Vehicle obstacle avoidance method, device, equipment and storage medium

Abstract

The present invention discloses a vehicle obstacle avoidance method, device, equipment and storage medium, the method comprising: filtering the point cloud data collected by the target sensor through a preset vehicle contour model corresponding to the vehicle load state to obtain a target point cloud data set; filtering the target point cloud data set based on the current vehicle motion state to obtain an obstacle point cloud data set; determining the spatial position relationship between the vehicle and the obstacle corresponding to the obstacle point cloud data set based on the current vehicle motion state and the obstacle point cloud data set; and controlling the vehicle's moving speed according to the spatial position relationship and preset obstacle avoidance parameters. Since the present invention filters the point cloud data through the vehicle contour model and the current vehicle motion state to obtain an obstacle point cloud data set, and then controls the vehicle's moving speed according to the spatial position relationship between the vehicle and the obstacle corresponding to the obstacle point cloud data set and the obstacle avoidance parameters, accurate obstacle avoidance of the vehicle is achieved, ensuring the vehicle's driving safety.

Description

Technical Field

[0001] The present invention relates to the field of vehicle safety technology, and in particular to a vehicle obstacle avoidance method, device, equipment and storage medium. Background Art

[0002] With the development of science and technology and the advancement of automatic navigation technology, AGVs are increasingly used in life and industry. In the industrial production process, it is very important to ensure the driving safety of AGVs, so the requirements for obstacle avoidance technology are getting higher and higher.

[0003] The existing technology can obtain the distance between the AGV and the obstacle based on the data in the sensor and avoid the obstacle. However, due to the inaccuracy of the sensor data transmission, the distance feedback between the AGV and the obstacle is inaccurate, which causes the AGV to collide with the obstacle while driving and cause damage to the AGV. Therefore, how to accurately avoid obstacles and improve vehicle driving safety has become an urgent problem to be solved.

[0004] The above contents are only used to assist in understanding the technical solution of the present invention and do not constitute an admission that the above contents are prior art. Summary of the invention

[0005] The main purpose of the present invention is to provide a vehicle obstacle avoidance method, device, equipment and storage medium, aiming to solve the technical problem that the obstacle avoidance technology in the prior art cannot achieve accurate obstacle avoidance of the vehicle and ensure the driving safety of the vehicle.

[0006] To achieve the above object, the present invention provides a vehicle obstacle avoidance method, the method comprising the following steps:

[0007] The point cloud data collected by the target sensor is filtered through a preset vehicle contour model corresponding to the vehicle load state to obtain a target point cloud data set;

[0008] Filtering the target point cloud data set based on the current vehicle motion state to obtain an obstacle point cloud data set;

[0009] Determine a spatial position relationship between the vehicle and an obstacle corresponding to the obstacle point cloud data set based on the current vehicle motion state and the obstacle point cloud data set;

[0010] The moving speed of the vehicle is controlled according to the spatial position relationship and preset obstacle avoidance parameters.

[0011] Optionally, the step of filtering the point cloud data collected by the target sensor by using a preset vehicle profile model corresponding to the vehicle load state to obtain a target point cloud data set includes:

[0012] Establishing a vehicle body coordinate system by using a preset vehicle contour model corresponding to a vehicle load state, wherein the vehicle load state includes a loaded state or an unloaded state;

[0013] Acquire point cloud data collected by a target sensor, wherein the target sensor includes a laser radar and a stereo camera;

[0014] Establishing a sensor coordinate system based on the target sensor, and acquiring a first coordinate position of the point cloud data in the sensor coordinate system;

[0015] Performing coordinate transformation on the first coordinate position through a coordinate transformation matrix between the target sensor and the center of the vehicle to obtain a second coordinate position of the point cloud data in the vehicle body coordinate system;

[0016] Determining whether the second coordinate position is located in a preset area of ​​the vehicle body coordinate system;

[0017] If so, the point cloud data corresponding to the second coordinate position is screened to obtain a target point cloud data set.

[0018] Optionally, the step of screening the target point cloud data set based on the current vehicle motion state to obtain the obstacle point cloud data set includes:

[0019] Generate a corresponding walking area based on the current vehicle motion state and the preset vehicle contour model;

[0020] Determining whether the target point cloud data in the target point cloud data set is located in the walking area;

[0021] If so, the target point cloud data is screened to obtain an obstacle point cloud data set.

[0022] Optionally, the step of determining the spatial position relationship between the vehicle and an obstacle corresponding to the obstacle point cloud data set based on the current vehicle motion state and the obstacle point cloud data set comprises:

[0023] Establishing a vehicle rotation center coordinate system based on the current vehicle motion state;

[0024] Obtain obstacle projection point cloud data of the obstacle point cloud data set on the preset vehicle contour model;

[0025] Obtaining the obstacle point cloud coordinates corresponding to the obstacle point cloud data in the vehicle rotation center coordinate system, and the obstacle projection point cloud coordinates corresponding to the obstacle projection point cloud data in the vehicle rotation center coordinate system;

[0026] The spatial position relationship between the vehicle and the obstacle corresponding to the obstacle point cloud data set is determined based on the obstacle point cloud coordinates and the obstacle projection point cloud coordinates.

[0027] Optionally, the current vehicle motion state includes a spinning motion state, and the step of determining the spatial position relationship between the vehicle and the obstacle corresponding to the obstacle point cloud data set based on the obstacle point cloud coordinates and the obstacle projection point cloud coordinates includes:

[0028] Determining a straight-line distance between the obstacle point cloud coordinates and the rotation center coordinates of the vehicle rotation center coordinate system based on the obstacle point cloud coordinates and the rotation center coordinates of the vehicle rotation center coordinate system;

[0029] Acquire a first angle between the obstacle point cloud data and the target axis of the vehicle rotation center coordinate system based on the straight-line distance and the obstacle point cloud coordinates;

[0030] Acquire a second angle between the obstacle projection point cloud data and the target axis of the vehicle rotation center coordinate system based on the straight-line distance and the obstacle projection point cloud coordinates;

[0031] Determine a target angle between the vehicle and an obstacle corresponding to the obstacle point cloud data set according to the first angle and the second angle;

[0032] The step of controlling the moving speed of the vehicle according to the spatial position relationship and the preset obstacle avoidance parameters comprises:

[0033] The moving speed of the vehicle is controlled according to the target angle and preset obstacle avoidance parameters.

[0034] Optionally, the current vehicle motion state includes a non-spinning motion state, and the step of determining the spatial position relationship between the vehicle and the obstacle corresponding to the obstacle point cloud data set based on the obstacle point cloud coordinates and the obstacle projection point cloud coordinates includes:

[0035] Determine a target distance between the vehicle and an obstacle corresponding to the obstacle point cloud data set based on a straight-line distance between the obstacle point cloud coordinates and the rotation center coordinates and the target angle;

[0036] The step of controlling the moving speed of the vehicle according to the spatial position relationship and the preset obstacle avoidance parameters comprises:

[0037] The moving speed of the vehicle is controlled according to the target distance and preset obstacle avoidance parameters.

[0038] Optionally, the step of filtering the point cloud data collected by the target sensor by using a preset vehicle profile model corresponding to the vehicle load state to obtain a target point cloud data set includes:

[0039] The point cloud data collected by the target sensor is filtered through a preset vehicle contour model corresponding to the vehicle load state to obtain target point cloud data;

[0040] The target point cloud data is subjected to a preset filtering process, and a target point cloud data set is obtained based on the filtered target point cloud data.

[0041] In addition, to achieve the above-mentioned purpose, the present invention also provides a vehicle obstacle avoidance device, the device comprising:

[0042] A first data screening module is used to screen the point cloud data collected by the target sensor through a preset vehicle profile model corresponding to the vehicle load state to obtain a target point cloud data set;

[0043] A second data screening module is used to screen the target point cloud data set based on the current vehicle motion state to obtain an obstacle point cloud data set;

[0044] a position relationship determination module, configured to determine a spatial position relationship between the vehicle and an obstacle corresponding to the obstacle point cloud data set based on the current vehicle motion state and the obstacle point cloud data set;

[0045] A speed control module is used to control the moving speed of the vehicle according to the spatial position relationship and preset obstacle avoidance parameters.

[0046] In addition, to achieve the above-mentioned purpose, the present invention also proposes a vehicle obstacle avoidance device, which includes: a memory, a processor, and a vehicle obstacle avoidance program stored in the memory and executable on the processor, wherein the vehicle obstacle avoidance program is configured to implement the steps of the vehicle obstacle avoidance method described above.

[0047] In addition, to achieve the above-mentioned purpose, the present invention also proposes a storage medium, on which a vehicle obstacle avoidance program is stored. When the vehicle obstacle avoidance program is executed by a processor, the steps of the vehicle obstacle avoidance method described above are implemented.

[0048] The present invention discloses a vehicle obstacle avoidance method, device, equipment and storage medium, the method comprising: filtering point cloud data collected by a target sensor through a preset vehicle contour model corresponding to a vehicle load state to obtain a target point cloud data set; filtering the target point cloud data set based on a current vehicle motion state to obtain an obstacle point cloud data set; determining the spatial position relationship between the vehicle and an obstacle corresponding to the obstacle point cloud data set based on the current vehicle motion state and the obstacle point cloud data set; controlling the vehicle's moving speed according to the spatial position relationship and preset obstacle avoidance parameters; compared with the prior art in which the distance between a vehicle and an obstacle is directly determined based on inaccurate data in a vehicle sensor to avoid obstacles, the present invention filters the collected point cloud data according to a vehicle contour model to obtain a target point cloud data set, filters the target point cloud data set according to the current vehicle motion state to obtain an obstacle point cloud data set, and then determines the spatial position relationship between the vehicle and the obstacle according to the current vehicle motion state, and controls the vehicle's moving speed according to the spatial position relationship and the obstacle avoidance parameters, thereby solving the technical problem that the obstacle avoidance method in the prior art cannot achieve accurate obstacle avoidance of the vehicle, thereby ensuring the driving safety of the vehicle. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 It is a structural schematic diagram of a vehicle obstacle avoidance device in a hardware operating environment involved in an embodiment of the present invention;

[0050] Figure 2 It is a schematic diagram of the flow chart of the first embodiment of the vehicle obstacle avoidance method of the present invention;

[0051] Figure 3 It is a flow chart of a second embodiment of the vehicle obstacle avoidance method of the present invention;

[0052] Figure 4 A schematic diagram of a flow chart of a third embodiment of a vehicle obstacle avoidance method according to the present invention;

[0053] Figure 5 It is a schematic diagram of the obstacle avoidance principle of the vehicle when the vehicle is in a spinning motion state in the third embodiment of the vehicle obstacle avoidance method of the present invention;

[0054] Figure 6 This is a schematic diagram of the vehicle's obstacle avoidance principle when the vehicle is in a non-spinning motion state in the third embodiment of the vehicle obstacle avoidance method of the present invention;

[0055] Figure 7 It is a structural block diagram of the first embodiment of the vehicle obstacle avoidance device of the present invention.

[0056] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0057] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0058] Reference Figure 1 , Figure 1 A schematic diagram of the structure of a vehicle obstacle avoidance device in a hardware operating environment according to an embodiment of the present invention.

[0059] like Figure 1 As shown, the vehicle obstacle avoidance device may include: a processor 1001, such as a central processing unit (CPU), a communication bus 1002, a user interface 1003, a network interface 1004, and a memory 1005. Among them, the communication bus 1002 is used to realize the connection and communication between these components. The user interface 1003 may include a display screen (Display), an input unit such as a keyboard (Keyboard), and the optional user interface 1003 may also include a standard wired interface and a wireless interface. The network interface 1004 may optionally include a standard wired interface and a wireless interface (such as a wireless fidelity (Wireless-Fidelity, Wi-Fi) interface). The memory 1005 may be a high-speed random access memory (Random Access Memory, RAM), or a stable non-volatile memory (Non-Volatile Memory, NVM), such as a disk storage. The memory 1005 may also be a storage device independent of the aforementioned processor 1001.

[0060] Those skilled in the art will understand that Figure 1 The structure shown in the figure does not constitute a limitation on the vehicle obstacle avoidance device, and may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently.

[0061] like Figure 1 As shown, the memory 1005 as a storage medium may include an operating system, a network communication module, a user interface module, and a vehicle obstacle avoidance program.

[0062] exist Figure 1 In the vehicle obstacle avoidance device shown, the network interface 1004 is mainly used for data communication with the network server; the user interface 1003 is mainly used for data interaction with the user; the processor 1001 and the memory 1005 in the vehicle obstacle avoidance device of the present invention can be set in the vehicle obstacle avoidance device, and the vehicle obstacle avoidance device calls the vehicle obstacle avoidance program stored in the memory 1005 through the processor 1001, and executes the vehicle obstacle avoidance method provided by the embodiment of the present invention.

[0063] The embodiment of the present invention provides a vehicle obstacle avoidance method, referring to Figure 2 , Figure 2 FIG. 4 is a flow chart of a first embodiment of a vehicle obstacle avoidance method according to the present invention.

[0064] In this embodiment, the vehicle obstacle avoidance method includes the following steps:

[0065] Step S10: filtering the point cloud data collected by the target sensor through a preset vehicle contour model corresponding to the vehicle load state to obtain a target point cloud data set.

[0066] It should be noted that the method of this embodiment can be applied in the scenario of AGV vehicle obstacle avoidance, or in other scenarios where the vehicle needs to avoid obstacles. The executor of the method of this embodiment can be a vehicle obstacle avoidance device for performing obstacle avoidance in the vehicle, or other vehicle obstacle avoidance systems that can achieve the same or similar functions and include the vehicle obstacle avoidance device. Here, the vehicle obstacle avoidance method provided in this embodiment and the following embodiments is specifically described with a vehicle obstacle avoidance system (hereinafter referred to as the system).

[0067] It should be understood that the above-mentioned vehicle load state can be a vehicle loaded state or a vehicle unloaded state, wherein the vehicle loaded state can be the corresponding driving state of the vehicle when transporting goods, and the vehicle unloaded state can be the corresponding initial driving state of the vehicle.

[0068] It can be understood that the above-mentioned preset vehicle contour model can be a rectangular diagram drawn according to the vehicle contour. When the vehicle is transporting goods, the size of the goods may be larger than the size of the vehicle. At this time, the corresponding vehicle contour model will be larger than the initial contour model. Therefore, it is necessary to obtain the corresponding vehicle contour model according to the current load status of the vehicle.

[0069] It should be noted that the above-mentioned target sensors can be sensors corresponding to lidar and stereo cameras, among which the lidar can be placed symmetrically on both sides of the bottom of the vehicle left and right or front and back, and the stereo camera can be placed on the top of the vehicle. Since the stereo camera has a wide field of view, it can obtain point cloud data of obstacles that are close and high in position. The lidar has a long detection distance and high accuracy, and can detect point cloud data of lower obstacles. The integration of the point cloud data of the sensors in the lidar and stereo camera can detect obstacles at higher positions, and can also accurately detect obstacles farther away, so that the system has a wider range when collecting point cloud data of obstacles, and the collected point cloud data is more accurate.

[0070] It should be understood that the above-mentioned target point cloud data set may be a set of point cloud data located outside the vehicle.

[0071] In a specific implementation, the current load status of the vehicle can be obtained first. If the vehicle is currently in a loaded state, the vehicle contour model corresponding to the vehicle in the loaded state is obtained, and the point cloud data within a certain range collected by the laser radar and stereo camera in the vehicle is obtained. The collected point cloud data is then classified in combination with the vehicle contour model into point cloud data located inside the vehicle and point cloud data located outside the vehicle. Since the point cloud data located inside the vehicle will not affect the driving of the vehicle, it can be eliminated. At this time, the point cloud data located outside the vehicle that may affect the driving of the vehicle is fused to obtain the target point cloud data set.

[0072] Step S20: Filtering the target point cloud data set based on the current vehicle motion state to obtain an obstacle point cloud data set.

[0073] It should be noted that the above-mentioned current vehicle motion state can be a non-spinning motion state and a spinning motion state, wherein the non-spinning motion state can be forward, backward, lateral movement and turning states, and the spinning motion state can be spinning in place, etc. This embodiment does not limit this.

[0074] It should be understood that the above obstacle point cloud data set can be a set of point cloud data that affects the vehicle's driving in the point cloud data located outside the vehicle. Since the vehicle drives in different ways or ranges in different motion states, the obstacles corresponding to the target point cloud data have different effects on the vehicle's driving. Therefore, the data in the target point cloud set can be filtered according to the current vehicle motion state. For example, when the vehicle's motion state is forward, the point cloud data located behind the vehicle can be eliminated; when the vehicle is moving sideways, the point cloud data located in front of and behind the vehicle can be eliminated. After eliminating the point cloud data that does not affect the vehicle's driving, the remaining point cloud data is the obstacle point cloud data, and the obstacle point cloud data is then fused to obtain the obstacle point cloud data set.

[0075] Furthermore, in order to accurately obtain point cloud data corresponding to obstacles that may affect vehicle driving, the above method of this embodiment also includes: generating a corresponding walking area based on the current vehicle motion state and the preset vehicle contour model; determining whether the target point cloud data in the target point cloud data set is located in the walking area; if so, screening the target point cloud data to obtain an obstacle point cloud data set.

[0076] It can be understood that the above-mentioned walking area can be the area where the vehicle is about to travel or may travel. For example, when the vehicle is in a forward state, the area where the vehicle travels is the road section in front of the vehicle; when the vehicle is in a lateral state, the area where the vehicle travels is the left and right sides; when the vehicle is in a turning state, the rotation radius is determined according to the wheel speeds of the left and right wheels of the vehicle, and the center of the circle is determined according to the direction of movement of the vehicle. At this time, an inner circle and an outer circle can be obtained, and the area between the two circles is the walking area of ​​the vehicle.

[0077] In a specific implementation, the system can obtain the walking area corresponding to the vehicle based on the vehicle contour model and the current motion state of the vehicle, and judge the point cloud data in the target point cloud data that may affect the vehicle's driving based on the walking area, thereby obtaining an obstacle point cloud data set. For example, when the vehicle's motion state is forward, obstacles behind the vehicle will not affect the vehicle's driving, and the point cloud data corresponding to the rear obstacles can be eliminated. However, obstacles in front of or on the side of the vehicle may affect the vehicle's driving, thereby obtaining an obstacle point cloud data set located in front of the vehicle.

[0078] Step S30: determining a spatial position relationship between the vehicle and an obstacle corresponding to the obstacle point cloud data set based on the current vehicle motion state and the obstacle point cloud data set.

[0079] It should be noted that the above-mentioned spatial position relationship can be an angle relationship or a distance relationship between the vehicle and the obstacle. When the current motion state of the vehicle is a non-spinning motion state, the relationship between the vehicle and the obstacle is a distance relationship. When the current motion state of the vehicle is a spinning motion state, the relationship between the vehicle and the obstacle is an angle relationship.

[0080] Step S40: controlling the moving speed of the vehicle according to the spatial position relationship and preset obstacle avoidance parameters.

[0081] It should be understood that the above-mentioned preset obstacle avoidance parameters can be preset parking distances or parking angles. In practical applications, the system can compare the shortest distance or minimum angle between the vehicle and the obstacle with the preset parking distance or parking angle. If the distance or angle between the vehicle and the obstacle is less than the preset parking distance or parking angle, it is necessary to stop the vehicle by controlling the vehicle's moving speed. If the distance or angle between the vehicle and the obstacle is greater than the preset parking distance or parking angle, the vehicle drives normally. In addition, a deceleration parameter can also be set, that is, the vehicle deceleration distance or deceleration angle. When the distance or angle between the vehicle and the obstacle is less than the preset deceleration distance or deceleration angle, the vehicle's moving speed is reduced to ensure driving safety.

[0082] Furthermore, in order to reduce unnecessary point cloud computing and resource waste and improve data processing efficiency, the above method of this embodiment also includes: filtering the point cloud data collected by the target sensor through a preset vehicle contour model corresponding to the vehicle load state to obtain target point cloud data; performing preset filtering processing on the target point cloud data, and obtaining a target point cloud data set based on the filtered target point cloud data.

[0083] It is understandable that the above-mentioned preset filtering process can be voxel filtering or other processing methods for filtering and denoising the data, and this embodiment does not limit this. By filtering the point cloud data, the influence of noise points on the distance calculation between the vehicle and the obstacle can be reduced, thereby making the calculated distance more accurate and further ensuring the safety of vehicle driving.

[0084] The present embodiment discloses filtering the point cloud data collected by the target sensor through a preset vehicle contour model corresponding to the vehicle load state to obtain a target point cloud data set; filtering the target point cloud data set based on the current vehicle motion state to obtain an obstacle point cloud data set; determining the spatial position relationship between the vehicle and the obstacle corresponding to the obstacle point cloud data set based on the current vehicle motion state and the obstacle point cloud data set; controlling the vehicle's moving speed according to the spatial position relationship and preset obstacle avoidance parameters; compared with the prior art of directly judging the distance between the car and the obstacle according to the inaccurate data in the vehicle sensor to avoid the obstacle, since the present embodiment filters the collected point cloud data according to the vehicle contour model to obtain the target point cloud data set, filters the target point cloud data set according to the current vehicle motion state to obtain the obstacle point cloud data set, and then determines the spatial position relationship between the vehicle and the obstacle according to the current vehicle motion state, and controls the vehicle's moving speed according to the spatial position relationship and the obstacle avoidance parameters, the technical problem that the obstacle avoidance method in the prior art cannot achieve accurate obstacle avoidance of the car is solved, thereby ensuring the driving safety of the vehicle. At the same time, the preset vehicle contour model can be dynamically adjusted according to the vehicle's load status without restricting the vehicle's driving status, so that the vehicle can avoid obstacles in different driving conditions.

[0085] refer to Figure 3 , Figure 3 FIG. 4 is a flow chart of a second embodiment of a vehicle obstacle avoidance method according to the present invention.

[0086] Based on the above first embodiment, in order to accurately obtain the point cloud data located outside the vehicle, in this embodiment, step S10 includes:

[0087] Step S101: establishing a vehicle body coordinate system through a preset vehicle contour model corresponding to a vehicle load state, wherein the vehicle load state includes a loaded state or an unloaded state.

[0088] It should be noted that the above-mentioned vehicle body coordinate system can be a spatial coordinate system established based on the vehicle contour model, consisting of an x-axis, a y-axis and a z-axis, wherein the x-axis points to the front of the vehicle body, the y-axis points to the left side of the vehicle body, and the z-axis points to the top of the vehicle body.

[0089] Step S102: Acquire point cloud data collected by a target sensor, where the target sensor includes a laser radar and a stereo camera.

[0090] It can be understood that the point cloud data acquired by the above laser radar can be a two-dimensional point cloud data set P laser , the point cloud data acquired by the above stereo camera can be a three-dimensional point cloud data set P camera .

[0091] Step S103: establishing a sensor coordinate system based on the target sensor, and acquiring a first coordinate position of the point cloud data in the sensor coordinate system.

[0092] It should be understood that the above-mentioned sensor coordinate system can be a coordinate system established based on the actual coordinate position and installation angle of the target sensor relative to the center of the vehicle body. Therefore, before establishing the sensor coordinate system, the actual coordinate position and installation angle of the laser radar and the stereo camera relative to the center of the vehicle body can be measured.

[0093] It should be noted that the above-mentioned first coordinate position can be the two-dimensional coordinate position corresponding to the two-dimensional point cloud data obtained by the laser radar in the sensor coordinate system and / or the three-dimensional coordinate position corresponding to the three-dimensional point cloud data obtained by the stereo camera in the sensor coordinate system.

[0094] Step S104: performing coordinate transformation on the first coordinate position through a coordinate transformation matrix between the target sensor and the center of the vehicle to obtain a second coordinate position of the point cloud data in the vehicle body coordinate system.

[0095] It should be understood that the above coordinate conversion matrix may be a coordinate change matrix of the sensor coordinate system and the vehicle coordinate system, and the above coordinate conversion matrix may convert the coordinate position of the point cloud data in the sensor coordinate system into the coordinate position in the vehicle coordinate system.

[0096] It can be understood that the above-mentioned second coordinate position can be the two-dimensional coordinate position corresponding to the point cloud data in the lidar, for example: coordinate (x1, y1), and / or the coordinate position corresponding to the point cloud data obtained by the stereo camera in the vehicle body coordinate system, for example: (x2, y2, z2).

[0097] Step S105: determining whether the second coordinate position is located in a preset area of ​​the vehicle body coordinate system.

[0098] It should be noted that the above-mentioned preset area can be the area outside the vehicle body. After the coordinate positions of the point cloud data collected by the laser radar and the stereo camera are converted into the vehicle body coordinate system, the point cloud data can be compared with the size of the coordinate value of the point cloud data and the size of the individual axes in the vehicle body coordinate system to determine whether the point cloud data is inside or outside the vehicle body, and the point cloud data located inside the vehicle body can be eliminated to obtain the target point cloud data located outside the vehicle body.

[0099] Step S106: If yes, the point cloud data corresponding to the second coordinate position is screened to obtain a target point cloud data set.

[0100] In the specific implementation, the vehicle body coordinate system can be divided into four areas based on the center of the vehicle body, namely the front side of the vehicle body, the rear side of the vehicle body, the left side of the vehicle body and the right side of the vehicle body. At this time, the distance between the center of the vehicle body and the front side of the vehicle body can be x max , the distance from the rear end of the vehicle body can be x min , the distance from the leftmost side of the vehicle body can be y max , the distance from the rightmost side of the vehicle body can be y min , and then extract and convert it into a two-dimensional point cloud data set P in the vehicle body coordinate system laser and the 3D point cloud data set P camera , and filter the point cloud data based on the coordinate position of the point cloud data. For example: for the two-dimensional point cloud data with coordinates (x1, y1), if x1 is less than x max and is greater than x min , then the point cloud is removed, otherwise it is retained. If y1 is less than y max and is greater than y min , then the point cloud is removed, otherwise it is retained; for the three-dimensional point cloud data with coordinates (x2, y2, z2), if x2 is less than x max and is greater than x min , then the point cloud is removed, otherwise it is retained. If the y2 of the obstacle point cloud is less than y max and is greater than y min , then the point cloud is discarded, otherwise it is retained. If z2 is less than or equal to zero, it is regarded as a ground point cloud, then the ground point cloud is discarded, otherwise it is retained.

[0101] This embodiment establishes a vehicle body coordinate system based on the vehicle contour model, establishes a sensor coordinate system based on the target sensor, and converts the first coordinate position of the point cloud data in the sensor coordinate system into a second coordinate position in the vehicle body coordinate system through a coordinate conversion matrix, and then screens the point cloud data based on the second coordinate position and a preset area of ​​the vehicle body coordinate system to obtain a target point cloud set located outside the vehicle body, so that the position of the point cloud data collected by the sensor can be corresponded to the vehicle body coordinate system, and the coordinate position of the point cloud data is compared with the maximum distance from the center point of the vehicle body coordinate system to the vehicle body in all directions, so as to accurately eliminate the point cloud data located inside the vehicle body, prevent the point cloud data located inside the vehicle from interfering with the system's calculation of the distance between the vehicle and the obstacle, and further improve the accuracy of the obstacle avoidance distance calculation.

[0102] refer to Figure 4 , Figure 4 FIG. 4 is a flow chart of a third embodiment of a vehicle obstacle avoidance method according to the present invention.

[0103] Based on the above embodiments, in order to determine the positional relationship between the obstacle and the vehicle, in this embodiment, step S30 includes:

[0104] Step S301: establishing a vehicle rotation center coordinate system based on the current vehicle motion state.

[0105] It should be noted that the above-mentioned vehicle rotation center coordinate system can be a coordinate system established with the center point of rotation of the vehicle when it is in motion as the rotation center coordinate system. In practical applications, when the vehicle is in a spinning motion state, the rotation center coordinate system is as follows: Figure 5 As shown, the vehicle rotation center coordinate system is the vehicle body coordinate system xoy; when the vehicle is in a non-spinning motion state, the rotation center coordinate system is as follows Figure 6 As shown, the rotation center point O(x0, y0) of the vehicle can be determined according to the current motion state of the vehicle, and then the rotation center coordinate system XOY can be established with the rotation center point O as the coordinate point.

[0106] Step S302: Obtain obstacle projection point cloud data of the obstacle point cloud data in the obstacle point cloud data set on the preset vehicle contour model.

[0107] It should be understood that the above obstacle point cloud data may be a set of obstacle point cloud data P obs , obtain a point cloud data P in the obstacle point cloud data set.

[0108] It can be understood that the above-mentioned obstacle projection point cloud data can be the projection point cloud data of the above-mentioned obstacle point cloud data on the preset vehicle contour model. For example, the projection of the obstacle point cloud data P on the preset vehicle contour model is P', and P' is the obstacle projection point cloud data corresponding to the above-mentioned obstacle point cloud data in the preset vehicle contour model.

[0109] Step S303: Obtain the obstacle point cloud coordinates corresponding to the obstacle point cloud data in the vehicle rotation center coordinate system, and the obstacle projection point cloud coordinates corresponding to the obstacle projection point cloud data in the vehicle rotation center coordinate system.

[0110] It should be noted that the above-mentioned obstacle point cloud coordinates can be the coordinate positions of the above-mentioned obstacle point cloud data in the vehicle rotation center coordinate system. For example: the coordinates of the obstacle point cloud data P in the vehicle rotation center coordinate system are (x, y), and correspondingly, the coordinates of the obstacle point cloud data P' in the vehicle rotation center coordinate system are (x', y').

[0111] Step S304: determining the spatial position relationship between the vehicle and the obstacle corresponding to the obstacle point cloud data set based on the obstacle point cloud coordinates and the obstacle projection point cloud coordinates.

[0112] It can be understood that when the vehicle is in different motion states, the spatial position relationship between the vehicle and the obstacle is also different. When the current motion state of the vehicle is a spinning motion state, the above step S304 includes: determining the straight-line distance between the obstacle point cloud coordinates and the rotation center coordinates of the vehicle rotation center coordinate system based on the obstacle point cloud coordinates and the rotation center coordinates of the vehicle rotation center coordinate system; obtaining the first angle between the obstacle point cloud data and the target axis of the vehicle rotation center coordinate system based on the straight-line distance and the obstacle point cloud coordinates; obtaining the second angle between the obstacle projection point cloud data and the target axis of the vehicle rotation center coordinate system based on the straight-line distance and the obstacle projection point cloud coordinates; determining the target angle between the vehicle and the obstacle corresponding to the obstacle point cloud data set based on the first angle and the second angle. The above step S40 includes: controlling the moving speed of the vehicle according to the target angle and preset obstacle avoidance parameters.

[0113] It should be noted that the above rotation center coordinates may be the coordinates (x0, y0) of the rotation center point O in the vehicle rotation center coordinate system XOY.

[0114] It is understandable that if Figure 5 As shown, the straight-line distance can be the distance r between the obstacle point cloud data P and the rotation center O, and the first angle can be the angle α between the obstacle point cloud data and the Y axis in the vehicle rotation center coordinate system, that is, Figure 5 The second angle can be the angle β between the obstacle projection point cloud data and the Y axis in the vehicle rotation center coordinate system, that is, Figure 5 The angle between the midline OP' and oy, the target angle can be Figure 5 The included angle γ between the straight line OP and the straight line OP', where γ = |α-β|.

[0115] In a specific implementation, the distance r between the obstacle point cloud data and the rotation center point can be calculated based on P(x,y) and O(x0,y0). After the straight-line distance r is calculated, the angle α is calculated based on the straight-line distance r and P(x,y), and the angle β is calculated based on the straight-line distance r and P'(x',y'), so as to obtain the target angle γ. The obtained target angle γ is compared with the preset deceleration angle and the preset obstacle avoidance angle. If the target angle γ is greater than the preset deceleration angle, the vehicle drives normally. If the target angle γ is less than the preset deceleration angle, the vehicle is controlled to decelerate. If the target angle γ is less than the preset obstacle avoidance angle, the vehicle is controlled to stop and avoid the obstacle.

[0116] It should be noted that when the vehicle is in a non-spinning motion state, the above step S304 includes: determining the target distance between the vehicle and the obstacle corresponding to the obstacle point cloud data set based on the straight-line distance between the obstacle point cloud coordinates and the rotation center coordinates and the target angle; the above step S40 includes: controlling the moving speed of the vehicle according to the target distance and preset obstacle avoidance parameters.

[0117] It should be understood that the above target distance may be the distance between the obstacle point cloud data and the preset vehicle contour model, that is, the distance between the obstacle point cloud data and the obstacle projection point cloud data, such as Figure 6 As shown, the target distance may be the distance d between P and P', wherein after the angle γ between OP and OP' and the length r of OP are calculated, the target distance d may be calculated according to the formula d=γ*r. After the target distance d is calculated, the distance d may be compared with the preset deceleration distance and the preset obstacle avoidance distance. If the target distance d is greater than the preset deceleration distance, the vehicle drives normally. If the target distance d is less than the preset deceleration distance, the vehicle is controlled to decelerate. If the target distance d is less than the preset obstacle avoidance distance, the vehicle is controlled to stop and avoid obstacles.

[0118] This embodiment establishes a vehicle rotation center coordinate system according to the vehicle's motion state, and then obtains the target angle based on the coordinate position of the obstacle point cloud data and the obstacle projection point cloud data in the vehicle rotation center coordinate system and the coordinate position of the rotation center, and when the vehicle is in a non-spinning motion state, the target distance is obtained based on the target angle and the coordinate position of the obstacle point cloud data, so that the vehicle's moving speed can be controlled by comparing the obtained target distance or target angle with the preset deceleration parameters or preset obstacle avoidance parameters, thereby ensuring the safe driving of the vehicle.

[0119] In addition, an embodiment of the present invention further proposes a storage medium, on which a vehicle obstacle avoidance program is stored. When the vehicle obstacle avoidance program is executed by a processor, the steps of the vehicle obstacle avoidance method described above are implemented.

[0120] Reference Figure 7 , Figure 7 It is a structural block diagram of the first embodiment of the vehicle obstacle avoidance device of the present invention.

[0121] like Figure 7 As shown, the vehicle obstacle avoidance device proposed in the embodiment of the present invention includes:

[0122] The first data screening module 701 is used to screen the point cloud data collected by the target sensor through a preset vehicle profile model corresponding to the vehicle load state to obtain a target point cloud data set;

[0123] A second data screening module 702 is used to screen the target point cloud data set based on the current vehicle motion state to obtain an obstacle point cloud data set;

[0124] A position relationship determination module 703, configured to determine a spatial position relationship between the vehicle and an obstacle corresponding to the obstacle point cloud data set based on the current vehicle motion state and the obstacle point cloud data set;

[0125] The speed control module 704 is used to control the moving speed of the vehicle according to the spatial position relationship and preset obstacle avoidance parameters.

[0126] The vehicle obstacle avoidance device of the present embodiment discloses filtering the point cloud data collected by the target sensor through a preset vehicle contour model corresponding to the vehicle load state to obtain a target point cloud data set; filtering the target point cloud data set based on the current vehicle motion state to obtain an obstacle point cloud data set; determining the spatial position relationship between the vehicle and the obstacle corresponding to the obstacle point cloud data set based on the current vehicle motion state and the obstacle point cloud data set; and controlling the moving speed of the vehicle according to the spatial position relationship and the preset obstacle avoidance parameters. Compared with the prior art in which the distance between the vehicle and the obstacle is obtained and the obstacle is avoided according to the data in the vehicle sensor, the present embodiment filters the collected point cloud data according to the vehicle contour model to obtain the target point cloud data set, filters the target point cloud data set according to the current vehicle motion state to obtain the obstacle point cloud data set, and then determines the spatial position relationship between the vehicle and the obstacle according to the current vehicle motion state, and controls the moving speed of the vehicle according to the spatial position relationship and the obstacle avoidance parameters, thereby solving the technical problem that the obstacle avoidance distance acquisition method in the prior art cannot accurately obtain the distance between the vehicle and the obstacle, thereby ensuring the driving safety of the vehicle.

[0127] Other embodiments or specific implementations of the vehicle obstacle avoidance device of the present invention can refer to the above-mentioned method embodiments and will not be described in detail here.

[0128] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or system including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or system. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, article or system including the element.

[0129] The serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages or disadvantages of the embodiments.

[0130] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course can also be implemented by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as a read-only memory / random access memory, a magnetic disk, or an optical disk), and includes a number of instructions for a terminal device (which can be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.) to execute the methods described in each embodiment of the present invention.

[0131] The above are only preferred embodiments of the present invention, and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A vehicle obstacle avoidance method, characterized in that: The vehicle obstacle avoidance method comprises: The point cloud data collected by the target sensor is filtered through a preset vehicle contour model corresponding to the vehicle load state to obtain a target point cloud data set; Filtering the target point cloud data set based on the current vehicle motion state to obtain an obstacle point cloud data set; Determine a spatial position relationship between the vehicle and an obstacle corresponding to the obstacle point cloud data set based on the current vehicle motion state and the obstacle point cloud data set; Controlling the moving speed of the vehicle according to the spatial position relationship and preset obstacle avoidance parameters; The step of determining the spatial position relationship between the vehicle and the obstacle corresponding to the obstacle point cloud data set based on the current vehicle motion state and the obstacle point cloud data set comprises: Establishing a vehicle rotation center coordinate system based on the current vehicle motion state; Obtain obstacle projection point cloud data of the obstacle point cloud data set on the preset vehicle contour model; Obtaining the obstacle point cloud coordinates corresponding to the obstacle point cloud data in the vehicle rotation center coordinate system, and the obstacle projection point cloud coordinates corresponding to the obstacle projection point cloud data in the vehicle rotation center coordinate system; Determine a spatial position relationship between the vehicle and an obstacle corresponding to the obstacle point cloud data set based on the obstacle point cloud coordinates and the obstacle projection point cloud coordinates; The current vehicle motion state includes a spinning motion state, and the step of determining the spatial position relationship between the vehicle and the obstacle corresponding to the obstacle point cloud data set based on the obstacle point cloud coordinates and the obstacle projection point cloud coordinates includes: Determining a straight-line distance between the obstacle point cloud coordinates and the rotation center coordinates of the vehicle rotation center coordinate system based on the obstacle point cloud coordinates and the rotation center coordinates of the vehicle rotation center coordinate system; Acquire a first angle between the obstacle point cloud data and the target axis of the vehicle rotation center coordinate system based on the straight-line distance and the obstacle point cloud coordinates; Acquire a second angle between the obstacle projection point cloud data and the target axis of the vehicle rotation center coordinate system based on the straight-line distance and the obstacle projection point cloud coordinates; Determine a target angle between the vehicle and an obstacle corresponding to the obstacle point cloud data set according to the first angle and the second angle; The step of controlling the moving speed of the vehicle according to the spatial position relationship and the preset obstacle avoidance parameters comprises: The moving speed of the vehicle is controlled according to the target angle and preset obstacle avoidance parameters.

2. The vehicle obstacle avoidance method according to claim 1, characterized in that: The step of filtering the point cloud data collected by the target sensor by using the preset vehicle profile model corresponding to the vehicle load state to obtain the target point cloud data set includes: Establishing a vehicle body coordinate system by using a preset vehicle contour model corresponding to a vehicle load state, wherein the vehicle load state includes a loaded state or an unloaded state; Acquire point cloud data collected by a target sensor, wherein the target sensor includes a laser radar and a stereo camera; Establishing a sensor coordinate system based on the target sensor, and acquiring a first coordinate position of the point cloud data in the sensor coordinate system; Performing coordinate transformation on the first coordinate position through a coordinate transformation matrix between the target sensor and the center of the vehicle to obtain a second coordinate position of the point cloud data in the vehicle body coordinate system; Determining whether the second coordinate position is located in a preset area of ​​the vehicle body coordinate system; If so, the point cloud data corresponding to the second coordinate position is screened to obtain a target point cloud data set.

3. The vehicle obstacle avoidance method according to claim 1, characterized in that: The step of screening the target point cloud data set based on the current vehicle motion state to obtain the obstacle point cloud data set includes: Generate a corresponding walking area based on the current vehicle motion state and the preset vehicle contour model; Determining whether the target point cloud data in the target point cloud data set is located in the walking area; If so, the target point cloud data is screened to obtain an obstacle point cloud data set.

4. The vehicle obstacle avoidance method according to claim 1, characterized in that: The current vehicle motion state includes a non-spin motion state, and the step of determining the spatial position relationship between the vehicle and the obstacle corresponding to the obstacle point cloud data set based on the obstacle point cloud coordinates and the obstacle projection point cloud coordinates includes: Determine a target distance between the vehicle and an obstacle corresponding to the obstacle point cloud data set based on a straight-line distance between the obstacle point cloud coordinates and the rotation center coordinates and the target angle; The step of controlling the moving speed of the vehicle according to the spatial position relationship and the preset obstacle avoidance parameters comprises: The moving speed of the vehicle is controlled according to the target distance and preset obstacle avoidance parameters.

5. The vehicle obstacle avoidance method according to any one of claims 1 to 4, characterized in that: The step of filtering the point cloud data collected by the target sensor by using the preset vehicle profile model corresponding to the vehicle load state to obtain the target point cloud data set includes: The point cloud data collected by the target sensor is filtered through a preset vehicle contour model corresponding to the vehicle load state to obtain target point cloud data; The target point cloud data is subjected to a preset filtering process, and a target point cloud data set is obtained based on the filtered target point cloud data.

6. A vehicle obstacle avoidance device, characterized in that: The device comprises: A first data screening module is used to screen the point cloud data collected by the target sensor through a preset vehicle profile model corresponding to the vehicle load state to obtain a target point cloud data set; A second data screening module is used to screen the target point cloud data set based on the current vehicle motion state to obtain an obstacle point cloud data set; a position relationship determination module, configured to determine a spatial position relationship between the vehicle and an obstacle corresponding to the obstacle point cloud data set based on the current vehicle motion state and the obstacle point cloud data set; A speed control module, used to control the moving speed of the vehicle according to the spatial position relationship and preset obstacle avoidance parameters; The position relationship determination module is further used to establish a vehicle rotation center coordinate system based on the current vehicle motion state; obtain obstacle projection point cloud data of the obstacle point cloud data set in the preset vehicle contour model; obtain obstacle point cloud coordinates corresponding to the obstacle point cloud data in the vehicle rotation center coordinate system, and obstacle projection point cloud coordinates corresponding to the obstacle projection point cloud data in the vehicle rotation center coordinate system; determine the spatial position relationship between the vehicle and the obstacle corresponding to the obstacle point cloud data set based on the obstacle point cloud coordinates and the obstacle projection point cloud coordinates; The current vehicle motion state includes a spinning motion state, and the position relationship determination module is further used to determine the straight-line distance between the obstacle point cloud coordinates and the rotation center coordinates of the vehicle rotation center coordinate system based on the obstacle point cloud coordinates and the rotation center coordinates of the vehicle rotation center coordinate system; obtain a first angle between the obstacle point cloud data and a target axis of the vehicle rotation center coordinate system based on the straight-line distance and the obstacle point cloud coordinates; obtain a second angle between the obstacle projection point cloud data and a target axis of the vehicle rotation center coordinate system based on the straight-line distance and the obstacle projection point cloud coordinates; and determine a target angle between the vehicle and an obstacle corresponding to the obstacle point cloud data set according to the first angle and the second angle; The speed control module is also used to control the moving speed of the vehicle according to the target angle and preset obstacle avoidance parameters.

7. A vehicle obstacle avoidance device, characterized in that: The device comprises: a memory, a processor, and a vehicle obstacle avoidance program stored in the memory and executable on the processor, wherein the vehicle obstacle avoidance configuration is configured to implement the steps of the vehicle obstacle avoidance method according to any one of claims 1 to 5.

8. A storage medium, characterized in that: The storage medium stores a vehicle obstacle avoidance program, which, when executed by the processor, implements the steps of the vehicle obstacle avoidance method according to any one of claims 1 to 5.

Images