Vehicle driving obstacle avoidance method and device

Abstract

The embodiment of the invention discloses a vehicle driving obstacle avoidance method which is used for improving the accuracy of vehicle driving obstacle avoidance. The method comprises the steps that a first obstacle map is obtained, the first obstacle map is used for predicting the position of a road surface obstacle area for a vehicle at a far sight distance, and the far sight distance comprises the distance with the vehicle larger than a first threshold value; when the distance between the vehicle and the road surface obstacle area is smaller than or equal to a first threshold value, first road surface information is collected, and existence of the road surface obstacle area is detected according to the first road surface information. When the distance between the vehicle and the road surface obstacle area is smaller than or equal to a second threshold value, second road surface information is collected, a real-time obstacle map is generated according to the second road surface information, and the second threshold value is smaller than the first threshold value; and obtaining a second obstacle map, and determining a driving path of the vehicle based on the second obstacle map, the second obstacle map being an obstacle map generated by the cloud side device based on the real-time obstacle map.

Description

Technical Field

[0001] The embodiments of the present application relate to the field of vehicles, and in particular to a method and device for avoiding obstacles while driving a vehicle. Background Art

[0002] Unmanned driving technology is a comprehensive technology based on artificial intelligence. With the development of unmanned driving technology, unmanned driving technology has been applied in many scenarios, among which open-pit mines are one of the first scenarios where unmanned driving technology has been implemented. In the unmanned driving scenario of open-pit mines, due to the heavy load of vehicles, the road surface on which the vehicles travel is mainly soft rock environments such as mudstone and silt conglomerate, which are prone to potholes. In addition, water is easily accumulated during the rainy season, and potholes on the road sections can easily cause vehicles to overturn. Therefore, how unmanned vehicles can safely and efficiently pass through potholes has become a pain point in open-pit mines.

[0003] In the current vehicle driving obstacle avoidance solution, the vehicle collects the road surface bump information through sensors and generates a bump map of the road surface. When the vehicle drives on the bumpy road surface again, the vehicle can plan a path based on the bump map, thereby slowing down or detouring when passing through the bumpy road obstacle area. However, in the open-pit mining scenario, due to the soft rock environment and the relatively soft road surface, the bumpy road obstacle area of ​​the road surface will change with the driving of the vehicle and the changes in the weather, resulting in poor accuracy of the vehicle's path planning based on the bump map for obstacle avoidance. Summary of the invention

[0004] The embodiment of the present application provides a vehicle driving obstacle avoidance method, in which the vehicle can collect road surface information of the road obstacle area in real time and perform path planning when approaching the road obstacle area, thereby improving the accuracy of vehicle driving obstacle avoidance. The embodiment of the present application also provides electronic equipment, vehicles, computer-readable storage media and computer program products corresponding to the vehicle driving obstacle avoidance method.

[0005] In a first aspect, an embodiment of the present application provides a vehicle driving obstacle avoidance method, which can be executed by a vehicle driving obstacle avoidance device, or by a component of the vehicle driving obstacle avoidance device, such as a processor, a chip or a chip system, or can be implemented by a logic module or software that can realize all or part of the functions of the vehicle driving obstacle avoidance device. The method provided in the first aspect includes: the vehicle driving obstacle avoidance device obtains a first obstacle map, and the first obstacle map is used to predict the position of a road obstacle area for the vehicle at a long sight distance, and the long sight distance includes a distance from the vehicle greater than a first threshold. When the distance between the vehicle and the road obstacle area is less than or equal to the first threshold, the vehicle driving obstacle avoidance device collects first road information and detects the existence of the road obstacle area based on the first road information. When the distance between the vehicle and the road obstacle area is less than or equal to the second threshold, the vehicle driving obstacle avoidance device collects second road information and generates a real-time obstacle map based on the second road information, and the second threshold is less than the first threshold. The vehicle driving obstacle avoidance device obtains the second obstacle map, and determines the driving path of the vehicle based on the second obstacle map, and the second obstacle map is an obstacle map generated by the cloud-side device based on the real-time obstacle map.

[0006] The vehicle obstacle avoidance device in the embodiment of the present application can re-collect real-time road surface information of the road obstacle area when the vehicle approaches the road obstacle area, and update the obstacle map based on the real-time road surface information, thereby avoiding obstacle avoidance failure caused by changes in the road obstacle area. Compared with the current method of performing vehicle obstacle avoidance based on a fixed obstacle map, the vehicle obstacle avoidance method provided in the embodiment of the present application improves the accuracy of vehicle obstacle avoidance.

[0007] In some possible implementations, when the vehicle determines that the road obstacle area exists based on the first road information, the vehicle is controlled to decelerate, or the vehicle's driving path is adjusted. The distance from the vehicle between the first threshold and the second threshold can be called the middle sight distance, that is, when the vehicle's driving obstacle avoidance device enters the vehicle's middle sight distance in the road obstacle area, the vehicle's driving obstacle avoidance device can determine whether the road obstacle area exists based on the first road information collected by the vehicle, and control the vehicle to decelerate in advance or control the vehicle to adjust the driving path.

[0008] In the embodiment of the present application, the vehicle obstacle avoidance device can determine the distance between the vehicle and the road obstacle area based on the first obstacle map at a long sight distance, and control the vehicle to slow down before the vehicle approaches the road obstacle area, so that the vehicle can prepare for re-collecting real-time road information at a close sight distance, thereby improving the accuracy of road information collection and improving the safety of vehicle driving.

[0009] In some possible implementations, after the vehicle obstacle avoidance device generates a real-time obstacle map according to the second road surface information, the real-time obstacle map is sent to the cloud-side device. The vehicle obstacle avoidance device receives the second obstacle map sent by the cloud-side device, where the second obstacle map is an obstacle map generated by the cloud-side device based on the fusion of the real-time obstacle map and the crowdsourced obstacle map, and the crowdsourced obstacle map is an obstacle map generated by different vehicles.

[0010] In the embodiment of the present application, the cloud-side device generates a second obstacle map based on the fusion of the real-time obstacle map and the crowdsourced obstacle map, thereby improving the accuracy of the second obstacle map.

[0011] In some possible implementations, when the vehicle obstacle avoidance device acquires the first obstacle map, the vehicle obstacle avoidance device performs clustering operations on the road information collected by the vehicle based on a clustering algorithm to identify road obstacle areas, which include raised areas and sunken areas. The vehicle obstacle avoidance device marks the road obstacle areas in the map to obtain the first obstacle map. Furthermore, the vehicle obstacle avoidance device can also identify the obstacle types of the road obstacle areas based on the clustering algorithm, such as falling rocks, ruts, and potholes.

[0012] In the implementation of the present application, the vehicle driving device can accurately determine the distance between the vehicle and the road obstacle area based on the position of the road obstacle area marked in the obstacle map, thereby improving the accuracy of the vehicle's driving and obstacle avoidance.

[0013] In some possible implementations, the vehicle obstacle avoidance device calculates risk values ​​for different areas of the road based on road surface information, and the risk values ​​are used to indicate the difficulty of driving in different areas of the road. The vehicle obstacle avoidance device provides a display interface of a first obstacle map, and the display interface is used to display the risk values ​​of different areas of the road.

[0014] In the embodiment of the present application, the vehicle obstacle avoidance device can provide a display interface of the obstacle map, and display the risk values ​​of different road obstacle areas in the obstacle map through the display interface, thereby improving the visibility of the obstacle map.

[0015] In some possible implementations, the vehicle obstacle avoidance device converts the coordinate system of the first obstacle map into the vehicle's driving coordinate system, and the vehicle's driving coordinate system is used to calculate the distance between the vehicle and the road obstacle area. The vehicle obstacle avoidance device annotates the laser point cloud acquired by the vehicle sensor in the driving coordinate system based on the converted first obstacle map.

[0016] In the embodiment of the present application, the vehicle obstacle avoidance device obtains the annotation data of the road obstacle area based on the vehicle sensor, and uses the annotation data to correct the posture of the road obstacle area, thereby improving the accuracy of the road obstacle area in the obstacle map.

[0017] In some possible implementations, after the vehicle obstacle avoidance device establishes the first obstacle map, the vehicle obstacle avoidance device can operate by grouping the vehicles. The vehicle obstacle avoidance device updates and generates a second obstacle map based on the first road surface information collected by the first vehicle in the group, and plans the driving path of the vehicle group based on the second obstacle map.

[0018] In the embodiment of the present application, the vehicle driving obstacle avoidance device can plan the path for the vehicle grouping based on the second obstacle map. Since the vehicles are grouped, the vehicle driving obstacle avoidance device only needs to configure laser radar and other sensors on the first vehicle to collect road surface information, thereby reducing the configuration of vehicle sensors and reducing the cost of vehicle driving obstacle avoidance.

[0019] In some possible implementations, when the vehicle obstacle avoidance device determines the vehicle's driving path based on the second obstacle map, the vehicle obstacle avoidance device generates a waypoint of the vehicle according to the size of the road obstacle area in the second obstacle map and the wheel spacing of the vehicle, and the waypoint is used to indicate the coordinates through which the center of the vehicle passes. The vehicle obstacle avoidance device determines the vehicle's driving route based on the waypoint.

[0020] In the embodiment of the present application, the vehicle driving obstacle avoidance device can plan the vehicle's driving path according to the size of the road obstacle area in the obstacle map, thereby improving the accuracy of the vehicle driving obstacle avoidance.

[0021] In some possible implementations, when the wheel spacing can cover the size of the road obstacle area, the waypoint of the vehicle is controlled to be the center of the road obstacle area.

[0022] The vehicle obstacle avoidance device in the embodiment of the present application can control the vehicle to cross the road obstacle area, thereby improving the accuracy of the vehicle's obstacle avoidance.

[0023] In a second aspect, an embodiment of the present application provides a vehicle obstacle avoidance device, including an acquisition unit and a processing unit. The acquisition unit is used to acquire a first obstacle map, and the first obstacle map is used to predict the position of a road obstacle area for the vehicle at a long sight distance, and the long sight distance includes a distance from the vehicle greater than a first threshold. The processing unit is used to collect first road information when the distance between the vehicle and the road obstacle area is less than or equal to the first threshold, and detect the existence of the road obstacle area based on the first road information. The processing unit is also used to collect second road information when the distance between the vehicle and the road obstacle area is less than or equal to a second threshold, and generate a real-time obstacle map based on the second road information, and the second threshold is less than the first threshold. The acquisition unit is also used to acquire a second obstacle map. The processing unit is used to determine the driving path of the vehicle based on the second obstacle map, and the second obstacle map is an obstacle map generated by the cloud-side device based on the real-time obstacle map.

[0024] In some possible implementations, the processing unit is further configured to control the vehicle to decelerate, or adjust the vehicle's driving path, when the vehicle determines that a road obstacle area exists based on the first road information.

[0025] In some possible implementations, the acquisition unit is further used to send a real-time obstacle map to the cloud-side device. The acquisition unit is specifically used to receive a second obstacle map sent by the cloud-side device, where the second obstacle map is an obstacle map generated by the cloud-side device based on a fusion of the real-time obstacle map and the crowdsourced obstacle map, where the crowdsourced obstacle map is an obstacle map generated by different vehicles.

[0026] In some possible implementations, the processing unit is further used to perform clustering operations on the road information collected by the vehicle based on a clustering algorithm to identify road obstacle areas, which include raised areas and sunken areas. The road obstacle areas are marked in the map to obtain a first obstacle map.

[0027] In some possible implementations, the processing unit is further configured to calculate risk values ​​of different road areas based on the road surface information, and the risk values ​​are used to indicate the driving difficulty of different road areas. The processing unit is further configured to provide a display interface of the first obstacle map, and the display interface is used to display the risk values ​​of different road areas.

[0028] In some possible implementations, the processing unit is further configured to convert the coordinate system of the first obstacle map into a driving coordinate system of the vehicle, and annotate the laser point cloud acquired by the vehicle sensor in the driving coordinate system based on the converted first obstacle map.

[0029] In some possible implementations, the processing unit is specifically configured to generate a waypoint of the vehicle according to the size of the road obstacle area in the second obstacle map and the wheel spacing of the vehicle, the waypoint being used to indicate the coordinates through which the center of the vehicle passes, and determine the driving route of the vehicle based on the waypoint.

[0030] In some possible implementations, the processing unit is further configured to control the waypoint of the vehicle to be the center of the road obstacle area when the wheel spacing can cover the size of the road obstacle area.

[0031] In a third aspect, an embodiment of the present application provides a vehicle-mounted device, including a processor, which is coupled to a memory, and the processor is used to execute instructions in the memory. When the instructions are executed by the processor, the vehicle-mounted device executes the method described in the first aspect or any possible implementation method of the first aspect.

[0032] In a fourth aspect, an embodiment of the present application provides a vehicle, which is used to execute the method described in the first aspect or any possible implementation method of the first aspect.

[0033] In a fifth aspect, an embodiment of the present application provides a vehicle obstacle avoidance system, which includes a vehicle and a cloud-side device, and the vehicle is used to execute the method described in the first aspect or any possible implementation method of the first aspect.

[0034] In a sixth aspect, an embodiment of the present application provides a computer-readable storage medium having instructions stored thereon. When the instructions are executed, the computer executes the method described in the first aspect or any possible implementation manner of the first aspect.

[0035] In a seventh aspect, an embodiment of the present application provides a computer program product, which includes instructions. When the instructions are executed, the computer implements the method described in the first aspect or any possible implementation method of the first aspect.

[0036] It can be understood that the beneficial effects that can be achieved by any of the vehicle obstacle avoidance devices, on-board equipment, vehicles, vehicle obstacle avoidance systems, computer-readable media or computer program products provided above can be referred to the beneficial effects in the corresponding methods and will not be repeated here. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 A schematic diagram of the system architecture of a vehicle obstacle avoidance system provided in an embodiment of the present application;

[0038] Figure 2 A schematic diagram of a process flow of a vehicle obstacle avoidance method provided in an embodiment of the present application;

[0039] Figure 3 A schematic diagram of a flow chart of another vehicle obstacle avoidance method provided in an embodiment of the present application;

[0040] Figure 4 A schematic diagram of generating an obstacle map provided in an embodiment of the present application;

[0041] Figure 5 A schematic diagram of laser point cloud annotation provided in an embodiment of the present application;

[0042] Figure 6 A schematic diagram of updating an obstacle map provided in an embodiment of the present application;

[0043] Figure 7 A schematic diagram of a vehicle driving path planning provided in an embodiment of the present application;

[0044] Figure 8 A schematic diagram of a vehicle avoiding obstacles provided in an embodiment of the present application;

[0045] Fig. 9 A schematic diagram of the structure of a vehicle obstacle avoidance device provided in an embodiment of the present application;

[0046] Fig.10 A schematic diagram of the structure of a vehicle-mounted device provided in an embodiment of the present application;

[0047] Fig.11 A schematic diagram of the structure of a vehicle provided in an embodiment of the present application. DETAILED DESCRIPTION

[0048] The embodiment of the present application provides a vehicle driving obstacle avoidance method for improving the accuracy of vehicle driving obstacle avoidance.

[0049] The terms "first", "second", "third", "fourth", etc. (if any) in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0050] In the embodiments of the present application, words such as "exemplary" or "for example" are used to indicate examples, illustrations or descriptions. Any embodiment or design described as "exemplary" or "for example" in the embodiments of the present application should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "exemplary" or "for example" is intended to present related concepts in a specific way.

[0051] First, some terms involved in the embodiments of the present application are introduced to facilitate understanding by those skilled in the art.

[0052] An inertial measurement unit (IMU) is a device that integrates multiple sensors. By measuring and fusing data from multiple sensors, it provides accurate motion status and attitude information for the vehicle, supporting the vehicle's control, navigation, and positioning functions. The multiple sensors include a three-axis accelerometer, a three-axis gyroscope, and a three-axis magnetometer. The IMU processes and fuses the measurement data from the three-axis accelerometer, three-axis gyroscope, and three-axis magnetometer to provide information such as the vehicle's acceleration, angular velocity, and attitude change.

[0053] Clustering algorithm is an unsupervised learning method used to group data objects into similar groups or clusters. Clustering algorithm calculates the similarity between data objects and classifies similar objects into the same cluster, while classifying dissimilar objects into different clusters.

[0054] The following is an illustrative description of the application scenarios of the vehicle obstacle avoidance method provided in the embodiments of the present application.

[0055] The embodiments of the present application are applied to unmanned driving scenarios in open-pit mines. The open-pit mine scenario has high requirements for the safe and efficient passage of unmanned mine trucks through the road. Potholes are prone to occur when mine trucks pass through soft roads. The road conditions are complex, and open-pit mines have a high frequency of rain and snow, which can easily lead to vehicle overturning and production suspension. In the current unmanned driving solutions for open-pit mines, the cloud-side equipment can collect road surface bump information based on the vehicle's sensor to generate a road bump map. The cloud server plans the vehicle's driving path based on the bump map, thereby slowing down or detouring when passing through bumpy road obstacle areas. However, in the open-pit mine scenario, the vehicle is heavily loaded and the unpaved road surface is relatively soft. The road surface is prone to sinking and forming potholes, which makes the road obstacle area of ​​the road surface change frequently as the vehicle travels, resulting in poor accuracy in obstacle avoidance based on obstacle map planning.

[0056] In view of this, the implementation of the present application provides a vehicle obstacle avoidance method. The cloud-side device of the embodiment of the present application can establish an obstacle map based on the road surface information collected by the vehicle, and plan the vehicle's driving path based on the obstacle map. At the same time, the vehicle can predict the obstacle area at a long sight distance based on the obstacle map, achieve long-distance deceleration, and re-collect road surface information in real time at a close sight distance. The cloud-side device updates the obstacle map based on the newly collected real-time road surface information, and adjusts the vehicle's driving trajectory according to the updated map, thereby improving the accuracy of the vehicle passing through road obstacle areas such as potholes.

[0057] In order to make the technical solution of the present application clearer and easier to understand, the system architecture of the present application is introduced below with reference to the accompanying drawings.

[0058] See also Figure 1 , Figure 1 A schematic diagram of the system architecture of the vehicle obstacle avoidance system provided in the embodiment of the present application. Figure 1 In the example shown, the vehicle obstacle avoidance system 10 includes a vehicle end 100, a cloud device 200 and a client 300. The vehicle end 100 includes a radar module 101, a positioning module 102, a communication module 103 and an inertial navigation module 104. The specific functions of each part of the vehicle obstacle avoidance system 10 are described in detail below.

[0059] The vehicle side 100 is used to collect road surface information of the vehicle driving on the road, and generate an obstacle map based on the collected road surface information. The vehicle side 100 is also used to receive the obstacle map sent by the cloud side device 200. The vehicle side 101 is also used to plan the vehicle's driving path according to the obstacle map.

[0060] The vehicle end 100 includes a radar module 101, a positioning module 102, a communication module 103 and an inertial navigation module 104. The radar module 101 is used to obtain road surface information, which includes point cloud data of the road surface. The radar module 101 is, for example, a laser radar and a millimeter wave radar. The laser radar is a high-precision sensor of the vehicle end 100, which acquires three-dimensional point cloud data of the vehicle's surrounding environment in real time by emitting a laser beam and receiving its reflection.

[0061] The positioning module 102 is used to obtain the position of the vehicle in the map. The positioning module 102 integrates multiple technologies to achieve accurate positioning of the vehicle. Specifically, the positioning module 102 can calculate the precise position, speed and attitude information of the vehicle by receiving the global positioning system GPS satellite signal and using the inertial measurement unit IMU.

[0062] The communication module 103 is used for the vehicle end 100 to communicate with the cloud side device 200. Specifically, the communication module 103 uploads the obstacle map generated by the vehicle end 100 to the cloud side device 200, and receives the updated obstacle map sent by the cloud side device 200.

[0063] The inertial navigation module 104 is used to measure the body position of the vehicle 100. Specifically, the inertial navigation module 104 can calculate the position, speed and attitude information of the vehicle 100 through the inertial measurement unit IMU, and provide this information to the positioning module 102 for accurate positioning of the vehicle.

[0064] The cloud-side device 200 is used to receive obstacle maps sent by one or more vehicle terminals 100, and fuse the obstacle maps sent by different vehicle terminals to generate an updated obstacle map.

[0065] The client 300 is used to provide the user with a visual interactive interface for interacting with the cloud-side device 200. The user can view the obstacle map generated by the cloud-side device 200 and the driving status of the vehicle 100 through the client 300, and the user can also control the vehicle 100 in real time through the client 300.

[0066] It is understandable that the client 300 can be software or an application deployed on the vehicle side 100, the cloud side device 200 or other terminal devices, without specific limitation.

[0067] based on Figure 1The vehicle driving obstacle avoidance system 10 shown in the figure, the present application also provides a vehicle driving obstacle avoidance method. The vehicle driving obstacle avoidance method provided by the present application embodiment is introduced below in conjunction with an embodiment.

[0068] See also Figure 2 , Figure 2 A flow chart of a vehicle obstacle avoidance method provided in an embodiment of the present application. Figure 2 In the example shown, the method includes the following steps:

[0069] S201. The vehicle end 100 obtains a first obstacle map, where the first obstacle map is used to predict the location of a road obstacle area for the vehicle at a long viewing distance.

[0070] The vehicle end 100 first collects road surface information, which is used to indicate the height difference at different positions of the road surface. The road surface information includes point cloud data at different positions of the road surface. In the process of collecting road surface information, the vehicle end 100 scans the road surface on which the vehicle is traveling through a laser radar, thereby obtaining a laser point cloud of the road surface on which the vehicle is traveling. At the same time, the vehicle end 100 collects the vehicle's posture information through satellite navigation real-time dynamic positioning and inertial measurement unit IMU. The laser point cloud and posture information are both used to analyze the height difference of concave and convex areas at different positions on the road surface.

[0071] See also Figure 3 , Figure 3 A flow chart of another vehicle obstacle avoidance method provided in an embodiment of the present application. Figure 3 In the example shown, the vehicle end 100 collects the vehicle's posture information based on satellite navigation real-time dynamic positioning and inertial measurement unit IMU, wherein the posture information includes the coordinate position of the vehicle end 100 in the map and the vehicle's heading angle (yaw).

[0072] exist Figure 3 In the example shown, the vehicle end 100 collects point cloud data of the road surface based on the laser radar. For example, when an unmanned mining vehicle is driving in an open-pit mine area, the vehicle continuously emits laser beams through the laser radar and records the distance and intensity of each laser beam to the surrounding environment. When the laser beam shines on the potholes on the road surface, the laser radar detects the reflected light and measures the distance to the object, converting these measurement data into point cloud data.

[0073] After the vehicle end 100 collects the road surface information, a first obstacle map is generated based on the collected road surface information. The first obstacle map can display the location of the road obstacle area in the map, so that the vehicle can avoid the road obstacle area based on the first obstacle map. The following specifically describes the process of the vehicle end 100 generating the first obstacle map of the road based on the road surface information collected by the vehicle.

[0074] After the vehicle end 100 collects road information, it performs clustering operation on the road information collected by the vehicle based on a clustering algorithm to identify road obstacle areas, which include raised areas and sunken areas. The vehicle end 100 marks the road obstacle areas in the map to obtain a first obstacle map.

[0075] Specifically, the vehicle end 100 projects the collected laser point cloud into a horizontal two-dimensional map, rasterizes the two-dimensional map, and marks the height of each collection point in the grid of the horizontal two-dimensional map. The vehicle end 100 performs clustering operations on the collection points in the horizontal two-dimensional map to identify the raised areas and recessed areas of the road surface. Furthermore, the vehicle end 100 can also identify the obstacle types of the road obstacle area based on the clustering algorithm, such as fallen rocks, rutting, and potholes. For example, when the collection point is raised and the grid occupancy is small, it is identified as fallen rocks. When the collection point is raised and the grid occupancy is distributed along a straight line, it is identified as rutting. When the collection point is raised and recessed at the same time, it is identified as a pothole.

[0076] The vehicle end 100 uses a clustering algorithm to analyze and calculate the laser point cloud data, and groups the point cloud points into different clusters. Each cluster has similar features or attributes, which can be position, shape, size, etc. In short, the goal of the clustering algorithm is to aggregate similar points together and separate different points, so as to better understand and analyze the point cloud data and identify the features of the point cloud data.

[0077] See also Figure 4 , Figure 4 A schematic diagram of a cloud-side device establishing a first obstacle map provided in an embodiment of the present application. In the example shown in Figure 4, after the vehicle end 100 obtains the vehicle's posture information and laser point cloud, the vehicle end 100 analyzes the laser point clouds at different locations and clusters them to generate a first obstacle map.

[0078] For example, the vehicle end 100 is clustered according to the height of the collection point, where Figure 4 The dark-colored collection points are collection points above the road surface, i.e., the raised areas of the road surface, and the light-colored collection points are collection points below the road surface, i.e., the sunken areas of the road surface. The vehicle end 100 marks these dark-colored collection points and light-colored collection points in the two-dimensional horizontal map to generate a first obstacle map.

[0079] In one possible implementation, risk values ​​of different road surface areas are calculated based on road surface information, and the risk values ​​are used to indicate the difficulty of driving in different road surface areas. The vehicle end 100 can also provide a display interface of the first obstacle map through the client 300, and the display interface is used to display the risk values ​​of different road surface areas.

[0080] Specifically, in the process of calculating the risk values ​​of different areas of the road surface based on the road surface information, the vehicle end 100 calculates the slope and roughness of the collection point based on the posture information and the laser point cloud computing, and determines the risk value of the collection point based on the slope and roughness of the collection point. Further, after the vehicle end 100 calculates the risk values ​​of different collection points, the risk values ​​of different collection points can be marked in the first obstacle map, and the first obstacle map marked with the risk values ​​can be displayed through the display interface.

[0081] The above slope θ satisfies the following formula:

[0082] θ=cos -1 (n1*ng)

[0083] Among them, θ is the angle between n1 and ng, n1 is the normal vector of the slope on which the vehicle is traveling, n1 = (x, y, z), and ng is the direction of gravity, that is, ng = (0, 0, 1).

[0084] The calculation of the above roughness satisfies the following formula:

[0085]

[0086] Among them, hi is the height of the i-th sampling point, and h is the average height of the sampling points.

[0087] In the embodiment of the present application, the slope and roughness of the collection point are positively correlated with the risk value. The greater the slope or roughness, the greater the risk value corresponding to the area in the first obstacle map. Figure 4 ,exist Figure 4 In the example shown, the vehicle end 100 uses different colors to mark the risk values ​​of different areas of the first obstacle map. For example, the risk value of the dark area in the first obstacle map is higher, and the risk value of the light area is lower. The user can view the risk values ​​of different areas in the first obstacle map through the client 300.

[0088] It should be noted that the process of calculating the risk value of the first obstacle map area can also be performed in the cloud-side device 200. After the vehicle end 100 generates the first obstacle map, the first obstacle map is sent to the cloud-side device 100. The cloud-side device 200 calculates the risk values ​​of different areas of the road surface based on the road surface information.

[0089] In one possible implementation, after the vehicle side 100 generates the first obstacle map, the vehicle side 100 can send the first obstacle map to the cloud side device 200. The cloud side device 200 updates the first obstacle map according to the obstacle maps collected by different vehicles, and sends the updated first obstacle map to the vehicle side 100, so that the vehicle side 100 can predict the distance between the vehicle and the road obstacle area at a long sight distance according to the updated first obstacle map, and plan the driving path based on the updated first obstacle map.

[0090] In a possible implementation, the cloud-side device 200 converts the coordinate system of the first obstacle map to the vehicle's driving coordinate system. The cloud-side device 200 annotates the laser point cloud collected by the vehicle 100 based on the converted first obstacle map to obtain annotated data, which is used to annotate the road obstacle area corresponding to the laser point cloud. For example, the cloud-side device 200 annotates different areas of the laser point cloud as fallen rocks, ruts, potholes, etc. based on the converted first obstacle map. The coordinate matrix T of the road obstacle area in the vehicle's driving coordinate system is C Satisfies the following formula:

[0091]

[0092] Among them, T m is the vehicle coordinate matrix, T p is the coordinate matrix of the road obstacle area.

[0093] See also Figure 5 , Figure 5 A schematic diagram of a laser point cloud provided in an embodiment of the present application. Figure 5 In the example shown, the vehicle end 100 annotates the laser point cloud collected by the vehicle, wherein the content annotated in the annotation box in the laser point cloud is a pothole area. In the process of determining the boundary of the annotation box, the vehicle end 100 extracts the point cloud data of the area, selects the point P1 with the lowest height as the reference point, calculates the boundary of the annotation box with the reference point as the origin, fits the x-direction straight line based on the reference point P1, and takes the center point of the straight line to generate a vertical vector, and calculates the nearest point in the x-direction as the boundary P2, determines the length of the annotation box, and based on the line connecting P1 and P2, evenly samples 5 equal parts to calculate the nearest boundary point, and determines the width of the annotation box.

[0094] It should be noted that the vehicle end 100 can use a variety of methods to determine the boundaries of the annotation box of the laser point cloud, which are not specifically limited. A variety of annotation box boundary calculation methods, for example, minimum bounding box, least squares method, curve fitting and contour extraction methods.

[0095] S202. When the distance between the vehicle and the road obstacle area is less than or equal to a first threshold, the vehicle end 100 collects first road surface information and detects the existence of the obstacle area based on the first road surface information.

[0096] After the vehicle end 100 obtains the first obstacle map, the vehicle end 100 drives on the road based on the first obstacle map. Specifically, the vehicle end 100 can plan the vehicle's path based on the first obstacle map, or can plan the vehicle's driving path based on the updated first obstacle map sent by the cloud side device 200.

[0097] In a possible implementation, the vehicle end 100 may also receive a vehicle control instruction generated by the cloud-side device 200 based on the first obstacle map, and control the driving path of the vehicle according to the vehicle control instruction.

[0098] When the distance between the vehicle and the road obstacle area is less than or equal to a first threshold, the vehicle end 100 collects first road information and detects the existence of the obstacle area based on the first road information, wherein the first road information includes one or more of the following: posture information, laser point cloud or visual image.

[0099] In the embodiment of the present application, the distance from the vehicle end 100 that is greater than the first threshold is called the far sight distance, the distance from the vehicle end 100 that is greater than the second threshold and less than or equal to the first threshold is called the medium sight distance, and the distance from the vehicle end 100 that is less than the second threshold is called the near sight distance. The first threshold is greater than the second threshold, for example, the first threshold is 50 meters and the second threshold is 25 meters.

[0100] Among them, neither lidar nor camera can collect accurate road surface information at long sight distance. The vehicle's obstacle avoidance device predicts the location of road obstacle areas based on the obstacle map. At medium sight distance, the camera can collect accurate road surface information, while lidar cannot. At near sight distance, both camera and lidar can collect accurate road surface information.

[0101] In a possible implementation, after the vehicle end 100 collects the first road surface information, when the vehicle end 100 determines that a road obstacle area exists based on the first road surface information, the vehicle is controlled to slow down, or the vehicle is controlled to adjust the driving path of the vehicle.

[0102] In one possible implementation, after the vehicle end 100 collects the first road surface information, it sends the first road surface information to the cloud-side device 200. After the cloud-side device 200 receives the first road surface information sent by the vehicle end 100, when the cloud-side device 200 determines that a road obstacle area exists based on the first road surface information, it sends a deceleration instruction to the vehicle end 100. The vehicle end 100 controls the vehicle to decelerate according to the deceleration instruction sent by the cloud-side device 200.

[0103] Alternatively, when the cloud-side device 200 determines that a road obstacle area exists based on the first road surface information, it sends a vehicle driving path adjustment instruction to the vehicle end 100 , and the vehicle end 100 controls the vehicle's driving path according to the vehicle driving path adjustment instruction sent by the cloud-side device 200 .

[0104] S203. When the distance between the vehicle and the road obstacle area is less than or equal to the second threshold, the vehicle end 100 collects second road information and generates a real-time obstacle map according to the second road information.

[0105] When the distance between the vehicle and the road obstacle area is less than or equal to the second threshold, the vehicle end 100 collects the second road information and generates a real-time obstacle map based on the second road information. Specifically, the second road information includes the vehicle's posture information and the laser point cloud corresponding to the posture information.

[0106] The vehicle end 100 regenerates the real-time obstacle map based on the second road surface information, and the second road surface information is used to determine the real-time height difference of the road obstacle area. The process of the vehicle end 100 regenerating the real-time obstacle map based on the second road surface information is similar to the process of the vehicle end 100 generating the first obstacle map in step 201 above, and will not be described in detail.

[0107] Please continue reading Figure 3 ,exist Figure 3 In the example shown, when the vehicle 100 is driving based on the first obstacle map, the vehicle 100 performs long-distance detection based on the first obstacle map, that is, the distance between the vehicle 100 and the road obstacle area in the first obstacle map is determined according to the first obstacle map. When the distance is less than 50m, the cloud-side device 200 controls the vehicle 100 to slow down. When the distance is less than 25m, the vehicle 100 performs close-range detection of the road obstacle area, that is, the vehicle 100 re-collects the second road surface information, and the vehicle 100 generates a real-time obstacle map based on the second road surface information, and sends the real-time obstacle map to the cloud-side device 200.

[0108] S204. The vehicle side 100 sends a real-time obstacle map to the cloud side device 200.

[0109] The vehicle side 100 sends a real-time obstacle map to the cloud side device 200 .

[0110] S205. The cloud-side device 200 fuses the real-time obstacle map and the crowdsourced obstacle map to obtain a second obstacle map.

[0111] The cloud-side device 200 fuses the real-time obstacle map and the crowdsourced obstacle map to obtain a second obstacle map. Specifically, after receiving the real-time obstacle map sent by the vehicle end 100, the cloud-side device 103 fuses the crowdsourced obstacle map that meets the fusion conditions with the regenerated real-time obstacle map to obtain a second obstacle map. The fusion conditions include that the crowdsourced obstacle map and the real-time obstacle map have the same data format, the crowdsourced obstacle map and the real-time obstacle map have the same collection status, or the crowdsourced obstacle map and the real-time obstacle map have the same collection coordinates.

[0112] See also Figure 6 , Figure 6 A schematic diagram of a process for fusing an obstacle map provided in an embodiment of the present application. Figure 6 In the example shown, after the cloud-side device 200 receives the real-time obstacle map sent by the vehicle end 100, the cloud-side device 200 obtains the crowdsourced obstacle map stored in the object storage (OBS) bucket. The crowdsourced obstacle map is an obstacle map generated by the cloud-side device 200 based on road information collected by multiple vehicles.

[0113] exist Figure 6 In the example shown, when the crowdsourced obstacle map obtained by the cloud-side device 200 meets the fusion condition, the cloud-side device 200 fuses the crowdsourced obstacle map with the real-time obstacle map to obtain a second obstacle map. The cloud-side device 200 sends the second obstacle map to the vehicle end 100. The cloud-side device 200 fuses the crowdsourced obstacle map with the real-time obstacle map to calculate the following formula:

[0114]

[0115]

[0116]

[0117] Among them, α 2 represents a high degree of uncertainty, s and p represent the crowdsourced obstacle map and the real-time obstacle map, respectively. s The Jacobian matrix, J, represents the crowdsourced obstacle map p The Jacobian matrix representing the real-time obstacle map, v z and denote the mean and covariance of the heights respectively.

[0118] In the embodiment of the present application, the cloud side device 200 can receive obstacle maps sent by one or more vehicle terminals 100, thereby aggregating obstacle maps sent by different vehicles to achieve confidence updates of road obstacle areas. For example, when the road information of a certain vehicle misdetects a road obstacle area due to environmental influences such as dust, the cloud side gradually reduces the confidence of the misdetected road obstacle area through road information from multiple vehicles multiple times. The following example introduces the process of calculating confidence.

[0119] For example, the cloud-side device 200 rasterizes the collected obstacle maps separately, where the grid corresponding to the ground area is 0, and each grid corresponding to the non-ground area stores the height of the obstacle area, where the ground area refers to the area below 0.2m, and the non-ground area refers to the area above 0.2m. Afterwards, the cloud-side device 200 calculates the number of times the non-ground area grid and the ground area grid appear in the grid corresponding to the road obstacle area. If the non-ground area grid appears once, the obstacle score is increased by a threshold value s. If the ground area grid appears once, the obstacle score is -1. The initial obstacle score is 50. If the obstacle score is less than 0 or greater than 100, it is truncated to ensure that the value is in the range of 0 to 100. Where s satisfies the following formula:

[0120]

[0121] Among them, α1 is the variance of the height of the grid-stored obstacles corresponding to the non-ground area, and α2 is the variance of the height of the grid-stored obstacles corresponding to the ground area.

[0122] The cloud-side device 200 iterates the data multiple times and calculates the confidence of the road obstacle area, that is, the obstacle probability of the road obstacle area. The calculation formula is obstacle score / 100. If the confidence of the road obstacle area is greater than 0.6, it is retained as an obstacle area.

[0123] S206. The cloud-side device 200 sends a second obstacle map to the vehicle end 100.

[0124] The cloud-side device 200 sends the second obstacle map to the vehicle side 100 .

[0125] S207. The vehicle end 100 collects and determines the vehicle's driving path based on the second obstacle map.

[0126] After the vehicle end 100 receives the second obstacle map sent by the cloud-side device 200, the vehicle end 100 can plan the vehicle's path based on the second obstacle map sent by the cloud-side device 200.

[0127] In a possible implementation, the vehicle end 100 may also receive a vehicle control instruction generated by the cloud-side device 200 based on the second obstacle map, and control the vehicle's driving path based on the vehicle control instruction.

[0128] It should be noted that after the cloud-side device 200 establishes the obstacle map, the vehicle side 100 can be operated in a group, and only needs to configure sensors such as lidar on the first vehicle to achieve the reduction of vehicle sensors.

[0129] In some possible implementations, when the vehicle end 100 or the cloud side device 200 determines the driving path of the vehicle based on the second obstacle map, the vehicle end 100 or the cloud side device 200 generates a waypoint of the vehicle according to the size of the road obstacle area in the second obstacle map and the wheel spacing of the vehicle, and the waypoint is used to indicate the coordinates through which the center of the vehicle passes. The vehicle end 100 or the cloud side device 200 determines the driving route of the vehicle based on the waypoint.

[0130] See also Figure 7 , Figure 7 A schematic diagram of an obstacle map provided in an embodiment of the present application. Figure 7 In the example shown, the dashed box is the predicted range of the vehicle's wheel spacing in the obstacle map, the triangle represents the vehicle's driving orientation angle, and the dark box is the range of the road obstacle area in the obstacle map. The vehicle end 100 or the cloud side device 200 can determine the vehicle's waypoint based on the relative position relationship between the vehicle's wheel spacing prediction range, the range of the road obstacle area, and the road boundary.

[0131] In some possible implementations, when the wheel spacing can cover the size of the road obstacle area, the vehicle side 100 or the cloud side device 200 controls the vehicle's waypoint to be the center of the road obstacle area.

[0132] from Figure 7 It can be seen from the example shown that when the predicted range of the vehicle wheel spacing can cover the range of the road obstacle area, the vehicle end 100 or the cloud side device 200 can use the center of the road obstacle area as the vehicle's waypoint. When the predicted range of the vehicle wheel spacing cannot cover the range of the road obstacle area, the vehicle end 100 or the cloud side device 200 can allow the vehicle's wheel spacing predicted range to bypass the road obstacle area when determining the vehicle's waypoint.

[0133] See also Figure 8 , Figure 8 A schematic diagram of a driving obstacle avoidance method provided in an embodiment of the present application. Figure 8In the example shown, the vehicle end 100 controls the waypoint of the vehicle to be the center of the road obstacle area, and rides through the road obstacle area. For example, when the wheel spacing of the vehicle is greater than the road obstacle area, the vehicle can ride through the road obstacle area. For another example, when the wheel spacing of the vehicle exceeds the road width, the vehicle chooses another route to bypass. For another example, when the height difference of the road obstacle area exceeds the preset height value, a parking alarm process is performed.

[0134] exist Figure 8 In the example shown, for example, the front and rear wheelbase of the vehicle is 5.1m, and the left and right wheel tracks are 3m, then the predicted wheel spacing range is 5.1m*3m, and if the road obstacle area is 1m*2.1m, the vehicle can ride through the road obstacle area.

[0135] It can be seen from the above embodiments that in the embodiments of the present application, when the vehicle approaches a road obstacle area, the vehicle end 100 can re-collect real-time road information of the road obstacle area and update the obstacle map based on the real-time road information, thereby avoiding obstacle avoidance failures caused by changes in the road obstacle area, and improving the vehicle driving obstacle avoidance method and the accuracy of vehicle driving obstacle avoidance.

[0136] Based on the above method embodiment, the embodiment of the present application also provides a vehicle obstacle avoidance device. The vehicle obstacle avoidance device provided by the embodiment of the present application is described in detail below.

[0137] See also Fig. 9 , Fig. 9 This is a schematic diagram of the structure of a vehicle obstacle avoidance device provided in an embodiment of the present application. Fig. 9 In the example shown, the vehicle obstacle avoidance device 900 is used to implement the various steps performed by the cloud-side device 200 in the above-mentioned embodiments. The vehicle obstacle avoidance device 900 includes an acquisition unit 901 and a processing unit 902 .

[0138] Among them, the acquisition unit 901 is used to acquire a first obstacle map, and the first obstacle map is used to predict the position of a road obstacle area for the vehicle at a long sight distance, and the long sight distance includes a distance from the vehicle that is greater than a first threshold. The processing unit 902 is used to collect first road information when the distance between the vehicle and the road obstacle area is less than or equal to the first threshold, and detect the existence of the road obstacle area based on the first road information. The processing unit 902 is also used to collect second road information when the distance between the vehicle and the road obstacle area is less than or equal to a second threshold, and generate a real-time obstacle map based on the second road information, and the second threshold is less than the first threshold. The acquisition unit 901 is also used to acquire a second obstacle map. The processing unit 902 is used to determine the driving path of the vehicle based on the second obstacle map, and the second obstacle map is an obstacle map generated by the cloud-side device based on the real-time obstacle map.

[0139] In some possible implementations, the processing unit 902 is further configured to control the vehicle to decelerate, or adjust the vehicle's driving path, when the vehicle determines that a road obstacle area exists based on the first road information.

[0140] In some possible implementations, the acquisition unit 901 is further used to send a real-time obstacle map to the cloud-side device. The acquisition unit 901 is specifically used to receive a second obstacle map sent by the cloud-side device, where the second obstacle map is an obstacle map generated by the cloud-side device based on the fusion of the real-time obstacle map and the crowdsourced obstacle map, and the crowdsourced obstacle map is an obstacle map generated by different vehicles.

[0141] In some possible implementations, the processing unit 902 is further configured to perform clustering operations on the road information collected by the vehicle based on a clustering algorithm, identify road obstacle areas, which include raised areas and sunken areas, and mark the road obstacle areas in the map to obtain a first obstacle map.

[0142] In some possible implementations, the processing unit 902 is further configured to calculate risk values ​​of different road areas based on the road surface information, and the risk values ​​are used to indicate the driving difficulty of different road areas. The processing unit 902 is also configured to provide a display interface of the first obstacle map, and the display interface is used to display the risk values ​​of different road areas.

[0143] In some possible implementations, the processing unit 902 is further configured to convert the coordinate system of the first obstacle map into the driving coordinate system of the vehicle, and annotate the laser point cloud acquired by the vehicle sensor in the driving coordinate system based on the converted first obstacle map.

[0144] In some possible implementations, the processing unit 902 is specifically configured to generate a waypoint of the vehicle according to the size of the road obstacle area in the second obstacle map and the wheel spacing of the vehicle, wherein the waypoint is used to indicate the coordinates through which the center of the vehicle passes, and determine the driving route of the vehicle based on the waypoint.

[0145] In some possible implementations, the processing unit 902 is further configured to control the waypoint of the vehicle to be the center of the road obstacle area when the wheel spacing can cover the size of the road obstacle area.

[0146] It should be understood that the division of units or modules in the above device is only a division of logical functions. In actual implementation, they can be fully or partially integrated into one physical entity, or they can be physically separated. And the units or modules in the device can all be implemented in the form of software calling through processing elements, or they can all be implemented in the form of hardware, or some units or modules can be implemented in the form of software calling through processing elements, and some units or modules can be implemented in the form of hardware. For example, each unit or module can be a separately established processing element, or it can be integrated in a certain chip of the device for implementation. In addition, it can also be stored in a memory in the form of a program, and a certain processing element of the device calls and executes the function of the unit or module. In addition, these units or modules can be integrated together in whole or in part, or they can be implemented independently. The processing element described here can also be a processor, which can be an integrated circuit with signal processing capabilities. In the implementation process, each step of the above method or each unit or module above can be implemented by an integrated logic circuit of hardware in a processor element or in the form of software calling through a processing element.

[0147] It is worth noting that, for the above method embodiments, for the sake of simplicity of description, they are all expressed as a series of action combinations, but those skilled in the art should know that the present application is not limited to the described order of actions. Secondly, those skilled in the art should also know that the embodiments described in the specification are all preferred embodiments, and the actions involved are not necessarily required for the present application.

[0148] Other reasonable step combinations that can be thought of by those skilled in the art based on the above description also fall within the scope of protection of this application. Secondly, those skilled in the art should also be familiar with the fact that the embodiments described in the specification are all preferred embodiments, and the actions involved are not necessarily required by this application.

[0149] See also Fig.10 , Fig.10 This is a schematic diagram of the structure of a vehicle-mounted device provided in an embodiment of the present application. Fig.10 As shown, the vehicle-mounted device 1000 includes: a processor 1001, a memory 1002, a communication interface 1003 and a bus 1004. The processor 1001, the memory 1002 and the communication interface 1003 are coupled via a bus (not marked in the figure). The memory 1002 stores instructions. When the execution instructions in the memory 1002 are executed, the vehicle-mounted device 1000 executes the method executed by the terminal device 100 or the cloud-side device 200 in the above method embodiment.

[0150] The vehicle-mounted device 1000 may be one or more integrated circuits configured to implement the above method, such as: one or more application specific integrated circuits (ASIC), or one or more microprocessors (digital signal processors, DSP), or one or more field programmable gate arrays (FPGA), or a combination of at least two of these integrated circuit forms. For another example, when the unit in the device can be implemented in the form of a processing element scheduler, the processing element can be a general-purpose processor, such as a central processing unit (CPU) or other processor that can call a program. For another example, these units can be integrated together and implemented in the form of a system-on-a-chip (SOC).

[0151] The processor 1001 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application specific integrated circuits (ASIC), field programmable gate arrays (FPGA) or other programmable logic devices, transistor logic devices, hardware components or any combination thereof. A general-purpose processor may be a microprocessor or any conventional processor.

[0152] The memory 1002 may be a volatile memory or a nonvolatile memory, or may include both volatile and nonvolatile memories. Among them, the nonvolatile memory may be a read-only memory (ROM), a programmable ROM (PROM), an erasable programmable ROM (EPROM), an electrically erasable programmable ROM (EEPROM), or a flash memory. The volatile memory may be a random access memory (RAM), which is used as an external cache. By way of example and not limitation, many forms of RAM are available, such as static RAM (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link dynamic random access memory (SLDRAM), and direct rambus RAM (DRRAM).

[0153] The memory 1002 stores executable program codes, and the processor 1001 executes the executable program codes to respectively implement the functions of the aforementioned terminal device 100 or the cloud-side device 200, thereby implementing the aforementioned content recommendation method. That is, the memory 1002 stores instructions for executing the aforementioned content recommendation method.

[0154] The memory 1002 stores executable program codes, and the processor 1001 executes the executable program codes to respectively implement the functions of the aforementioned acquisition unit 901 and the processing unit 902, thereby implementing the above-mentioned vehicle obstacle avoidance method. That is, the memory 1002 stores instructions for executing the above-mentioned vehicle obstacle avoidance method.

[0155] The communication interface 1003 uses a transceiver module such as, but not limited to, a network interface card or a transceiver to implement communication between the vehicle-mounted device 1000 and other devices or a communication network.

[0156] In addition to the data bus, the bus 1004 may also include a power bus, a control bus, a status signal bus, etc. The bus may be a peripheral component interconnect express (PCIe) bus, or an extended industry standard architecture (EISA) bus, a unified bus (Ubus or UB), a compute express link (CXL), a cache coherent interconnect for accelerators (CCIX), etc. The bus may be divided into an address bus, a data bus, a control bus, etc.

[0157] See also Fig.11 , Fig.11 A schematic diagram of the structure of a vehicle is provided for an embodiment of the present application. Fig.11 As shown, the vehicle 1100 includes at least one processor 1101 and a communication interface 1103. The communication interface 1103 and the at least one processor 1101 are interconnected via lines. The at least one processor 1101 is used to run a computer program or instruction. When the computer program or instruction is executed, the vehicle 1100 executes the method executed by the vehicle end 100 in the above method embodiment.

[0158] In a possible implementation, the vehicle 1100 described above in the present application further includes at least one memory 1102, in which executable program codes are stored, and the processor 1101 executes the executable program codes to respectively implement the functions of the aforementioned acquisition unit 901 and the processing unit 902, thereby implementing the above-mentioned vehicle driving obstacle avoidance method. That is, the memory 1102 stores instructions for executing the above-mentioned vehicle driving obstacle avoidance method.

[0159] In another embodiment of the present application, a computer-readable storage medium is also provided, in which computer-readable storage medium is stored computer execution instructions. When the processor of the device executes the computer execution instructions, the device executes the method executed by the vehicle side 100 or the cloud side device 200 in the above method embodiment.

[0160] In another embodiment of the present application, a computer program product is further provided, the computer program product including computer executable instructions, the computer executable instructions being stored in a computer readable storage medium. When the processor of the device executes the computer executable instructions, the device executes the method executed by the vehicle side 100 or the cloud side device 200 in the above method embodiment.

[0161] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0162] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0163] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0164] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0165] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, read-only memory), random access memory (RAM, random access memory), disk or optical disk and other media that can store program code.

Claims

1. A vehicle obstacle avoidance method, characterized in that: Applied to vehicle obstacle avoidance devices, including: Acquire a first obstacle map, where the first obstacle map is used to predict the position of a road obstacle area for the vehicle at a long sight distance, where the long sight distance includes a distance from the vehicle that is greater than a first threshold; When the distance between the vehicle and the road obstacle area is less than or equal to a first threshold, collecting first road information, and detecting the existence of the road obstacle area according to the first road information; When the distance between the vehicle and the road obstacle area is less than or equal to a second threshold, second road information is collected, and a real-time obstacle map is generated based on the second road information, and the second threshold is less than the first threshold; a second obstacle map is obtained, and a driving path of the vehicle is determined based on the second obstacle map, and the second obstacle map is an obstacle map generated by the cloud-side device based on the real-time obstacle map.

2. The method according to claim 1, characterized in that The method further comprises: When the vehicle determines that the road obstacle area exists based on the first road information, the vehicle is controlled to decelerate, or the driving path of the vehicle is adjusted.

3. The method according to claim 1 or 2, characterized in that: After generating the real-time obstacle map according to the second road surface information, the method further includes: Sending the real-time obstacle map to the cloud-side device; The obtaining of the second obstacle map comprises: Receive the second obstacle map sent by the cloud-side device, where the second obstacle map is an obstacle map generated by the cloud-side device based on the fusion of the real-time obstacle map and the crowdsourced obstacle map, and the crowdsourced obstacle map is an obstacle map generated by different vehicles.

4. The method according to any one of claims 1 to 3, characterized in that The obtaining of the first obstacle map further comprises: Performing clustering operation on the road surface information collected by the vehicle based on a clustering algorithm to identify the road surface obstacle area, wherein the road surface obstacle area includes a raised area and a sunken area; The first obstacle map is obtained by marking the road obstacle area in a map.

5. The method according to any one of claims 1 to 4, characterized in that The method further comprises: Calculating risk values ​​of different areas of the road surface based on the road surface information, wherein the risk values ​​are used to indicate the driving difficulty of different areas of the road surface; A display interface of the first obstacle map is provided, wherein the display interface is used to display risk values ​​of different areas of the road surface.

6. The method according to any one of claims 1 to 5, characterized in that The method further comprises: Converting the coordinate system of the first obstacle map to the driving coordinate system of the vehicle; The laser point cloud acquired by the vehicle sensor in the driving coordinate system is annotated based on the converted first obstacle map.

7. The method according to any one of claims 1 to 6, characterized in that The determining the driving path of the vehicle based on the second obstacle map comprises: Generating a waypoint of the vehicle according to the size of the road obstacle area in the second obstacle map and the wheel spacing of the vehicle, wherein the waypoint is used to indicate the coordinates through which the center of the vehicle passes; A driving route of the vehicle is determined based on the waypoints.

8. The method according to claim 7, characterized in that The method further comprises: When the wheel spacing can cover the size of the road obstacle area, the waypoint of the vehicle is controlled to be the center of the road obstacle area.

9. A vehicle obstacle avoidance device, characterized in that: include: an acquisition unit, configured to acquire a first obstacle map, wherein the first obstacle map is used to predict a position of a road obstacle area for a vehicle at a long sight distance, wherein the long sight distance includes a distance from the vehicle that is greater than a first threshold; a processing unit, configured to collect first road surface information when the distance between the vehicle and the road surface obstacle area is less than or equal to a first threshold, and detect the existence of the road surface obstacle area according to the first road surface information; The processing unit is further configured to collect second road surface information and generate a real-time obstacle map according to the second road surface information when the distance between the vehicle and the road surface obstacle area is less than or equal to a second threshold, and the second threshold is less than the first threshold; The acquisition unit is further used to acquire a second obstacle map. The processing unit is used to determine the driving path of the vehicle based on the second obstacle map, where the second obstacle map is an obstacle map generated by the cloud-side device based on the real-time obstacle map.

10. The device according to claim 9, characterized in that The processing unit is also used for: When the vehicle determines that the road obstacle area exists based on the first road information, the vehicle is controlled to decelerate, or the driving path of the vehicle is adjusted.

11. The device according to claim 9 or 10, characterized in that The acquisition unit is also used for: Sending the real-time obstacle map to the cloud-side device; The acquisition unit is specifically used for: Receive the second obstacle map sent by the cloud-side device, where the second obstacle map is an obstacle map generated by the cloud-side device based on the fusion of the real-time obstacle map and the crowdsourced obstacle map, and the crowdsourced obstacle map is an obstacle map generated by different vehicles.

12. The device according to any one of claims 9 to 11, characterized in that The processing unit is also used for: Performing clustering operation on the road surface information collected by the vehicle based on a clustering algorithm to identify the road surface obstacle area, wherein the road surface obstacle area includes a raised area and a sunken area; The first obstacle map is obtained by marking the road obstacle area in a map.

13. The device according to any one of claims 9 to 12, characterized in that The processing unit is also used for: Calculating risk values ​​of different areas of the road surface based on the road surface information, wherein the risk values ​​are used to indicate the driving difficulty of different areas of the road surface; A display interface of the first obstacle map is provided, wherein the display interface is used to display risk values ​​of different areas of the road surface.

14. The device according to any one of claims 9 to 13, characterized in that The processing unit is also used for: Converting the coordinate system of the first obstacle map to the driving coordinate system of the vehicle; The laser point cloud acquired by the vehicle sensor in the driving coordinate system is annotated based on the converted first obstacle map.

15. The device according to any one of claims 9 to 14, characterized in that The processing unit is specifically used for: Generating a waypoint of the vehicle according to the size of the road obstacle area in the second obstacle map and the wheel spacing of the vehicle, wherein the waypoint is used to indicate the coordinates through which the center of the vehicle passes; A driving route of the vehicle is determined based on the waypoints.

16. The device according to claim 15, characterized in that The processing unit is also used for: When the wheel spacing can cover the size of the road obstacle area, the waypoint of the vehicle is controlled to be the center of the road obstacle area.

17. A vehicle-mounted device, characterized in that: The vehicle-mounted device comprises a processor coupled to a memory, wherein the processor is used to execute instructions in the memory. When the instructions are executed by the processor, the vehicle-mounted device executes the method according to any one of claims 1 to 8.

18. A vehicle, characterized in that: The vehicle is used to perform the method according to any one of claims 1 to 8 above.

19. A vehicle obstacle avoidance system, characterized in that: The invention comprises a vehicle and a cloud-side device, wherein the vehicle is used to execute the method described in any one of claims 1 to 8.

20. A computer-readable storage medium having instructions stored thereon, characterized in that: When the instructions are executed, the computer is caused to perform the method according to any one of claims 1 to 8.

21. A computer program product, comprising instructions, characterized in that: When the instructions are executed, the computer implements the method according to any one of claims 1 to 8.

Images