providing system, a traffic information providing system, and a weather information providing system in the facility as described with reference to FIG. 9 (refer to the second embodiment for details thereof).
[0174]
Other configurations are the same as those in FIGS. 3 to 5 in the first embodiment. The duplicate description here is omitted.
[0175]
(Overall operation of the fourth embodiment)
FIG. 18 is a sequence diagram in the fourth embodiment.
[0176]
  Referring to FIG. 18, in the sequence of the fourth embodiment, as in the sequence of the third embodiment shown in FIG. 14, when any one of the wheelchair 1 that is a plurality of types of artificial objects travels and travels. In addition, the vehicle travels to a position in the vicinity of the target subject m by detouring while detecting an obstacle. In this case, based on the support information from the support information providing system 6 as in the second embodiment shown in FIG.IntentionalRecognizing that the externalization of the subject mIn the figureCorrespondingly, one of wheelchair 1 to umbrella stand 3 is selected.
[0177]
The other operations of the autonomous movement control system 4 and the position / posture action control device 5 are the same as those in the second embodiment and the third embodiment, and redundant description thereof is omitted.
[0178]
(Advantages of the fourth embodiment)
As described above, in the fourth embodiment, any one of the wheelchair 1 to the umbrella stand 3 which is a plurality of kinds of artifacts is selected by inference / search based on a human externalized intention, and the position / Attitude control is performed to automatically move to a position near the subject m. At this time, any one of the wheelchair 1 to the umbrella stand 3 detects an obstacle on traveling ahead of the movement, and moves to a position near the target person m while performing a detour. Therefore, it becomes possible to perform an appropriate autonomous movement.
[0179]
(Modification of the embodiment)
In the first to fourth embodiments described above, an example in which any one of the wheelchair 1 from the wheelchair 1 is selectively moved has been described. However, a plurality of artifacts (for example, two from the wheelchair 1 to the umbrella stand 3 Can be moved simultaneously (in parallel).
[0180]
  in this case,IntentionalAt the time of recognition, similar determination is made based on this recognition (for example, the wheelchair 1 and the book transporter 2 are selected simultaneously). Furthermore, in order to avoid a collision while the selected artificial object is traveling, the traveling (see the third and fourth embodiments) based on the detection of the front obstacle described above may be applied.
[0181]
In the first to fourth embodiments described above, traveling has been described with two systems of drive units 48a and 49a and wheel rotation motors (M) 48b and 49b (see FIG. 4). The above-described traveling and attitude control may be executed by employing the motor for rotating the wheel and the wheel (M). In this case, it becomes possible to travel more precisely and a heavy artificial object.
[0182]
In the first to fourth embodiments described above, the book transporter 2 can be configured to automatically open and close the lid. As a result, the book stored during traveling is prevented from popping out, and the lid is automatically opened when traveling is completed, further improving the convenience of use.
[0183]
As this structure, the book transporter 2 is comprised by a cover part, an accommodating part, a cover opening / closing mechanism part, a cover opening / closing mechanism control part, etc.
[0184]
In the book transporter 2 having this configuration, the lid opening / closing mechanism section and the lid opening / closing mechanism control section are operated during traveling that automatically moves toward the subject m, and the lid section is closed. The opening / closing mechanism unit and the lid opening / closing mechanism control unit operate to execute control for automatically opening the lid.
[0185]
In the first to fourth embodiments described above, the wireless section has been described using a communication line using weak radio waves. However, the embodiment can also be performed using an infrared communication line. This infrared communication line needs to be designed in consideration of the transmission range (service area) due to directivity.
[0186]
Such modifications of the first to fourth embodiments are design matters that can be easily considered by those skilled in the art, and are all included in the present invention.
[0187]
【The invention's effect】
  As is clear from the above description, according to the autonomous movement control method of an artificial object based on the recognition of the human intention of the present invention, and its communication system and program, a plurality of types of artificial objects are externalized. Selectively by inference and search based on intentionAutonomousPosition / posture control is performed, and it becomes possible to move autonomously to a human position (predetermined range position).
[0188]
As a result, there is an effect that convenience as a system that is easy to use and can adapt various artifacts to various people is improved.
[Brief description of the drawings]
FIG. 1 is a block diagram for explaining functions in a first embodiment of the present invention.
FIG. 2 is an external view / block diagram showing a specific configuration example of the first embodiment of the present invention.
FIG. 3 is a block diagram showing an example of an internal electrical configuration of the artifact control system in FIG. 2;
4 is a block diagram showing an example of an internal electrical configuration of the artifact drive system in FIG. 2. FIG.
5 is a block diagram showing an example of an internal electrical configuration of the autonomous mobile control system in FIG. 2. FIG.
FIG. 6 is a sequence diagram of the operation of the first embodiment.
FIG. 7 is a flowchart showing a processing procedure of operations of the artifact control system and the artifact drive system of the first embodiment.
FIG. 8 is a flowchart showing a processing procedure of an operation of the autonomous mobile control system of the first embodiment.
FIG. 9 is a block diagram for explaining functions in the second embodiment;
FIG. 10 is a block diagram showing a specific example of a main part of the second embodiment.
FIG. 11 is a sequence diagram in the second embodiment.
FIG. 12 is a flowchart showing a main processing procedure of an operation of the autonomous mobile control system in the second embodiment.
FIG. 13 is a block diagram for explaining functions in the third embodiment.
FIG. 14 is a sequence diagram in the third embodiment.
FIG. 15 is a flowchart illustrating a main part processing procedure of operations of the artifact control system and the artifact drive system of the third embodiment.
FIG. 16 is a flowchart showing the main part processing procedure of the operation of the autonomous mobile control system in the third embodiment.
FIG. 17 is a block diagram for explaining functions in the fourth embodiment;
FIG. 18 is a sequence diagram in the fourth embodiment.
[Explanation of symbols]
1-3 Autonomous mobile artifacts (wheelchairs, book transporters, umbrella stands)
1a Artifact (wheelchair) control system (wheelchair control system)
1b Artifact (wheelchair) drive system (wheelchair drive system)
2a Artifact (book transporter) control system (book transporter control system)
2b Artifact (book transporter) drive system (book transporter drive system)
3a Artifact (Umbrella Stand) Control System (Umbrella Stand Control System)
3b Artifact (Umbrella Stand) Drive System (Umbrella Stand Drive System)
4 Autonomous movement control system
5 Position / Attitude Action Control Device
31, 51 MPU
32 Weak radio wave transmitter / receiver
33, 47a Imaging device
36,54 D / B equipment
37a microphone
37b Voice processing unit
47b Laser detector
48a, 49a Drive unit
48b, 49b Motor (M)
54a Inference processing unit
54b Search processing unit
55a, 55b Weak radio wave transmitter / receiver
m Target audience

Claims (17)

This is an autonomous movement control method for an artificial object based on recognition of human intentions, for selective movement of multiple types of artificial objects , implemented in a communication system including multiple types of artificial objects and an autonomous movement control system. And
The autonomous mobile control system is
(A) at least one of captured of a plurality of types of artifacts, and stages that will receive either one or both of the imaging data and audio data to human spaced locations,
(B) Using one or both of the received imaging data and audio data,
When using the imaging data, by comparing the degree of similarity between the pre-stored standard pattern for image recognition including physical movements and physical features and the imaging data, the similarity is determined, Or
When using the voice data, by comparing the voice data with a standard pattern for voice recognition with respect to the voice of the target person stored in advance, the voice data is recognized,
Recognizing the human intention;
A stage you automatically selected at least one from a plurality of kinds of artifacts (c) recognized to correspond to the intention and externalizing the human
; (D) selected by autonomously displacing the position and orientation of the artifact, the stage is moved in purpose position indicated by the recognized intended
Autonomous movement control method for artifacts due to the recognition of human intention, characterized by the execution. When the autonomous mobile control system recognizes the human intention,
Capturing map information including information indicating passages in facilities provided on a communication network, support information including traffic information and weather information for consideration when an artifact moves , and
Moving at least one of the plurality of types of artifacts based on the map information, and identifying the destination human from imaging data received from the moved artifacts;
The method for autonomously moving an artifact by recognizing a human intention according to claim 1, further comprising : When the autonomous movement control system automatically selects at least one of a plurality of types of artifacts,
  The artificial object according to claim 2, wherein the artificial object is automatically selected based on any one of the map information, the traffic information, and the weather information included in the support information. Autonomous movement control method of things. And autonomously displacing the position and orientation of the artifact, when Before moving to the purposes position,
The autonomous mobile control system is
A stage that will receive obstacle detection data for which the artifact is detected, an obstacle on the travel of the mobile front,
To bypass the obstacles it detects, and the stage to move autonomously artifacts to the purpose position
Autonomous movement control method for artifacts due to the recognition of human intention according to any one of claims 1 to 3, characterized by the further execution. The artifacts by displacing the autonomous position and posture, when moved to the purposes position,
A stage of control for storing movement path data according to the displacement of the movement position and posture;
The stage of control for returning to the original position from which the movement path by the displacement of the stored movement position / posture is moved backwards,
The autonomous movement control method for an artifact by recognizing a human intention according to claim 1, further comprising: Includes an autonomous motion control system and a plurality of types of artifacts, by the recognition that includes an inference-search for intended that externalized human, artifacts more is an artifact mobile communication system which moves autonomously,
At least one of the plurality of types of artifacts, the audio data for recognition of prior SL and outputs the recognition to order of imaging data intended externalized was between human imaging means and said intention of spaced locations One or both of the output sound collecting means and control / driving means for moving to the target position indicated by the recognized intention based on the movement instruction data received from the autonomous movement control system > And
The autonomous mobile control system is
It means you recognize the intention of the person from one or both of the imaging data and the audio data received from the artifact,
Hand stage you select artifacts corresponding to intent recognized before Symbol plural kinds,
Means for transmitting to calculate the movement instructing data Ru displaces the position and posture of the front hexene-option the artifact
Artifacts mobile communication system comprising: a. The autonomous mobile control system is :
Provided from the communication network, the map information having the information for displaying the path in the facility, the resulting manually setup-assistance information including traffic information and weather information for consideration when artifact is moved further Prepared ,
The means for recognizing the human intention moves at least one of the plurality of types of artifacts based on the map information and identifies the destination person from the imaging data received from the moved artifact. artifacts mobile communication system according to claim 6, characterized in <br/> that. The means for selecting the artifact included in the support information automatically selects the artifact based on any of the map information, the traffic information, and the weather information.
The artifact mobile communication system according to claim 7, wherein: The artificial object comprises a detecting means for detecting an obstacle on the run ahead of movement and transmitting obstacle detection data ,
It means for transmitting to calculate the movement instructing data to which the autonomous moving control system included in the using obstacle detection data, bypassing the obstacle detected by the detecting means, artifacts on the purposes position The movement instruction data for moving autonomously is calculated . The artificial object mobile communication system according to any one of claims 6 to 8, wherein the movement instruction data is calculated . The control / drive means is
And autonomously displacing the position and orientation of the artifact, for moving the purposes position, comprise at least two wheels and, a rotary drive means for rotating the two wheels individually The artifact mobile communication system according to claim 6 . The detection means is configured to include a photographing device,
Artifacts mobile communication system according to claim 9 wherein the imaging device, wherein the benzalkonium generates imaging data as the obstacle data. The detection means comprises a laser detection device;
The laser detection device, the laser transceiver detects the reflection intensity, that generates the obstacle data from the reflection intensity with a threshold value
Artifacts mobile communication system according to claim 9, wherein the this. Humans who are able to move multiple types of artifacts selectively and autonomously by recognition including inference and search for human externalized intentions in a communication system with multiple types of artifacts and autonomous movement control system A program comprising the following steps for causing a computer to execute autonomous movement control of an artifact by inferring and searching for an externalized intention of the object.
(A) receiving either one or both of imaging data and audio data for a human at a separated position captured by at least one of a plurality of types of artifacts;
(B) Using one or both of the received imaging data and audio data,
When using the imaging data, by comparing the degree of similarity between the pre-stored standard pattern for image recognition including physical movements and physical features and the imaging data, the similarity is determined, Or
When using the voice data, by comparing the voice data with a standard pattern for voice recognition with respect to the voice of the target person stored in advance, the voice data is recognized,
Recognizing the human intention;
(C) selecting at least one of a plurality of types of artifacts corresponding to the recognized externalized intention of the human,
Step (d) that said selected by autonomously displacing the position and orientation of the artifact, is moved to the purpose position indicated by the recognized intended. In the step of recognizing the intention,
A process of capturing support information including at least one of map information having information for displaying a passage in a facility provided from a communication network, traffic information, and weather information;
Based on the map information, a process of moving at least one of the plurality of types of artifacts and identifying the destination person from imaging data received from the moved artifacts ;
14. The program according to claim 13 for causing a computer to execute. In the step of selecting at least one of the plurality of types of artifacts,
  Processing for automatically selecting the artifact based on any of the map information, the traffic information, and the weather information included in the support information
15. The program according to claim 14 for causing a computer to execute. And autonomously displacing the position and orientation of the artifact, when Before moving to the purposes position,
Receiving a obstacle detection data for which the artifact is detected, an obstacle on the travel of the mobile front,
Bypassing the obstacle it detects a step of causing autonomously move artifact to the purposes position,
The program as described in any one of Claims 13-15 for making a computer perform further. And storing the moving path data by the displacement of said autonomously displacing the position and orientation of the artifact, the position and orientation of the moving the time of automatically moved to the purposes position,
Reversing the movement path by the displacement of the stored movement position / posture and returning to the original position,
The program according to claim 13 , further causing a computer to execute.

Images