WO1998020398A1 - Appareil permettant de prevenir la collision d'un objet en mouvement avec un obstacle - Google Patents

Appareil permettant de prevenir la collision d'un objet en mouvement avec un obstacle Download PDF

Info

Publication number
WO1998020398A1
WO1998020398A1 PCT/JP1997/004044 JP9704044W WO9820398A1 WO 1998020398 A1 WO1998020398 A1 WO 1998020398A1 JP 9704044 W JP9704044 W JP 9704044W WO 9820398 A1 WO9820398 A1 WO 9820398A1
Authority
WO
WIPO (PCT)
Prior art keywords
obstacle
moving object
obstacles
sequentially
time
Prior art date
Application number
PCT/JP1997/004044
Other languages
English (en)
Japanese (ja)
Inventor
Hiroyoshi Yamaguchi
Tetsuya Shinbo
Osamu Yoshimi
Seiichi Mizui
Original Assignee
Komatsu Ltd.
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Komatsu Ltd. filed Critical Komatsu Ltd.
Priority to AU48847/97A priority Critical patent/AU4884797A/en
Publication of WO1998020398A1 publication Critical patent/WO1998020398A1/fr

Links

Classifications

    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
    • G05D1/02—Control of position or course in two dimensions
    • G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
    • G05D1/0287—Control of position or course in two dimensions specially adapted to land vehicles involving a plurality of land vehicles, e.g. fleet or convoy travelling
    • G05D1/0289—Control of position or course in two dimensions specially adapted to land vehicles involving a plurality of land vehicles, e.g. fleet or convoy travelling with means for avoiding collisions between vehicles

Definitions

  • the present invention predicts a proximity time until an obstacle for a moving object such as an unmanned dump truck or a mobile robot is closest to the moving object, and a proximity distance between the obstacle and the closest obstacle.
  • the present invention relates to an obstacle collision prevention device for a moving body that avoids the moving body from colliding with an obstacle based on a prediction result.
  • a laser radar that scans a laser range finder to detect obstacles
  • a three-dimensional stereo camera There is one that detects an obstacle using a visual sensor.
  • a scan image and a captured image are acquired at a fixed time (sampling time), and a plurality of obstacles are determined from each image.
  • sampling time a fixed time
  • a plurality of obstacles are determined from each image.
  • it is no problem if it is one obstacle when trying to detect multiple obstacles from images acquired at different times, for example, it is possible to detect a moving object from an image at a certain time.
  • the relative position with respect to multiple obstacles is known, in the image acquired at the next time with a large time interval, the relative position has changed considerably from the previous position, It was difficult to respond to each obstacle.
  • the conventional obstacle detection device cannot accurately calculate the relative speed between the moving object and each obstacle, so that the time until the obstacle collides with the moving object can be calculated from the relative speed. Predicting and avoiding collisions would not be possible with high accuracy.
  • the collision between the moving object and each obstacle can be predicted by the conventional obstacle detection device because the number of obstacles is small, the relative speed between the moving object and the obstacle is small, and when the moving object moves. It was limited to the case where the shaking was small. Disclosure of the invention
  • the present invention has been made in view of such a situation, and is effective even when the number of obstacles is large, when the relative speed between the moving body and each obstacle is large, and when the moving body shakes greatly during traveling. It is an object of the present invention to accurately calculate the response of each sequentially moving obstacle and thereby accurately calculate the relative speed of each obstacle, so that the collision avoidance operation can be performed accurately.
  • a proximity time until an obstacle for a moving object is closest to the moving object a proximity distance to the obstacle when the object is closest to the moving object
  • a collision preventing device for a moving body which is configured to prevent the moving body from colliding with the obstacle based on the prediction result
  • the position measuring means By sequentially measuring the relative position of each obstacle existing outside the moving object at intervals of a fixed time or less by the position measuring means, the corresponding relative position data is associated with which obstacle data.
  • Speed calculating means for sequentially calculating the relative speed of each obstacle based on the relationship between each obstacle and the sequential relative position data associated by the associating means; Obtained obstacles and sequential relative position data Based on the relationship between each obstacle and the relative speed data obtained from the calculation result of the speed calculating means, and sequentially calculates the proximity time and the proximity distance for each obstacle.
  • Time and distance calculation means for sequentially calculating the relative speed of each obstacle based on the relationship between each obstacle and the sequential relative position data associated by the associating means; Obtained obstacles and sequential relative position data Based on the relationship between each obstacle and the relative speed data obtained from the calculation result of the speed calculating means, and sequentially calculates the proximity time and the proximity distance for each obstacle.
  • the relative positions P 1-2, P l-1, P 10... and P 2 of the obstacles 3 1 and 3 2 existing outside the moving body 1 -2, P2-1, P20 ... are measured by the position measuring means at intervals of less than a certain time (t-2, t-1 and t0-), as shown in Table 1 in Fig. 4.
  • the sequential relative position data is sequentially associated with which obstacle data.
  • the relative speed of each obstacle is sequentially calculated based on the relationship between each of the associated obstacles and the sequential relative position data.
  • the proximity time T until approaching the mobile object most and the proximity distance L when approaching the mobile object are sequentially calculated. Based on the calculation result, an operation and processing for preventing the moving body 1 from colliding with the obstacles 31 and 32 are performed.
  • the position measurement means since the position measurement means performs position measurement sequentially at intervals of a fixed time or less, it is possible to accurately associate the position measurement data with those of obstacles that have been displaced. be able to.
  • the associating means includes:
  • a three-dimensional image generating means for sequentially generating current three-dimensional images of each obstacle outside the moving body in a three-dimensional coordinate system viewed from the moving body at intervals of a fixed time or less, and a sequential generation by the three-dimensional image generating means.
  • By sequentially detecting the coordinate position of each obstacle in the three-dimensional image to be obtained it is possible to determine which obstacle data corresponds to the sequential coordinate position data. It consists of means for performing the application sequentially. Further, image information representing the characteristics of each obstacle in the generated three-dimensional image is extracted, and based on the extracted image information, the coordinate position data that is sequentially measured is used as the data of any obstacle. Is determined.
  • FIG. 1 is a perspective view showing an entire configuration of an embodiment of a moving object obstacle collision prevention device according to the present invention.
  • FIG. 2 is a block diagram showing a configuration of the collision prediction device according to the embodiment.
  • FIG. 3 is a diagram used to explain the correspondence between the relative position data and each obstacle.
  • FIG. 4 is a table showing the correspondence shown in FIG.
  • FIG. 5 is a diagram for explaining how a table of a correspondence relationship between each obstacle and relative speed data is obtained from the table shown in FIG.
  • FIG. 6 (a) is a diagram for explaining a method of obtaining a relative speed by interpolation
  • FIG. 6 (b) is a diagram for explaining a method of obtaining a proximity point by interpolation.
  • FIG. 7 is a diagram for explaining a first method for calculating the proximity time and the proximity distance.
  • FIG. 8 is a diagram illustrating a second method for calculating the proximity time and the proximity distance.
  • FIG. 9 is a diagram illustrating a third method for calculating the proximity time and the proximity distance.
  • FIG. 10 is a table for determining the risk of collision.
  • FIG. 11 is a diagram showing a distance image.
  • FIG. 12 is a diagram showing a three-dimensional distribution image of pixels obtained from a distance image.
  • Figure 13 shows the road surface and obstacles identified from the three-dimensional distribution image of pixels. It is a figure showing a two-dimensional image.
  • FIG. 1 when a moving body 1 such as an unmanned dump truck or a mobile robot travels along a traveling path 33, obstacles 3 such as rocks existing on the traveling path 3 3 1. It is assumed that each obstacle such as an obstacle 3 such as another vehicle 3 2 is detected by a collision prediction device 2 mounted on a moving body and a collision risk is determined. Although two obstacles are assumed for convenience of explanation, the present invention can naturally be applied to a case where there are three or more obstacles.
  • X—Y—Z indicates a vehicle body coordinate system that moves together with the moving body 1.
  • X is a coordinate axis corresponding to the vehicle width direction of the moving body 1
  • Z is a coordinate axis corresponding to the traveling direction of the moving body 1
  • Y is a vertical coordinate axis.
  • the origin of the vehicle body coordinate system corresponds to the origin of the camera of the obstacle detection unit 2 of the collision prediction device 2.
  • FIG. 2 shows the configuration of the collision prediction device 2 according to the embodiment of the present invention.
  • the collision prediction device 2 has a relative position P1-2, P1-1, P1-1 of each obstacle 31 1 and 32 2 existing outside the moving object 1 with respect to the moving object 1.
  • 10... and P2-2, P2-1, P20... are measured at high speed successively at intervals (t-2, t-1, t0- Obstacle detector 4 that sequentially associates the relative position data with which obstacle data, and the relationship between each of the associated obstacles 31 and 32 and the sequential relative position data
  • the relative speed calculation unit 5 that sequentially calculates the relative speed of each of the obstacles 3 1 and 3 2 with respect to the moving object 1 based on the following, and the relative position of each of the associated obstacles 3 1 and 3 2 sequentially Based on the relationship with the data and the relationship between the obstacles 3 1 and 3 2 and the sequential relative velocity data, each of the obstacles 3 1 and 3 2 comes closest to the moving object 1
  • the driving of the moving body 1 is controlled in accordance with the collision risk output from the determination section 7 so that the moving body 1 does not collide with an obstacle. In this way, the collision of the moving body 1 with the obstacles 31 and 32 is avoided.
  • the present inventors convert the distance image in the traveling direction of the moving object into a three-dimensional distribution of each pixel, specify a pixel group corresponding to the traveling path from the three-dimensional distribution state of the pixels, and
  • the present invention proposes an invention in which a road is regarded as a plane, and an object having a height higher than a predetermined height is detected as an obstacle based on the height of the plane, and attempts have been made to implement the present invention.
  • the obstacle detection unit 4 of the collision prediction device 2 can be configured by applying the obstacle detection method according to the proposal of the present inventors.
  • the obstacle detection unit 4 uses a multi-lens lens (multi-lens camera), for example, a traveling road surface 33 as shown in FIG. 11 and obstacles 31, 3 2 existing on the traveling road surface 33. Is generated.
  • Each pixel 50 of the distance image 30 has three-dimensional data indicating the two-dimensional coordinate position (i, j) in the i-j two-dimensional coordinate system and the distance d from the reference position (reference plane) of the moving body 1. (I, j, d) are associated with each other, and the pixel at each position i, j of the distance image 30 has brightness corresponding to the distance d.
  • a multi-lens multi-eye lens
  • the method used can be used.
  • the distance image 30 is obtained in this way, the distance image 30 is subjected to coordinate transformation to obtain a three-dimensional image 40 in the vehicle body coordinate system X—Y—Z as shown in FIG. Is generated.
  • each pixel 50 of the distance image 30 is associated with the three-dimensional information (i, j, d) as described above, this distance image data (i, j, d)
  • This distance image data (i, j, d) Each pixel 50 shown is moved together with the moving body 1 as shown in FIG. 12, and the three-dimensional coordinates on the body coordinate system X—Y—Z, which is a camera coordinate system with the origin of the camera of the moving body 1 as an origin. It can be converted to each pixel 60 corresponding to the position data (X, Y, ⁇ ). By performing such conversion, a three-dimensional distribution image 40 can be obtained as a distribution diagram of the three-dimensional coordinate position of the pixel 60.
  • a pixel group at the lowest point in the vertical direction is selected from all the pixels 60 of the three-dimensional distribution image 40, and these are approximated in a plane to correspond to the road surface 33.
  • a plane (X- ⁇ plane) can be detected.
  • the obstacles 3 1 and 3 2 are a group of objects, they should be observed as a group of pixels in the two-dimensional distribution of X- ⁇ , and should not be a group of pixels ( Objects other than obstacles).
  • this two-dimensional distribution is divided at predetermined intervals in the width X direction, and for each of the regions 71, 7 2. Search for the existence. And the area where the obstacle candidate point exists Is an obstacle candidate area. Then, it is searched whether or not the obstacle candidate area is continuous in the width direction X. In this way, the entirety of the obstacle candidate area that is continuous in the width direction X is set as the obstacle area.
  • two obstacle regions, one obstacle region corresponding to the obstacle 31 and another obstacle region corresponding to the obstacle 32, should be searched.
  • the three-dimensional coordinate position P1 of the pixel corresponding to the position (for example, the position of the center of gravity) representing the obstacle 31 is obtained.
  • the maximum length in the horizontal direction of the obstacle 31 is obtained as “width a l”.
  • the maximum length of the obstacle 31 in the direction perpendicular to the width a 1 is obtained as “depth b lj.
  • the maximum height of the obstacle 31 is obtained as“ height clj.
  • the color dl characterizing the obstacle 3 1 (obtained by averaging the brightness of the pixels 60 composing the obstacle 3 1) is obtained.
  • the same processing is performed for the other obstacle 3 2.
  • the representative position P2, width a2, depth b2, height c2, and color d2 are required.The representative positions of the obstacles 31 and 32? Good.
  • the traveling road surface 33 is detected as a group of pixels whose planes are approximated, and obstacles 31 and 32 are detected as a group of pixels having a continuous distribution.
  • 3D image 40 ' is acquired.
  • the above-described obstacle detection processing can be performed at high speed in real time, and the positions 1, P2 of the obstacles 31 and 32, that is, the moving body 1 (the camera of the moving body 1)
  • the relative positions Pl and P2 of the obstacles 31 and 32 with respect to can be measured at extremely short sampling intervals.
  • the relative positions P1-2, Pl-1, P10 ... and P2-2, P2-1, P20 ... of the obstacles 31 and 32 are extremely short sampling times (t-2 , Tl, tO ').
  • the position data closest to the position at the previous measurement time is sequentially determined to be the same obstacle position data.
  • the relative position data at each time is associated with each of the obstacles 31 and 3.
  • the position measurement is performed at a very short interval.
  • the above-mentioned association can be performed accurately and reliably even if the three-dimensional image 40 'is influenced by the vibration.
  • feature information such as width al, depth bl, height cl, and color dl is obtained along with the relative position data P1, and width a2, depth b2, along with the relative position data P2. Since feature information such as height c2 and color d2 has been acquired, it is possible to determine that relative position data having the same content of feature information is data of the same obstacle. It is possible to more accurately and more reliably associate one relative position data with one of the obstacles.
  • the characteristic information may be any one of width a, depth b, height c, color d, or Any combination of two or more can be used. It is also possible to perform the association without using the feature information.
  • the present invention is not limited to the above-described method of converting a distance image into a three-dimensional image.
  • a current 3D image of each obstacle outside the moving object in the 3D coordinate system X—Y—Z viewed from the moving object 1 can be generated sequentially at intervals of a fixed time or less, a 3D image
  • the method of generating is arbitrary.
  • the present invention is not necessarily limited to one that generates a three-dimensional image.
  • the relative speed calculation unit 5 calculates the relative speed of each of the obstacles 3 1 and 3 2 based on the correspondence between the obstacles 3 1 and 3 2 stored in Table 1 of FIG. 4 and the sequential relative position data. Are sequentially calculated, and the calculation result is stored and stored as Table 2 as shown in FIG.
  • any method can be applied, and for example, the following method can be used.
  • the difference between the components at the relative position between the last time t0 and the time t-1 immediately before the last time is calculated.
  • the relative speed V IO (VxlO, VylO, VzlO) of the obstacle 31 at the final time t0 is obtained by the following calculation.
  • VylO Y 10— Y 1-1
  • V zl0 Z 10— Z 1
  • the prediction unit 6 calculates the correspondence between the obstacles 3 1 and 3 2 stored in Table 1 and the sequential relative position data, and the sequential relative positions of the obstacles 3 1 and 3 2 stored in Table 2 Based on the correspondence with the speed data, the approach time T until the obstacle approaches the moving object 1 from the current time t 0, which is the last time, is calculated for each of the obstacles 3 1 and 3 2, and Then, the proximity distance L between the moving body 1 and the obstacle at the time of the closest approach is calculated.
  • any method can be applied. For example, the following methods can be used.
  • the predicted movement trajectory is defined as a straight line (the directional component of the relative speed vector at the last time). It is a method of calculating from the geometric relationship between this straight line and the body coordinate system X-Y-Z.
  • the obstacle 31 moves from the relative position P at the final time t O to the relative speed V at the final time t 0.
  • a straight line extending from the relative position P (X10, Y10, Z10) at the final time in the direction component direction of the relative velocity vector V (Vxl0, VylO, VzlO) is set as the predicted movement trajectory M.
  • the distance between the relative position P (X10, Y10, Z10) at the final time and the predicted relative position Q (XQ, YQ, ZQ) when approaching closest is calculated, and this distance is further moved at the relative speed V.
  • the time is calculated, and this is referred to as the proximity time T.
  • This method is a method of performing calculation processing in the above calculation method 1 while ignoring the component in the vertical Y axis.
  • This method can be applied when the running path is a flat road that is not a slope, and obstacles fly in the air or do not run on a slope, and the calculation processing is simplified by simplifying the calculation. Time can be reduced.
  • This straight line M has a slope (VzlO / VxlO) corresponding to the direction component of the relative velocity V (VxlO, VzlO), and the point P (X10, Z10), when the obstacle 31 comes closest to the moving body 1, Since it passes through the relative point Q point (XQ, ZQ), the following equation (1) is obtained.
  • the proximity time T is a first time
  • This method is a method of further shortening the calculation processing time by performing the calculation in the calculation method 2 more easily.
  • This method accurately predicts the predicted movement trajectory of an obstacle in consideration of acceleration.
  • the storage unit 8 determines the degree of danger of the obstacle colliding with the moving object 1 according to the combination of the magnitude of the proximity time ⁇ and the magnitude of the proximity distance L. Table 3 is stored and stored.
  • the contents of the table 3 stored in the storage unit 8 are compared with the proximity time and the proximity distance calculated for each of the obstacles 31 and 32 for each obstacle. This determines the risk of collision.
  • the proximity time ⁇ of the obstacle 3 1 is in the range between the threshold value Thl and the threshold value Th2, and the proximity distance L of the obstacle 3 1 is in the range between the threshold value Lhl and the threshold value Lh2. If it is within the range, it is judged as "low risk”.
  • the judgment results of the danger levels of the obstacles 31 and 32 are output from the judgment unit 7.
  • an arithmetic expression for calculating the risk of collision as a numerical value is stored in the storage unit 8, and the risk of collision can be obtained by substituting the proximity time T and the proximity distance L into this expression. Good.
  • the collision risk D can be calculated from the calculation formula shown in the following expression (8).
  • the danger level D of each of the obstacles 3 1 and 3 2 is output from the determination unit 7.
  • the controller 3 controls the force to apply the sudden braking and the relaxation. It determines the force to apply the brake or whether it should be avoided by tightening the steering wheel, etc., and drives and controls the moving body 1 so as to avoid the obstacle.
  • the judgment unit 7 judges which of the obstacles 3 1 and 3 2 has the highest risk. It is also possible to output a fixed result or numerical value D to judge and control.

Landscapes

  • Engineering & Computer Science (AREA)
  • Radar, Positioning & Navigation (AREA)
  • Remote Sensing (AREA)
  • Aviation & Aerospace Engineering (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Automation & Control Theory (AREA)
  • Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)
  • Length Measuring Devices By Optical Means (AREA)
  • Measurement Of Optical Distance (AREA)
  • Processing Or Creating Images (AREA)
  • Traffic Control Systems (AREA)

Abstract

Même lorsque le nombre d'obstacles est élevé, même lorsque les vitesses relatives d'un objet en mouvement se rapprochant des obstacles sont élevées, et même lorsque l'objet en mouvement est animé de grandes secousses, la présente invention permet d'obtenir de façon séquentielle et avec précision les correspondances vers les obstacles mobiles et de calculer exactement les vitesses relatives vers les obstacles en fonction de ces correspondances afin d'éviter la collision de manière fiable. On mesure de façon séquentielle les positions relatives des obstacles se dressant autour de l'objet en mouvement à des intervalles de temps inférieurs à une durée prédéterminée au moyen d'un dispositif de mesure de position afin de mettre en corrélation de façon séquentielle les données successives de position relative et les données relatives aux obstacles. On calcule ensuite de façon séquentielle les vitesses relatives vers les obstacles en fonction de la relation entre les obstacles et les données successives de position relative. En outre, on calcule successivement le temps d'approche de chaque obstacle, c'est-à-dire le temps qui est nécessaire pour que l'obstacle se rapproche le plus de l'objet en mouvement, et la distance rapprochée de chaque obstacle, c'est-à-dire la distance qui sépare l'obstacle de l'objet en mouvement lorsque l'obstacle se trouve le plus proche de l'objet en mouvement, en fonction de la relation entre les obstacles et les données successives de position relative et en fonction de la relation entre les obstacles et les données successives de vitesse relative. Selon le résultat de ces calculs, on effectue les opérations nécessaires afin de prévenir la collision de l'objet en mouvement avec l'un de ces obstacles.
PCT/JP1997/004044 1996-11-07 1997-11-06 Appareil permettant de prevenir la collision d'un objet en mouvement avec un obstacle WO1998020398A1 (fr)

Priority Applications (1)

Application Number Priority Date Filing Date Title
AU48847/97A AU4884797A (en) 1996-11-07 1997-11-06 Obstacle collision preventive apparatus of mobile object

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
JP8/295176 1996-11-07
JP8295176A JPH10143245A (ja) 1996-11-07 1996-11-07 移動体の障害物衝突防止装置

Publications (1)

Publication Number Publication Date
WO1998020398A1 true WO1998020398A1 (fr) 1998-05-14

Family

ID=17817218

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP1997/004044 WO1998020398A1 (fr) 1996-11-07 1997-11-06 Appareil permettant de prevenir la collision d'un objet en mouvement avec un obstacle

Country Status (3)

Country Link
JP (1) JPH10143245A (fr)
AU (1) AU4884797A (fr)
WO (1) WO1998020398A1 (fr)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20180345922A1 (en) * 2017-06-01 2018-12-06 Delphi Technologies, Inc. Automatic braking system for slow moving objects
CN111381594A (zh) * 2020-03-09 2020-07-07 兰剑智能科技股份有限公司 基于3d视觉的agv空间避障方法及系统

Families Citing this family (22)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP3986681B2 (ja) * 1998-08-19 2007-10-03 本田技研工業株式会社 車両の走行安全装置
JP3975009B2 (ja) * 1998-08-25 2007-09-12 本田技研工業株式会社 車両の走行安全装置
JP2004145479A (ja) * 2002-10-22 2004-05-20 Aisin Seiki Co Ltd 周辺車両情報提供装置
JP4089443B2 (ja) * 2003-01-21 2008-05-28 トヨタ自動車株式会社 衝突判定装置
JP4080932B2 (ja) 2003-03-31 2008-04-23 本田技研工業株式会社 ２足歩行ロボットの制御装置
JP3949628B2 (ja) * 2003-09-02 2007-07-25 本田技研工業株式会社 車両の周辺監視装置
JP4884806B2 (ja) * 2006-03-14 2012-02-29 本田技研工業株式会社 車両周辺監視システム
DE112007000468T5 (de) * 2006-03-27 2009-01-15 Murata Mfg. Co., Ltd., Nagaokakyo-shi Radarvorrichtung und mobiles Objekt
JP5112666B2 (ja) * 2006-09-11 2013-01-09 株式会社日立製作所 移動装置
JP4171510B2 (ja) * 2006-12-20 2008-10-22 本田技研工業株式会社 移動装置、ならびにその制御システム、制御プログラムおよび監督システム
EP1978432B1 (fr) 2007-04-06 2012-03-21 Honda Motor Co., Ltd. Appareil de routage pour unité mobile autonome
JP5080333B2 (ja) * 2007-04-06 2012-11-21 本田技研工業株式会社 自律移動体のための物体認識装置
JP4774401B2 (ja) * 2007-04-06 2011-09-14 本田技研工業株式会社 自律移動体の経路設定装置
JP2009151648A (ja) * 2007-12-21 2009-07-09 Mitsubishi Fuso Truck & Bus Corp 車両用警報装置
JP5018980B2 (ja) * 2010-04-08 2012-09-05 カシオ計算機株式会社 撮像装置、長さ測定方法、及びプログラム
CN102685382B (zh) 2011-03-18 2016-01-20 安尼株式会社 图像处理装置和方法、及移动体防碰撞装置
JP2016001464A (ja) * 2014-05-19 2016-01-07 株式会社リコー 処理装置、処理システム、処理プログラム、及び、処理方法
JP6723774B2 (ja) * 2016-03-15 2020-07-15 キヤノン株式会社 撮像素子、撮像装置、測距装置及び移動体
KR102355426B1 (ko) * 2017-12-08 2022-01-26 현대자동차주식회사 주행 경로 상의 장애물 탐색 및 회피를 위한 방법 및 장치
JP6728268B2 (ja) * 2018-04-26 2020-07-22 キヤノン株式会社 撮像装置、撮像システム、および、移動体
JP7258432B2 (ja) * 2019-05-30 2023-04-17 アルパイン株式会社 路面検出システム、パーソナルモビリティ及び路面検出方法
JP7179687B2 (ja) * 2019-06-17 2022-11-29 株式会社東芝 障害物検知装置

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPS6085897A (ja) * 1983-10-19 1985-05-15 株式会社日立製作所 自律移動ロボツトにおける視覚情報から三次元画像を得る方法
JPH04358209A (ja) * 1991-06-04 1992-12-11 Yamatake Honeywell Co Ltd 障害物回避装置
JPH07129898A (ja) * 1993-11-05 1995-05-19 Toyota Central Res & Dev Lab Inc 障害物検出装置

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPS6085897A (ja) * 1983-10-19 1985-05-15 株式会社日立製作所 自律移動ロボツトにおける視覚情報から三次元画像を得る方法
JPH04358209A (ja) * 1991-06-04 1992-12-11 Yamatake Honeywell Co Ltd 障害物回避装置
JPH07129898A (ja) * 1993-11-05 1995-05-19 Toyota Central Res & Dev Lab Inc 障害物検出装置

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20180345922A1 (en) * 2017-06-01 2018-12-06 Delphi Technologies, Inc. Automatic braking system for slow moving objects
US10913434B2 (en) * 2017-06-01 2021-02-09 Aptiv Technologies Limited Automatic braking system for slow moving objects
CN111381594A (zh) * 2020-03-09 2020-07-07 兰剑智能科技股份有限公司 基于3d视觉的agv空间避障方法及系统

Also Published As

Publication number Publication date
JPH10143245A (ja) 1998-05-29
AU4884797A (en) 1998-05-29

Similar Documents

Publication Publication Date Title
WO1998020398A1 (fr) Appareil permettant de prevenir la collision d'un objet en mouvement avec un obstacle
CN105425803B (zh) 自主避障方法、装置和系统
US11167753B2 (en) Vehicle control device, vehicle control method, and vehicle control program
KR101207903B1 (ko) 자율 이동 차량용 장애물 정보 제공장치 및 그 방법
JP5080333B2 (ja) 自律移動体のための物体認識装置
CN108536149A (zh) 一种基于Dubins路径的无人驾驶车辆避障控制装置及控制方法
US8019475B2 (en) Routing apparatus for autonomous mobile unit
JP4510554B2 (ja) 立体物監視装置
JP5982298B2 (ja) 障害物検出装置および障害物検出方法
KR101049906B1 (ko) 자율 이동 장치 및 이의 충돌 회피 방법
JP6959056B2 (ja) 移動ロボットの制御装置と制御方法
JP5402813B2 (ja) 車両運動制御装置及びプログラム
US20210387616A1 (en) In-vehicle sensor system
CN111942378A (zh) 车辆控制装置、车辆控制方法及存储介质
US20030097237A1 (en) Monitor system of vehicle outside and the method thereof
JP6985089B2 (ja) 立体物接地判定装置
KR102178992B1 (ko) 차량 동작 계획방법 및 그를 위한 장치
US20150262375A1 (en) Speed calculating device and speed calculating method, and collision determination device
Lindner et al. 3D LIDAR processing for vehicle safety and environment recognition
JP4956099B2 (ja) 壁検出装置
TWI680898B (zh) 近距離障礙物之光達偵測裝置及其方法
JP2009157668A (ja) 外界認識装置
JP5834428B2 (ja) 車両運動制御装置及びプログラム
CN109195849B (zh) 摄像装置
KR20230031344A (ko) 자동차를 둘러싼 영역 내 장애물을 검출하기 위한 시스템 및 방법

Legal Events

Date Code Title Description
AK Designated states

Kind code of ref document: A1

Designated state(s): AU US

DFPE Request for preliminary examination filed prior to expiration of 19th month from priority date (pct application filed before 20040101)