WO2017078001A1 - 障害物検知装置 - Google Patents

障害物検知装置 Download PDF

Info

Publication number
WO2017078001A1
WO2017078001A1 PCT/JP2016/082400 JP2016082400W WO2017078001A1 WO 2017078001 A1 WO2017078001 A1 WO 2017078001A1 JP 2016082400 W JP2016082400 W JP 2016082400W WO 2017078001 A1 WO2017078001 A1 WO 2017078001A1
Authority
WO
WIPO (PCT)
Prior art keywords
obstacle
parking space
unit
camera
area
Prior art date
Legal status (The legal status 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 status listed.)
Ceased
Application number
PCT/JP2016/082400
Other languages
English (en)
French (fr)
Japanese (ja)
Inventor
秀行 粂
將裕 清原
待井　君吉
範安 長谷島
吉孝 内田
健人 緒方
雄太 大泉
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Faurecia Clarion Electronics Co Ltd
Original Assignee
Clarion Co 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 Clarion Co Ltd filed Critical Clarion Co Ltd
Priority to EP16862061.5A priority Critical patent/EP3372456A4/en
Priority to US15/767,884 priority patent/US10242576B2/en
Publication of WO2017078001A1 publication Critical patent/WO2017078001A1/ja
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/16—Anti-collision systems
    • G08G1/168—Driving aids for parking, e.g. acoustic or visual feedback on parking space
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/50—Context or environment of the image
    • G06V20/56—Context or environment of the image exterior to a vehicle by using sensors mounted on the vehicle
    • G06V20/58—Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads
    • G06V20/586—Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads of parking space
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60R—VEHICLES, VEHICLE FITTINGS, OR VEHICLE PARTS, NOT OTHERWISE PROVIDED FOR
    • B60R21/00—Arrangements or fittings on vehicles for protecting or preventing injuries to occupants or pedestrians in case of accidents or other traffic risks
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B62—LAND VEHICLES FOR TRAVELLING OTHERWISE THAN ON RAILS
    • B62D—MOTOR VEHICLES; TRAILERS
    • B62D15/00—Steering not otherwise provided for
    • B62D15/02—Steering position indicators ; Steering position determination; Steering aids
    • B62D15/027—Parking aids, e.g. instruction means
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/50—Context or environment of the image
    • G06V20/56—Context or environment of the image exterior to a vehicle by using sensors mounted on the vehicle
    • G06V20/58—Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads
    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/14—Traffic control systems for road vehicles indicating individual free spaces in parking areas
    • G08G1/141—Traffic control systems for road vehicles indicating individual free spaces in parking areas with means giving the indication of available parking spaces
    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/16—Anti-collision systems
    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/16—Anti-collision systems
    • G08G1/166—Anti-collision systems for active traffic, e.g. moving vehicles, pedestrians, bikes

Definitions

  • the present invention relates to an obstacle detection device.
  • Patent Document 1 a mark having a predetermined positional relationship with respect to the parking section is installed, and it is determined whether or not the vehicle is stopped at the target initial stop position based on the image of the photographed mark.
  • a parking assist device that detects the presence or absence of an obstacle when it is determined that the vehicle is stopped at a position is disclosed.
  • the obstacle detection device includes a first camera that is mounted on a moving vehicle and images the front or rear of the vehicle, and a second camera that is mounted on the vehicle and images a side of the vehicle.
  • a parking space detection unit that detects a parking space of the vehicle based on an image obtained by photographing the first camera, a storage unit that stores an area in the real space of the parking space detected by the parking space detection unit,
  • An exercise acquisition unit that acquires information on the behavior of the vehicle, an obstacle calculation unit that calculates information on obstacles around the parking space based on a plurality of images obtained by the second camera taken at different times, and
  • a processing area setting unit that controls calculation in the obstacle calculation unit based on information on the real space of the parking space stored in the storage unit and information on the behavior of the vehicle acquired by the exercise acquisition unit.
  • the processing load due to the calculation of obstacle information can be reduced.
  • the figure which shows the structure of the obstruction detection apparatus 100 The figure which shows an example of the parking space information area 12a
  • the figure which represented the function which the program executed in obstacle detection device 100 has as a functional block
  • the flowchart which shows the process performed in the process area setting part 105 The figure which shows an example of the distance d of a parking space and the own vehicle
  • the flowchart which shows the process performed in the side obstacle detection part 106 The figure which represented as a functional block the function which the program performed in the obstacle detection apparatus 100a in 2nd Embodiment has.
  • FIG. 1 is a diagram illustrating a configuration of an obstacle detection device 100 built in a vehicle 500.
  • the vehicle 500 includes a CAN bus 20, and the obstacle detection device 100 is connected to the CAN bus 20.
  • Other devices not shown are also connected to the CAN bus 20.
  • a device that outputs the traveling direction of the vehicle to the CAN bus 20 a device that outputs the vehicle speed to the CAN bus 20, and a device that controls the vehicle 500 based on the obstacle information output by the obstacle detection device 100 are connected. Is done.
  • the obstacle detection apparatus 100 includes a front camera 101, a side camera 102, a CPU 10, a ROM 11, a RAM 12, and a CAN interface 13.
  • the front camera 101 is attached to the upper front of the vehicle 500 and photographs the front of the vehicle 500.
  • Side camera 102 is attached to the left side surface of vehicle 500 and photographs the left side of vehicle 500.
  • the CPU 10 calculates obstacle information from images obtained by the front camera 101 and the side camera 102 using a program to be described later at predetermined intervals, for example, every 0.1 second.
  • the calculated obstacle information is output to the CAN bus 20 via the CAN interface 13.
  • the above-described predetermined cycle is referred to as a “processing cycle”.
  • the ROM 11 stores a program and camera parameters of the front camera 101 and the side camera 102.
  • the program is developed from the ROM 11 to the RAM 12 by the CPU 10 and executed.
  • the camera parameters are internal parameters such as lens distortion and external parameters such as a camera mounting position / angle with respect to the vehicle.
  • the RAM 12 temporarily stores the parking space information area 12a or other information necessary for executing the program.
  • the parking space information area 12a will be described later.
  • the CAN interface 13 is a communication interface with the CAN bus 20 of the obstacle detection device 100.
  • the obstacle detection device 100 acquires movement information of the vehicle 500, that is, information related to the traveling direction and speed of the vehicle 500 via the CAN interface 13.
  • the obstacle detection device 100 outputs the calculated obstacle information to the CAN bus 20 via the CAN interface 13.
  • the parking space information area 12a is a predetermined area of the RAM in which information related to a parking space detected by a parking space detection unit 103 described later is stored.
  • the information regarding the parking space is the position and posture of the parking space in the real space with respect to the host vehicle.
  • the position and orientation of the parking space to be output are represented as a combination of a plurality of vertex coordinates that constitute the parking space, for example. Or you may represent as a combination of the center coordinate of a parking space, the magnitude
  • FIG. 2 is a diagram illustrating an example of information related to a parking space stored in the parking space information area 12a. In the example shown in FIG.
  • the parking space information area 12a stores three records, that is, information on three parking spaces.
  • the parking space is expressed as a combination of a plurality of vertex coordinates constituting the parking space. These coordinates represent a coordinate system around the host vehicle, that is, a relative position.
  • FIG. 3 is a diagram illustrating the functions of a program executed in the CPU 10 of the obstacle detection apparatus 100 as function blocks. That is, the obstacle detection device 100 is configured to execute a parking space detection function by the parking space detection unit 103, a vehicle motion acquisition function by the vehicle motion acquisition unit 104, and a processing region setting by the processing region setting unit 105 according to a program executed by the CPU 10. A function, a side obstacle detection function by the side obstacle detection unit 106, and an output function by the output unit 107.
  • the obstacle detection device 100 causes the front camera 101 and the side camera 102 to take images at a predetermined cycle, that is, every processing cycle, and performs processing by each functional block when an image is obtained by shooting.
  • the parking space detection unit 103 starts processing, and when the parking space detection unit 103 completes processing, the processing area setting unit 105 starts processing.
  • the side obstacle detection unit 106 starts the process. That is, each functional block operates every processing cycle.
  • the parking space detection unit 103 detects a parking space existing in front of the host vehicle from an image obtained by photographing the front camera 101, and adds information on the detected parking space to the parking space information area 12a.
  • the detection of a parking space is, for example, detecting a white line from an image and detecting a region between two white lines as a parking space.
  • an obstacle is detected from the image, and an area where no obstacle exists is used as a parking space.
  • the same processing as that of the side obstacle detection unit 106 described later can be used.
  • the vehicle motion acquisition unit 104 acquires information regarding the motion of the vehicle 500, that is, information regarding the traveling direction and speed of the vehicle 500 via the CAN interface 13. Then, the movement amount of the vehicle 500 from the previous processing cycle to the current processing cycle is calculated and output to the processing region setting unit 105.
  • the movement information of the vehicle 500 is acquired via the CAN bus 20 as information output from the wheel encoder that acquires the rotation amount of the wheel attached to the vehicle 500 and the steering angle. By executing the calculation according to the geometric model of the vehicle based on these pieces of information, the amount of movement of the vehicle, for example, 2 m to the right and 0.5 m to the front is calculated.
  • an output value of a sensor capable of acquiring posture / position information such as an acceleration / angular accelerometer and GPS mounted on the vehicle 500 is acquired via the CAN bus 20 to calculate the movement amount of the vehicle. Also good.
  • the processing area setting unit 105 sets a processing area C, which is an area on the image, to be processed by the side obstacle detection unit 106, which is a subsequent process.
  • the processing area C is set based on the information output from the parking space detection unit 103 and stored in the parking space information area 12a of the RAM and the output of the vehicle motion acquisition unit 104, that is, vehicle motion information. As will be described later, when the processing area C is blank, that is, when the processing area C does not exist, the side obstacle detection unit 106 does not perform processing.
  • the side obstacle detection unit 106 performs the following processing on a plurality of images obtained by photographing the side camera 102 at different times. That is, in the image obtained in the current processing cycle, the processing region C set as described above is specified, and the image information including luminance included in the processing region C and obtained in the past processing cycle. An obstacle present on the side of the vehicle is detected using image information including luminance included in the image, and a relative position with respect to the obstacle is output. When the processing area C is set to be blank by the processing area setting unit 105, the side obstacle detection unit 106 does not perform processing. Details of the processing will be described later.
  • the output unit 107 outputs the obstacle detection result to the CAN bus 20 via the CAN interface 13.
  • FIG. 4 is a flowchart showing processing executed in the processing area setting unit 105.
  • the execution subject of each step described below is the CPU 10.
  • step S210 processing area C is cleared and processing area C is left blank.
  • step S216 the side obstacle detection unit 106 does not calculate the three-dimensional position unless the processing region C is set in step S216 described later.
  • step S211 the process proceeds to step S211.
  • step S211 the information stored in the parking space information area 12a of the RAM 12 is updated using the movement amount of the vehicle from the immediately preceding processing cycle output by the vehicle motion acquisition unit 104 to the current processing cycle.
  • the parking space detected by the parking space detection unit 103 in the current processing cycle uses the information regarding the parking space output by the parking space detection unit 103 as it is.
  • step S212 The processes after the next step S212 are repeated for each parking space stored in the parking space information area 12a.
  • step S212 is executed for the first time, the first record stored in the parking space information area 12a is processed.
  • step S212 the distance d between the parking space and the host vehicle is calculated from the relative position / posture of the parking space and the host vehicle, and the process proceeds to step S213.
  • FIG. 5 is a diagram illustrating an example of a distance d between the parking space and the host vehicle.
  • a parked vehicle 501 parked between the white lines 502 exists around the host vehicle.
  • the parking space detection unit 103 detects the parking space 503.
  • the position of the parking space 503 is defined as the position closest to the host vehicle 500 in the parking space, and the position of the host vehicle is defined as the position of the side camera 102.
  • a distance 504 between the position of the parking space 503 and the position of the host vehicle is calculated, and this is defined as a distance d.
  • the position of the left camera is used as the position of the host vehicle, but the position of the host vehicle is not limited to this, and may be the center of the host vehicle, for example.
  • the position closest to the host vehicle is used as the position of the parking space
  • the position of the parking space is not limited to this, and may be the center of the parking space, for example.
  • step S213 it is determined whether or not the distance d between the parking space and the host vehicle is equal to or less than a preset threshold value d_th. If it is determined that the distance d is less than or equal to the threshold value d_th, the process proceeds to step S214. If it is determined that the distance d is greater than the threshold value d_th, the process proceeds to step S217.
  • step S214 a prediction area P, which is an area where the parking space and the vicinity of the parking space are expected to be reflected in the image obtained by photographing with the side camera 102, is calculated.
  • the prediction area P can be obtained by projection calculation according to a perspective projection model or the like using the parking space, the relative position / posture of the host vehicle, and external / internal parameters of the side camera 102.
  • step S215 it is determined whether or not the entire prediction region P calculated in step S214 exists in the determination region A set in advance.
  • the determination area A is defined as follows. The peripheral portion of the image obtained by photographing with the side camera 102 has a large distortion. For this reason, the area excluding the peripheral portion of the image is set as the determination area A. If it is determined that the entire prediction region P exists within the determination region A, the process proceeds to step S216. If it is determined that a part of the prediction region P does not exist in the determination region A or that the prediction region P and the determination region A do not overlap at all, the process proceeds to step S217.
  • FIG. 6 is a diagram illustrating an example of the prediction region P and the determination region A.
  • a parked vehicle 511 and a white line 512 are shown in an image 510 obtained by photographing with the side camera 102.
  • a prediction area P which is an area where the parking space and the surroundings of the parking space are expected to be reflected, is calculated by projection calculation.
  • step S215 it is determined whether or not the prediction area P exists within the predetermined determination area A. In the example shown in FIG. 6, since the entire prediction area P exists in the determination area A, the process proceeds to step S216. Returning to FIG. 4, the description will be continued.
  • step S216 the prediction region P calculated in step S214 is set as the processing region C.
  • a region wider than the prediction region P may be set as the processing region C in consideration of the detection position of the parking space, the moving amount of the vehicle, and the calibration error.
  • step S217 it is determined whether all the records stored in the parking space information area 12a of the RAM 12 have been processed. If it is determined that all the records have been processed, the process represented by the flowchart of FIG. 4 is terminated. If it is determined that there are any unprocessed records, the process returns to step S212. For example, when a plurality of records are stored in the parking space information area 12a, the respective prediction areas P corresponding to these records may be calculated, and an area obtained by combining them may be set as the processing area C. .
  • FIG. 7 is a flowchart showing processing executed in the side obstacle detection unit 106.
  • the side obstacle detection unit 106 calculates the three-dimensional position of the side obstacle based on the image obtained by the side camera 102 and the processing area C set by the processing area setting unit 105.
  • the execution subject of each step described below is the CPU 10.
  • step S200 it is determined whether or not the processing area C exists.
  • the side obstacle detection unit 106 when the processing area C is cleared in step S210 and step S216 is not executed once, the processing area C does not exist. If it is determined that the processing area C exists, the process proceeds to step S201. If it is determined that the processing area C does not exist, the process represented by the flowchart shown in FIG.
  • step S201 a feature point is extracted from an area included in the processing area C, which is an image obtained by the side camera 102 in the current processing cycle.
  • the feature points extracted in this step are used in the next step.
  • Feature points are extracted, for example, by the Harris corner detection method (C. Harris and M. Stephens: “A) that extracts points (feature points) that have large differences from surrounding points and can be easily matched from the corners and edges. combined corner and edge detector, ”Proc. Alvey Vision Conf., pp. 147-151, 1988.).
  • Harris corner detection method C. Harris and M. Stephens: “A) that extracts points (feature points) that have large differences from surrounding points and can be easily matched from the corners and edges. combined corner and edge detector, ”Proc. Alvey Vision Conf., pp. 147-151, 1988.
  • step S202 a point corresponding to the feature point extracted in step S201 is searched from the image of the side camera 102 n cycles before.
  • An arbitrary value is set in advance in n. Or you may set the period required for the movement of the preset distance according to the movement amount of the vehicle obtained from the vehicle movement acquisition part 104.
  • FIG. When a corresponding point is searched from the image of the side camera 102 before n cycles, it is determined that the point is an obstacle.
  • an efficient search can be performed by searching around the position where the feature points are extracted.
  • a search is performed by setting a certain range centered on the position where the feature points are extracted as the search range.
  • a neighboring image pattern centered on the position of the feature point is used, and a pattern most similar to the image pattern is searched for within the search range.
  • SSD Sum of Squared Difference
  • SAD Sum of Absolute Difference
  • NCC Normalized Cross Correlation
  • the LK method (Bruce D. Lucas and Takeo Kanade. An Iterative) is used to search for a point where the SSD value becomes small in the vicinity of the initial value by using the feature point extraction position as an initial value. Image Registration Technique with an Application to Stereo Vision. Int. Joint Conf. On Artificial Intelligence, pp. 674-679, 1981.).
  • the LK method can obtain the association at high speed and high accuracy when the movement between images is small, and is suitable for the association between temporally continuous images taken by the moving body.
  • step S203 the three-dimensional position of the feature point associated in step S202 is calculated.
  • the three-dimensional position of the feature point includes the position of the feature point in the image in the current processing cycle and the processing cycle before n cycles, the internal parameters of the side camera 102, and the amount of movement of the side camera 102 during the n cycles. And can be calculated based on the principle of triangulation.
  • the amount of movement of the side camera 102 can be calculated from the amount of movement of the vehicle acquired by the vehicle motion acquisition unit 104 and the attachment position / angle of the side camera 102 to the vehicle.
  • an SfM Structure from Motion
  • the side obstacle detection unit 106 detects a side obstacle as a three-dimensional point group.
  • a side obstacle may be obtained by connecting points that are close to each other.
  • the height of the detected obstacle is equal to or less than a predetermined value, it may be excluded from the obstacle because it does not hinder the traveling of the vehicle.
  • the obstacle detection device 100 is mounted on a moving vehicle 500 and captures the front of the vehicle, that is, the front camera 101, and the second camera mounted on the vehicle 500 and images the left side of the vehicle, That is, on the actual space of the parking space detected by the parking space detection unit 103 and the parking space detection unit 103 that detects the parking space of the vehicle based on the image obtained by the side camera 102 and the first camera.
  • An obstacle calculation unit that calculates information on obstacles existing around the parking space, that is, the side obstacle detection unit 106 and the storage unit Region in the real space ⁇ been parking space, and on the basis of the information about the motion of the vehicle 500 motion acquiring unit acquires, and a processing area setting unit 105 that controls the calculation of the obstacle calculator.
  • the processing area setting unit 105 performs a side obstacle based on a parking space detected from an image obtained by photographing the front camera 101 and vehicle movement information related to translation and rotation of the vehicle 500.
  • the detection unit 106 is controlled. Therefore, the processing load due to the obstacle information calculation performed by the side obstacle detection unit 106 can be reduced.
  • the processing area setting unit 105 is based on the real space area of the parking space stored in the storage unit and the information about the behavior of the vehicle acquired by the motion acquisition unit. If the distance is longer than a predetermined distance, it is determined that the obstacle calculation unit does not perform the obstacle information calculation process (FIG. 4, step S213). For this reason, when the distance between the side camera 102 and the parking space is long, the side obstacle detection unit 106 is not calculated for obstacle calculation by not setting the processing area C (FIG. 7, step S200: NO), obstacle information calculation processing can be reduced. This is because, when the distance between the side camera 102 and the parking space is long, the image obtained by shooting with the side camera 102 is shot only with a small parking space. This is because it is difficult to calculate the object information, or even if it can be calculated, the accuracy is considered low.
  • the processing area setting unit 105 is photographed by the side camera 102 based on information on the real space of the parking space stored in the storage unit and information on the motion of the vehicle 500 acquired by the vehicle motion acquisition unit 104. Whether to calculate the area of the parking space in the obtained image, that is, the prediction area P, and cause the obstacle calculation unit to calculate based on the relationship between the prediction area P and the predetermined determination area A in the image Is determined (FIG. 4, step S215). For this reason, in the image obtained by photographing with the side camera 102, the region excluding the peripheral portion is set as the determination region A. Therefore, when the prediction region P is not included in the determination region A, the processing region C is set. Without setting, the obstacle information calculation processing can be reduced.
  • the processing area setting unit 105 captures an image of the parking space in the image obtained by the second camera, that is, when at least a part of the prediction area P is included in the determination area A. In the obtained image, an area included in the determination area A is set as a processing area C.
  • the processing area setting unit 105 causes the side obstacle detection unit 106 to perform calculation using image information including luminance and the like included in the processing area C. Therefore, the side obstacle detection unit 106 calculates the obstacle information using the image included in the determination area A with a small distortion, so that the obstacle information with high accuracy can be obtained.
  • the processing area setting unit 105 captures the image of the parking space in the image obtained by capturing with the second camera, that is, when the determination area A includes at least a part of the prediction area P. In the obtained image, a region where the determination region A and the prediction region P overlap is set as the processing region C.
  • the processing area setting unit 105 causes the side obstacle detection unit 106 to perform calculation using image information including luminance included in the processing area C. Therefore, since only the area where the parking space which is the object for which information is to be obtained is photographed is set as the processing area C, the obstacle with high accuracy while reducing the obstacle information calculation processing. Information can be obtained.
  • the obstacle detection apparatus 100 may include a camera that captures the rear of the vehicle 500 instead of the front camera 101 that captures the front of the vehicle 500, and a camera that captures the rear of the vehicle 500 in addition to the front camera 101. You may prepare. When a camera that captures the rear of the vehicle 500 is provided instead of the front camera 101, an image obtained by photographing a rear of the vehicle 500 is used instead of an image obtained by photographing the front camera 101. The processing described in the first embodiment is used.
  • the processing described in the first embodiment may be individually performed on images obtained by capturing each camera. .
  • An image may be used.
  • the obstacle detection apparatus 100 may include a camera that captures the right side of the vehicle 500 instead of the side camera 102 that captures the left side of the vehicle 500. You may provide the camera which image
  • the use of the side camera 102 and the camera that captures the right side of the vehicle 500 is the same as the use of the front camera 101 and the camera that captures the rear of the vehicle 500 described above.
  • the obstacle detection target image is a right-side image based on the function of determining whether the parking space exists on the left or right side of the vehicle using the front camera image or the rear image and the presence direction of the parking space that is the determination result. This is a function for switching between the left image and the left image.
  • the obstacle detection device 100 is connected to other devices via the CAN bus 20 of the vehicle 500.
  • the connection relationship between the obstacle detection apparatus 100 and other devices is not limited to this.
  • the obstacle detection apparatus 100 may be connected to other devices via a communication bus other than CAN, or may be directly connected to other devices without using a communication bus. Further, the obstacle detection device 100 may be incorporated in a camera device or an integrated controller.
  • the processing region setting unit 105 sets the prediction region P as the processing region C when the entire prediction region P exists in the determination region A (FIG. 4, step S215: YES, step S216).
  • the processing region setting unit 105 may set the entire determination region A as the processing region C when the entire prediction region P exists in the determination region A, or may be obtained by photographing the side camera 102.
  • the entire area of the image may be set as the processing area C.
  • the processing region setting unit 105 may set the prediction region P as the processing region C or set the determination region A as the processing region C when at least a part of the prediction region P exists in the determination region A.
  • an area where the determination area A and the prediction area P overlap may be set as the processing area C, or the entire area of the image obtained by the side camera 102 is set as the processing area C. May be.
  • the distance threshold d_th (FIG. 4, step S213) used for processing of the processing area setting unit 105 is set to infinity, and only the predicted area P of the parking space is used regardless of the distance.
  • the processing area C may be set, or the entire image of the side camera 102 may be set as the determination area A. Further, a plurality of combinations of the distance threshold d_th and the determination region A are prepared, and the processing region C of the side obstacle detection unit 106 is set by repeatedly performing the processing from step S213 to step S217 on the plurality of combinations. May be.
  • the side obstacle detection unit 106 performs processing. Settings such as enabling can be made.
  • Modification 5 In the first embodiment described above, the side obstacle detection method based on the three-dimensional position calculation of the feature points using triangulation has been described, but the processing content of the side obstacle detection unit 106 is not limited to this.
  • a technique may be used in which an internal or external parameter of the camera is used to convert the surrounding environment into an image (overhead image) viewed from directly above, and an obstacle is detected on the overhead image based on an edge or the like. In order to realize this modification, it is necessary to add the following functions to the obstacle detection apparatus 100.
  • a camera that captures the forward direction, the rear direction, the right direction, and the left direction of the vehicle, an overhead image generation function that generates an overhead image by combining images captured by the multiple cameras, and an overhead image
  • an overhead image generation function that generates an overhead image by combining images captured by the multiple cameras
  • an overhead image This is a function for detecting obstacles on the side of the vehicle using the. Even in this case, it is possible to reduce the calculation cost by converting only the processing area C set by the processing area setting unit 105 into an overhead image.
  • the processing area setting unit 105 calculates the distance between the parking space and the host vehicle, and calculates the predicted area P of the parking space when the distance is equal to or less than a predetermined threshold (FIG. 4). Step S213: YES, S214).
  • the processing area setting unit 105 calculates the relative position / posture between the parking space and the host vehicle, and calculates the predicted area P of the parking space in the image obtained by the side camera 102 without performing any special determination. May be. That is, step S214 may be executed after step S211 in FIG.
  • step S215 it is next determined whether or not the area of the prediction region P on the image is larger than a predetermined area. If it is determined that the predicted area P is larger than the predetermined area, the process proceeds to step S215 and predetermined. When it is determined that the area is equal to or smaller than the determined area, the process proceeds to step S217. The determination may be made based on the area of the region where the prediction region P and the determination region A overlap.
  • the processing area setting unit 105 determines whether the second camera, that is, the side camera 102 is based on the information on the real space of the parking space stored in the storage unit and the vehicle behavior acquired by the motion acquisition unit.
  • the area of the parking space in the image obtained by photographing, that is, the area of the prediction region P is calculated, and when the calculated area is larger than the predetermined area, the processing region C is set, and the obstacle calculation unit, that is, the side
  • the direction obstacle detection unit 106 performs calculation.
  • FIGS. 1-10 A second embodiment of the obstacle detection apparatus according to the present invention will be described with reference to FIGS.
  • the same components as those in the first embodiment are denoted by the same reference numerals, and different points will be mainly described. Points that are not particularly described are the same as those in the first embodiment.
  • This embodiment differs from the first embodiment mainly in that it includes a front obstacle detection unit.
  • FIG. 8 is a diagram showing the functions of a program executed in the obstacle detection apparatus 100a as function blocks.
  • the differences between the second embodiment and the first embodiment are as follows.
  • the front obstacle detection part 208 is further provided.
  • a processing area setting unit 105 a is provided instead of the processing area setting unit 105.
  • An image obtained by photographing the front camera 101 is transmitted to the front obstacle detection unit 208, and the parking space detection unit 103a operates based on the processing result of the front obstacle detection unit 208.
  • the output unit 107 receives input not only from the side obstacle detection unit 106 but also from the front obstacle detection unit 208.
  • the front obstacle detection unit 208 detects an obstacle present in front of the vehicle from a plurality of images obtained by photographing the front camera 101 at different times, and outputs the position of the obstacle with respect to the host vehicle.
  • the processing of the front obstacle detection unit 208 is the same as that of the side obstacle detection unit 106, and the difference is not the image obtained by photographing the side camera 102 but the image obtained by photographing the front camera 101. It is a point to use.
  • the processing area C is assumed to be set in advance.
  • the entire image of the front camera 101 may be the processing area C.
  • the distortion at the edge of the image is large, such as when the image is taken with a camera equipped with a fisheye lens, it is effective to set the vicinity of the center of the image as the processing region C.
  • the output unit 107 outputs the obstacle position in front of the vehicle detected by the front obstacle detection unit 208 and the obstacle position on the side of the vehicle detected by the side obstacle detection unit 106 to the CAN bus 20. As will be described later, only the position where the obstacle is not detected by the front obstacle detection unit 208 becomes the processing target of the side obstacle detection unit 106. Therefore, the output unit 107 outputs the obstacle position output from the front obstacle detection unit 208 and the obstacle position output from the side obstacle detection unit 106 as they are.
  • FIG. 9 is a flowchart showing processing executed in the processing area setting unit 105a.
  • the difference between this flowchart and the flowchart shown in FIG. 2 in the first embodiment is the processing from the determination in step S215 to the step S217.
  • the execution subject of each step described below is the CPU 10.
  • step S250 which is executed when an affirmative determination is made in step S215, the front obstacle detection unit 208 executed in the current processing cycle is an area in the real space that exists behind the obstacle and cannot be measured.
  • a certain unmeasured area U is calculated. That is, the area that is divided by the obstacle detected by the front obstacle detection unit 208 and is farther from the front camera 101 than the obstacle is the unmeasured area U.
  • the process proceeds to step S216a.
  • FIG. 10 is a diagram illustrating an example of an unmeasured area.
  • FIG. 10 shows a state in which an obstacle 521 existing in front of the vehicle is detected using a plurality of images obtained by photographing the front camera 101 at different times. At this time, an area between the front camera 101 and the obstacle 521 is a measured area because it is considered that no other obstacle exists, and the remaining area is an unmeasured area U.
  • step 216a an area to be processed by the side obstacle detection unit 106 is set. That is, the unmeasured area U in the real space calculated in step S215 is projected onto an image obtained by photographing the side camera 102, and an unmeasured projection area W on the image is obtained.
  • This unmeasured projection area W is defined as an area processed by the side obstacle detection unit 106, that is, a processing area C.
  • the unmeasured projection area W the relative position / posture between the unmeasured area U and the host vehicle and the external / internal parameters of the front camera 101 and the side camera 102 are used.
  • an area wider than the unmeasured projection area W may be set as the processing area C in consideration of calculation errors.
  • step S217 The operation in step S217 is the same as that in the first embodiment.
  • the obstacle detection apparatus 100 is a first camera image processing unit that calculates obstacle information based on a plurality of images obtained by photographing the first camera, that is, the front camera 101 at different times, that is, the front obstacle.
  • An object detection unit 208 is provided.
  • the parking space detection unit 103a detects the parking space based on the calculation result of the forward obstacle detection unit 208.
  • the processing area setting unit 105a determines whether or not the obstacle calculation unit performs the calculation based on whether or not the obstacle information is calculated by the front obstacle detection unit 208. Therefore, the calculation process of the side obstacle detection unit 106 can be reduced by using the calculation result of the obstacle information by the front obstacle detection unit 208 performed for detecting the parking space. Specifically, when the obstacle information is already calculated by the forward obstacle detection unit 208, the processing region C is not set in the calculated region, and the calculation by the side obstacle detection unit 106 is performed. Processing can be reduced.
  • the processing region setting unit 105a sets the region that could not be measured by the front obstacle detection unit 208 in the current processing cycle as the unmeasured region U (FIG. 9, step S250).
  • an area that cannot be measured by the front obstacle detection unit 208 not only in the current processing period but also in the previous processing period may be set as the unmeasured area U. That is, the measured region in the past processing cycle is corrected using the vehicle motion from the past processing cycle output to the current processing cycle output by the vehicle motion acquisition unit 104. And the area
  • the processing area setting unit 105a may divide the real space into grids of a predetermined size, for example, a grid with a side of 1 m, and determine whether or not the front obstacle detection unit 208 can measure each grid. Good. In this case, the grid through which the line segment connecting the obstacle detected by the front obstacle detection unit 208 and the camera position has passed is set as measured. According to this modification, when the front and rear obstacles detected by the front obstacle detection unit 208 are a three-dimensional point group that is a collection of three-dimensional positions of feature points, a measured region for each point is obtained. Since it becomes a line, the problem that the measured area becomes sparse can be solved.
  • FIGS. A third embodiment of the obstacle detection apparatus according to the present invention will be described with reference to FIGS.
  • the same components as those in the first embodiment are denoted by the same reference numerals, and different points will be mainly described. Points that are not particularly described are the same as those in the first embodiment.
  • both the front obstacle detection unit and the side obstacle detection unit calculate obstacle information for the same point, and output any obstacle information based on the reliability. This is different from the first embodiment.
  • FIG. 11 is a block diagram illustrating a configuration of the obstacle detection apparatus 100b according to the third embodiment.
  • the third embodiment in addition to the function with which the obstacle detection apparatus 100 in 1st Embodiment is provided, the front obstacle detection part 208 and the integration part 309 are provided.
  • the integration unit 309 combines the obstacle position in front of the vehicle detected by the front obstacle detection unit 208 and the obstacle position on the side of the vehicle detected by the side obstacle detection unit 106 and outputs the result to the output unit 107. .
  • step S600 the reliability of the obstacle position in front of the vehicle detected by the front obstacle detector 208 and the obstacle position on the side of the vehicle detected by the side obstacle detector 106 is calculated.
  • the reliability is determined using the magnitude of the parallax.
  • the parallax is an angle formed by a vector from the position of the obstacle to the positions of the respective cameras corresponding to the plurality of images used for calculating the obstacle.
  • the reliability is expressed by an integer from 0 to 100, for example, for easy comparison.
  • FIG. 13 is a diagram illustrating an example of parallax.
  • the side obstacle detection unit 106 performs obstacle information at the position 532 by triangulation using the position 530 of the host vehicle in the current processing cycle and the position 531 of the host vehicle in the past processing cycle. Is output.
  • the parallax is calculated as an angle 533 formed by a vector from the position 532 to the two camera positions used for triangulation. If the position of the obstacle is different, the angle formed by the vectors to the two camera positions is also different, so the reliability has a different value for each position, that is, for each obstacle.
  • the process proceeds to step S601.
  • step S601 the real space around the host vehicle is divided into grids. For example, a 10 m ⁇ 10 m space is divided into 100 1 m ⁇ 1 m grids with the host vehicle as the center. However, the size of the surrounding space and the grid may be arbitrarily changed.
  • step S602. the reliability of each camera is calculated from the reliability of each obstacle present in a certain grid. For example, the average value of the reliability of the obstacle position measured from the image of each camera is set as the camera reliability. Further, a median value may be used instead of the average value.
  • the obstacle position used for the calculation of the reliability of the camera is not limited to the obstacle position measured in the current processing cycle, and an obstacle position measured in the past processing cycle may be used.
  • step S603 based on the reliability calculated in step S602, the obstacle position measured from the image of the camera having the higher reliability of the front camera 101 and the side camera 102 is output in a certain grid.
  • the obstacle information obtained from the front camera 101 has two points of reliability 10 and 20, and the obstacle information obtained from the side camera 102 has one point of reliability 20.
  • the following output is produced: That is, since the reliability of the front camera 101 is an average value of 15 and the reliability of the side camera 102 is 20, only one point obtained from the front camera 101 is output as the obstacle information of the grid.
  • Steps S602 and S603 are processed for all the grids, and the processing of the integration unit 309 is terminated.
  • the obstacle detection apparatus 100 divides the real space into a grid pattern, and a front obstacle detection unit 208 that calculates information on the obstacle based on a plurality of images obtained by the front camera 101 shooting at different times. For each region, the reliability of the obstacle information calculated by the front obstacle detection unit 208 and the reliability of the obstacle information calculated by the side obstacle detection unit 106 are calculated, and the calculated reliability in the region is calculated. And an integration unit 309 that outputs either one of the obstacle information of the higher degree to the output unit 107.
  • the parking space detection unit 103 detects a parking space based on the calculation result of the front obstacle detection unit 208.
  • the side obstacle detection unit 106 and the front obstacle detection unit 208 each calculate information on an obstacle based on two images and the principle of triangulation.
  • the integration unit 309 determines the forward obstacle based on the angle formed by the two straight lines connecting the position of the obstacle in the real space and the respective positions in the real space of the front camera 101 when the two images are captured.
  • the reliability of the obstacle information calculated by the object detection unit 208 is calculated, the position of the obstacle in the real space, and the respective positions in the real space of the side camera 102 when the two images are taken.
  • the reliability of the obstacle information calculated by the side obstacle detection unit 106 is calculated based on the angle formed by the two straight lines connecting the two. Therefore, the reliability can be calculated by a simple calculation.
  • the processing of the integration unit 309 is not limited to the above.
  • the integration unit 309 outputs the obstacle position in front of the vehicle output by the front obstacle detection unit 208 and the vehicle output by the side obstacle detection unit 106 regardless of the reliability or without calculating the reliability.
  • the side obstacle position may be output.
  • the integration unit 309 may calculate the reliability based on the number of obstacles, that is, the number of points that can be associated, instead of parallax. That is, the integration unit 309 compares the number of obstacles detected by the front obstacle detection unit 208 and the number of obstacles detected by the side obstacle detection unit 106 in a certain grid.
  • the integration unit 309 When the front obstacle detection unit 208 detects more obstacles, the integration unit 309 outputs the obstacle information detected by the front obstacle detection unit 208 as the obstacle information of the grid. When the side obstacle detection unit 106 detects more obstacles, the integration unit 309 outputs the obstacle information detected by the side obstacle detection unit 106 as the obstacle information of the grid. To do.
  • FIGS. A fourth embodiment of the obstacle detection apparatus according to the present invention will be described with reference to FIGS.
  • the same components as those in the third embodiment are denoted by the same reference numerals, and different points will be mainly described. Points that are not particularly described are the same as those in the third embodiment.
  • the present embodiment is different from the third embodiment in that the parking space is detected mainly using the obstacle detection result output by the obstacle detection device.
  • vehicle 500 further includes a parking space post-processing detection unit.
  • FIG. 14 is a diagram illustrating the functions of a program executed in vehicle 500 in the fourth embodiment as function blocks.
  • the parking space post-processing detection unit 410 represents the functions of a program executed in the ECU connected to the CAN bus 20 (see FIG. 1) as a function block.
  • the parking space post-processing detection unit 410 receives an obstacle detection result from the obstacle detection device 100b via the CAN bus 20, and detects a parking space existing around the host vehicle.
  • FIG. 15 is a diagram illustrating an example of a parking space detected by the parking space post-processing detection unit 410.
  • the parking space post-processing detection unit 410 detects a space without an obstacle as a parking space when there is a space without an obstacle between a plurality of obstacle positions.
  • a space 544 between the obstacle positions 542 and 543 is detected as a parking space. According to the fourth embodiment, it is possible to detect a parking space that is present around the host vehicle and has no obstacles.
  • the parking space post-process detection part 410 may detect all spaces where no obstacles exist as parking spaces.
  • the parking space post-processing detection unit 410 may define a minimum parking space size in advance and detect only a space where there is no obstacle more than the minimum parking space as a parking space.
  • the parking space post-processing detection unit 410 detects a parking frame such as a white line from the image of the front camera 101 that captures the front or rear of the vehicle and the image of the side camera 102 that images the side, and parks without an obstacle
  • the frame position may be detected as a parking space.
  • the parking space post-processing detection unit 410 corrects the obstacle position in the past processing cycle using the vehicle motion from the past processing cycle output to the current processing cycle output by the vehicle motion acquisition unit 104, and the current processing cycle. It may be used as an obstacle position by adding together with the obstacle position.
  • the parking space post-processing detection unit 410 has a function of an ECU different from the obstacle detection device 100b.
  • the obstacle detection device 100b may include the parking space post-processing detection unit 410 and output the detected position of the parking space together with the obstacle detection result.
  • the present invention is not limited to the above-described embodiments, and includes various modifications.
  • the above-described embodiments have been described in detail for easy understanding of the present invention, and are not necessarily limited to those having all the configurations described.
  • Other embodiments conceivable within the scope of the technical idea of the present invention are also included in the scope of the present invention.
  • a part of the configuration of one embodiment can be replaced with the configuration of another embodiment, and the configuration of another embodiment can be added to the configuration of one embodiment.
  • Each of the above-described configurations, functions, processing units, processing means, and the like may be realized by hardware by designing a part or all of them with, for example, an integrated circuit.
  • Each of the above-described configurations, functions, and the like may be realized by software by interpreting and executing a program that realizes each function by the processor.
  • Information such as programs, tables, and files that realize each function can be stored in a memory, a hard disk, a recording device such as an SSD (Solid State Drive), or a recording medium such as an IC card, an SD card, or a DVD.
  • SSD Solid State Drive

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Multimedia (AREA)
  • Theoretical Computer Science (AREA)
  • Mechanical Engineering (AREA)
  • Chemical & Material Sciences (AREA)
  • Combustion & Propulsion (AREA)
  • Transportation (AREA)
  • Traffic Control Systems (AREA)
  • Image Analysis (AREA)
  • Control Of Driving Devices And Active Controlling Of Vehicle (AREA)
PCT/JP2016/082400 2015-11-02 2016-11-01 障害物検知装置 Ceased WO2017078001A1 (ja)

Priority Applications (2)

Application Number Priority Date Filing Date Title
EP16862061.5A EP3372456A4 (en) 2015-11-02 2016-11-01 DEVICE FOR DETECTING OBSTACLES
US15/767,884 US10242576B2 (en) 2015-11-02 2016-11-01 Obstacle detection device

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
JP2015215752A JP6514624B2 (ja) 2015-11-02 2015-11-02 障害物検知装置
JP2015-215752 2015-11-02

Publications (1)

Publication Number Publication Date
WO2017078001A1 true WO2017078001A1 (ja) 2017-05-11

Family

ID=58662016

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP2016/082400 Ceased WO2017078001A1 (ja) 2015-11-02 2016-11-01 障害物検知装置

Country Status (4)

Country Link
US (1) US10242576B2 (https=)
EP (1) EP3372456A4 (https=)
JP (1) JP6514624B2 (https=)
WO (1) WO2017078001A1 (https=)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108961836A (zh) * 2017-05-24 2018-12-07 福特全球技术公司 用于周围车辆的低频定位的方法和设备
CN112172790A (zh) * 2020-06-24 2021-01-05 上汽通用五菱汽车股份有限公司 自动停车的控制方法、装置及计算机可读存储介质
US11787334B2 (en) 2018-09-05 2023-10-17 Volvo Car Corporation Driver assistance system and method for vehicle flank safety

Families Citing this family (28)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP6930152B2 (ja) * 2017-03-14 2021-09-01 トヨタ自動車株式会社 自動運転システム
US10332002B2 (en) * 2017-03-27 2019-06-25 GM Global Technology Operations LLC Method and apparatus for providing trailer information
JP6820075B2 (ja) * 2017-07-25 2021-01-27 日本電気株式会社 乗員数検知システム、乗員数検知方法、およびプログラム
WO2019039280A1 (ja) * 2017-08-22 2019-02-28 ソニー株式会社 情報処理装置、情報処理方法、プログラム、及び、車両
JP6926976B2 (ja) * 2017-11-15 2021-08-25 株式会社アイシン 駐車支援装置及びコンピュータプログラム
US11410557B2 (en) * 2018-03-22 2022-08-09 Hitachi Astemo, Ltd. Parking assistance device
JP6771517B2 (ja) * 2018-09-07 2020-10-21 本田技研工業株式会社 自動駐車装置及び自動駐車方法
JP7212486B2 (ja) * 2018-09-26 2023-01-25 フォルシアクラリオン・エレクトロニクス株式会社 位置推定装置
JP7195131B2 (ja) 2018-12-14 2022-12-23 株式会社デンソーテン 画像処理装置および画像処理方法
JP7236857B2 (ja) 2018-12-14 2023-03-10 株式会社デンソーテン 画像処理装置および画像処理方法
JP2020095624A (ja) 2018-12-14 2020-06-18 株式会社デンソーテン 画像処理装置、および画像処理方法
JP2020095631A (ja) 2018-12-14 2020-06-18 株式会社デンソーテン 画像処理装置および画像処理方法
JP7203586B2 (ja) 2018-12-14 2023-01-13 株式会社デンソーテン 画像処理装置および画像処理方法
JP7141940B2 (ja) 2018-12-14 2022-09-26 株式会社デンソーテン 画像処理装置および画像処理方法
JP7226986B2 (ja) 2018-12-14 2023-02-21 株式会社デンソーテン 画像処理装置および画像処理方法
JP7359541B2 (ja) 2018-12-14 2023-10-11 株式会社デンソーテン 画像処理装置および画像処理方法
JP7252750B2 (ja) 2018-12-14 2023-04-05 株式会社デンソーテン 画像処理装置および画像処理方法
JP2020095620A (ja) 2018-12-14 2020-06-18 株式会社デンソーテン 画像処理装置および画像処理方法
JP7203587B2 (ja) 2018-12-14 2023-01-13 株式会社デンソーテン 画像処理装置および画像処理方法
JP2020095623A (ja) 2018-12-14 2020-06-18 株式会社デンソーテン 画像処理装置および画像処理方法
JP2020107308A (ja) * 2018-12-27 2020-07-09 財團法人工業技術研究院Ｉｎｄｕｓｔｒｉａｌ Ｔｅｃｈｎｏｌｏｇｙ Ｒｅｓｅａｒｃｈ Ｉｎｓｔｉｔｕｔｅ 駐車スポット検出システム
US10769948B2 (en) * 2018-12-27 2020-09-08 Industrial Technology Research Institute Parking spot detection system and method thereof
JP2020166758A (ja) 2019-03-29 2020-10-08 株式会社デンソーテン 画像処理装置および画像処理方法
JP7323356B2 (ja) * 2019-06-28 2023-08-08 フォルシアクラリオン・エレクトロニクス株式会社 駐車支援装置及び駐車支援方法
KR20220099949A (ko) * 2019-09-22 2022-07-14 바야비전 센싱 리미티드 자율 주행에 있어 기능적 안정성
CN112634487B (zh) * 2019-09-24 2022-08-16 北京百度网讯科技有限公司 用于输出信息的方法和装置
JP7758531B2 (ja) 2021-10-29 2025-10-22 Ａｓｔｅｍｏ株式会社 画像処理装置および車載制御装置
US12110009B2 (en) * 2023-01-06 2024-10-08 Valeo Schalter Und Sensoren Gmbh Parking space detection method and system

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2004108944A (ja) * 2002-09-18 2004-04-08 Nissan Motor Co Ltd 障害物検出装置
JP2009101989A (ja) * 2007-10-01 2009-05-14 Nissan Motor Co Ltd 縦列駐車支援装置及び縦列駐車支援方法
JP2012040883A (ja) * 2008-12-19 2012-03-01 Panasonic Corp 車両周囲画像生成装置

Family Cites Families (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP4234374B2 (ja) * 2002-08-21 2009-03-04 三菱自動車工業株式会社 駐車支援装置
JP4428390B2 (ja) * 2007-02-15 2010-03-10 トヨタ自動車株式会社 駐車支援装置及び駐車支援方法
JP2010208358A (ja) 2009-03-06 2010-09-24 Toyota Industries Corp 駐車支援装置
JP5397321B2 (ja) * 2009-06-09 2014-01-22 株式会社デンソー 駐車支援システム
KR101305630B1 (ko) * 2011-09-30 2013-09-09 현대자동차주식회사 주차영역 자동 가변형 주차보조 시스템 및 그 방법
JP2014054912A (ja) * 2012-09-12 2014-03-27 Denso Corp 駐車支援システム
JP2015009646A (ja) * 2013-06-28 2015-01-19 アルパイン株式会社 運転支援装置

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2004108944A (ja) * 2002-09-18 2004-04-08 Nissan Motor Co Ltd 障害物検出装置
JP2009101989A (ja) * 2007-10-01 2009-05-14 Nissan Motor Co Ltd 縦列駐車支援装置及び縦列駐車支援方法
JP2012040883A (ja) * 2008-12-19 2012-03-01 Panasonic Corp 車両周囲画像生成装置

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP3372456A4 *

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108961836A (zh) * 2017-05-24 2018-12-07 福特全球技术公司 用于周围车辆的低频定位的方法和设备
US11787334B2 (en) 2018-09-05 2023-10-17 Volvo Car Corporation Driver assistance system and method for vehicle flank safety
CN112172790A (zh) * 2020-06-24 2021-01-05 上汽通用五菱汽车股份有限公司 自动停车的控制方法、装置及计算机可读存储介质
CN112172790B (zh) * 2020-06-24 2022-10-28 上汽通用五菱汽车股份有限公司 自动停车的控制方法、装置及计算机可读存储介质

Also Published As

Publication number Publication date
US10242576B2 (en) 2019-03-26
JP6514624B2 (ja) 2019-05-15
EP3372456A1 (en) 2018-09-12
JP2017087758A (ja) 2017-05-25
EP3372456A4 (en) 2019-07-10
US20180308364A1 (en) 2018-10-25

Similar Documents

Publication Publication Date Title
JP6514624B2 (ja) 障害物検知装置
JP5926228B2 (ja) 自律車両用の奥行き検知方法及びシステム
KR101725060B1 (ko) 그래디언트 기반 특징점을 이용한 이동 로봇의 위치를 인식하기 위한 장치 및 그 방법
KR101776622B1 (ko) 다이렉트 트래킹을 이용하여 이동 로봇의 위치를 인식하기 위한 장치 및 그 방법
KR101784183B1 (ko) ADoG 기반 특징점을 이용한 이동 로봇의 위치를 인식하기 위한 장치 및 그 방법
JP3719095B2 (ja) 挙動検出装置および勾配検出方法
KR101708659B1 (ko) 이동 로봇의 맵을 업데이트하기 위한 장치 및 그 방법
KR101776621B1 (ko) 에지 기반 재조정을 이용하여 이동 로봇의 위치를 인식하기 위한 장치 및 그 방법
KR101961001B1 (ko) 단일-카메라 거리 추정
JP4958279B2 (ja) 物体検出装置
KR20090103165A (ko) 모노큘러 모션 스테레오 기반의 주차 공간 검출 장치 및방법
WO2017134915A1 (ja) カメラキャリブレーション装置
JP2014137815A (ja) 歪みのあるカメラ画像を補正するシステム及び方法
EP3403216A1 (en) Systems and methods for augmenting upright object detection
JP2016152027A (ja) 画像処理装置、画像処理方法およびプログラム
WO2020137110A1 (ja) 移動量推定装置
KR102555269B1 (ko) 전방향 영상센서 및 관성측정센서의 자세추정 융합 방법 및 시스템
JP7727502B2 (ja) 画像処理装置
JPWO2017090097A1 (ja) 車両用外界認識装置
JP6174884B2 (ja) 車外環境認識装置および車外環境認識方法
KR102479253B1 (ko) 차량용 카메라 영상 기반 공차 보정 방법
KR20200102108A (ko) 차량의 객체 검출 장치 및 방법
WO2021106510A1 (ja) 障害物認識装置
JP2021033605A (ja) 画像処理装置、および、画像処理方法
JP7199337B2 (ja) 位置推定装置、位置推定方法およびプログラム

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 16862061

Country of ref document: EP

Kind code of ref document: A1

DPE1 Request for preliminary examination filed after expiration of 19th month from priority date (pct application filed from 20040101)
WWE Wipo information: entry into national phase

Ref document number: 15767884

Country of ref document: US

NENP Non-entry into the national phase

Ref country code: DE

WWW Wipo information: withdrawn in national office

Ref document number: 2016862061

Country of ref document: EP