Passable area detection method, device and mobile tool

Abstract

Embodiments of the present application relate to a kind of passable area detection method, comprising: laser radar point cloud data is handled, the point cloud grid map of target detection area is generated;Determine the initial ground height corresponding to the sliding window of preset initial direction;According to initial ground height, the search parameter of carrier is initialized setting, and the grid attribute of the area covered by sliding window is marked positively and negatively;According to the search parameter after initialization, respectively to preset first direction and preset second direction are traversed and searched, determine whether the ground height corresponding to the sliding window of preset first direction and the ground height corresponding to the sliding window of preset second direction are successfully acquired;If the ground height corresponding to the sliding window of preset first direction and / or the ground height corresponding to the sliding window of preset second direction are successfully acquired, search parameter is updated, and the grid attribute of the area covered by sliding window is marked positively and negatively;According to the search parameter after updating, to the direction parallel with preset initial direction is traversed and searched.

Description

Technical Field

[0001] This invention relates to the field of autonomous driving technology, and in particular to a method, apparatus and mobile tool for detecting passable areas. Background Technology

[0002] Autonomous vehicles need to identify passable road areas in the current environment, as well as potential positive obstacles (protruding obstacles such as rocks, trees, roadblocks, and raised curbs) and negative obstacles (potholes, cliffs, and sunken curbs). The ability to detect passable areas and both positive and negative obstacles becomes even more critical when high-precision map data of the environment cannot be obtained in advance. Summary of the Invention

[0003] The purpose of this invention is to address the shortcomings of existing technologies by providing a method for detecting passable areas. This method processes lidar point cloud data and then sets a sliding window to scan according to preset rules to obtain the ground height around the carrier. At the same time, it detects positive and negative obstacles in the carrier's operating environment, thus solving the problem of safe passage of the carrier without a high-precision map.

[0004] To achieve the above objectives, in a first aspect, the present invention provides a method for detecting passable areas, the method comprising:

[0005] Process the lidar point cloud data to generate a point cloud raster map of the target detection area;

[0006] Determine the initial ground height corresponding to the sliding window with the preset initial orientation;

[0007] Based on the initial ground height, the search parameters of the carrier are initialized, and the grid attributes of the sliding window coverage area are marked positively and negatively.

[0008] Based on the initialized search parameters, the system traverses and searches in the preset first direction and the preset second direction respectively to determine whether the ground height corresponding to the sliding window in the preset first direction and the ground height corresponding to the sliding window in the preset second direction have been successfully obtained.

[0009] If the ground height corresponding to the sliding window in the preset first direction and / or the ground height corresponding to the sliding window in the preset second direction are successfully obtained, the search parameters are updated, and the grid attributes of the sliding window coverage area are marked positively or negatively.

[0010] Based on the updated search parameters, a traversal search is performed in a direction parallel to the preset initial direction.

[0011] A second aspect of the present invention provides a passable area detection device, comprising:

[0012] The point cloud raster map generation module is used to process lidar point cloud data and generate a point cloud raster map of the target detection area.

[0013] The ground height determination module is used to determine the initial ground height corresponding to the sliding window with a preset initial direction;

[0014] The grid attribute marking module is used to initialize the search parameters of the carrier according to the initial ground height, and to mark the grid attributes of the sliding window coverage area with positive and negative labels.

[0015] The ground height determination module is also used to perform traversal searches in the preset first direction and the preset second direction according to the initialized search parameters, and to determine whether the ground height corresponding to the sliding window in the preset first direction and the ground height corresponding to the sliding window in the preset second direction have been successfully obtained.

[0016] The grid attribute marking module is also used to update the search parameters and mark the grid attributes of the sliding window coverage area positive and negative if the ground height corresponding to the sliding window in the preset first direction and / or the ground height corresponding to the sliding window in the preset second direction are successfully obtained.

[0017] The ground height determination module is also used to perform a traversal search in a direction parallel to the preset initial direction based on the updated search parameters.

[0018] In a third aspect, the present invention provides a computer server, comprising: a memory, a processor, and a transceiver;

[0019] The processor is used to couple with the memory, read and execute instructions in the memory to implement the passable area detection method described in any of the first aspects above;

[0020] The transceiver is coupled to the processor, and the processor controls the transceiver to send and receive messages.

[0021] In a fourth aspect, the present invention provides a chip system including a processor coupled to a memory storing program instructions, wherein when the program instructions stored in the memory are executed by the processor, the passable area detection method described in any of the first aspects is implemented.

[0022] In a fifth aspect, the present invention provides a computer system including a memory and one or more processors communicatively connected to the memory;

[0023] The memory stores instructions that can be executed by the one or more processors to cause the one or more processors to implement the passable area detection method as described in any of the first aspects above.

[0024] In a sixth aspect, the present invention provides a mobile tool comprising the computer server described in the third aspect above.

[0025] This invention provides a method for detecting passable areas. This method processes lidar point cloud data and then sets a sliding window to scan according to preset rules to obtain the ground height around the carrier. At the same time, it detects positive and negative obstacles in the carrier's operating environment, thus solving the problem of safe passage of the carrier without a high-precision map. Attached Figure Description

[0026] Figure 1 This is one of the flowcharts of the passable area detection method provided in Embodiment 1 of the present invention;

[0027] Figure 2 This is a top view of the carrier coordinate system provided in Embodiment 1 of the present invention;

[0028] Figure 3 This is a right view of the carrier coordinate system provided in Embodiment 1 of the present invention;

[0029] Figure 4 This is a schematic diagram of the operation process of the sliding window provided in Embodiment 1 of the present invention;

[0030] Figure 5 This is the second flowchart of the passable area detection method provided in Embodiment 1 of the present invention;

[0031] Figure 6 This is the third flowchart of the passable area detection method provided in Embodiment 1 of the present invention;

[0032] Figure 7 This is a schematic diagram of the sliding window search range provided in Embodiment 1 of the present invention;

[0033] Figure 8 Flowchart four of the passable area detection method provided in Embodiment 1 of the present invention;

[0034] Figure 9 Fifth flowchart of the passable area detection method provided in Embodiment 1 of the present invention;

[0035] Figure 10 A module structure diagram of the passable area detection device provided in Embodiment 2 of the present invention;

[0036] Figure 11 This is a flowchart illustrating the application scenario of the passable area detection method provided in Embodiment 7 of the present invention. Detailed Implementation

[0037] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this invention, and not all of them. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention.

[0038] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments.

[0039] The passable area detection method provided in this invention can be applied to autonomous vehicles to improve the safety of autonomous vehicle operation in the absence of high-precision maps.

[0040] Figure 1 This is a flowchart of the passable area detection method provided in Embodiment 1 of the present invention. The following is in conjunction with... Figure 1 The technical solution of the present invention will be described with reference to specific embodiments.

[0041] Example 1

[0042] The present invention provides a method for detecting passable areas, which mainly includes the following steps:

[0043] Step 101: Process the lidar point cloud data to generate a point cloud raster map of the target detection area.

[0044] Specifically, the LiDAR can be a single LiDAR or multiple LiDARs; this application does not impose a specific limitation. In other words, the point cloud data in this application can be a combination of point cloud data collected by different LiDARs. Since LiDAR collects point cloud data with the LiDAR as a reference, the collected point cloud data is three-dimensional coordinate data. However, for a vehicle, such as an autonomous vehicle, what is needed is the road surface condition that the vehicle will traverse, which directly corresponds to the horizontal coordinate system in which the vehicle is located. This horizontal coordinate system is then used to describe the environment in which the vehicle operates. Therefore, the LiDAR point cloud data must first undergo coordinate system transformation, that is, project the LiDAR point cloud data onto the vehicle's coordinate system, i.e., the horizontal coordinate system. Then, subsequent analysis of the passable area and positive and negative obstacles is performed in the horizontal coordinate system.

[0045] Furthermore, after generating a point cloud raster map of the target detection area, the number of points falling into each raster, the maximum height of the point cloud in the carrier coordinate system, and the minimum height are counted based on the point cloud raster map.

[0046] Specifically, based on the transformation relationship between the lidar coordinate system and the carrier coordinate system, the current ground elevation of the carrier can be obtained. It should be noted that the ground elevation is the elevation of the ground relative to the carrier's coordinate system at a preset distance around the carrier. The preset distance can be understood as a very close distance to the carrier, for example, 0.1m. It is understood that ground unevenness in the carrier's coverage area is ignored here. In a specific example, the ground elevation is 0.

[0047] Step 102: Determine the initial ground height corresponding to the sliding window with the preset initial direction.

[0048] Specifically, the preset initial direction can be selected as the front, left, or right of the vehicle. In this example, the front of the autonomous vehicle is used as the preset initial direction. Figure 2 and Figure 3 Different views of the carrier coordinate system are shown. (Combined) Figure 2 and Figure 3 As shown, taking an autonomous vehicle as an example, the origin of the carrier coordinate system is located at the center of the rear axle of the vehicle. The front of the autonomous vehicle is the X-axis direction, which is the longitudinal direction. The direction that is horizontal and perpendicular to the X-axis and to the left is the Y-axis direction, which is the lateral direction. The direction that is vertical and perpendicular to the X-axis and upward is the Z-axis direction, which is the height direction.

[0049] In this example, the sliding window specifications are the same, meaning the size of the input detection region (image) is the same each time a sliding window filtering calculation is performed. The sliding window here can be understood as the initial sliding window, such as... Figure 4 As shown in the image. At this moment, the coordinates directly facing the front of the vehicle are (X...). start ,0)(not shown in the figure).

[0050] By obtaining the initial ground height, it can be determined whether the area covered by the sliding window in the preset initial direction, i.e., the area directly in front of the autonomous vehicle, is a passable area, i.e., a road. In other words, the initial ground height refers to the ground height of the area where the window is first slid open.

[0051] Furthermore, the process of obtaining the initial ground height is as follows: Figure 5 As shown:

[0052] Step 10: Based on the number of point clouds, maximum height, and minimum height of each grid cell, determine whether there are planar grid cells in the sliding window coverage area and determine the first and second planar heights of the sliding window coverage area.

[0053] It is easy to understand that the sliding window coverage area includes multiple grids. By analyzing whether each grid is a plane, it can be determined whether there are planar grids in the sliding window coverage area.

[0054] like Figure 6As shown, in step S11, the number of grids in the area covered by the sliding window is determined according to the preset sliding window specifications and grid size.

[0055] The sliding window can be either rectangular or square; in this example, a square window will be used to explain the relevant technical solution. When the sliding window is square, its specifications are expressed by its radius.

[0056] First, calculate the sliding window search radius using Formula 1.

[0057] r = (r -meter / cell-size) (Formula 1)

[0058] Where r represents the sliding window search radius, r -meter This represents the sliding window radius, and cell_size represents the grid size.

[0059] Secondly, the number of grid cells in the sliding window coverage area is calculated using Formula 2.

[0060] M=(2r -meter +1)×(2r -meter +1) (Formula 2)

[0061] Where M represents the number of grid cells in the area covered by the sliding window. It should be noted that 1 represents the grid cell containing the center point of the sliding window.

[0062] Example, not limitation, r -meter =0.6 meters, cell_size =0.2 meters, then r =0.6 / 0.2 =3, then M = (3×2+1)×(3×2+1) =49. Figure 7 The text shows the sliding window search range when the sliding window search radius r is 3. That is, the grid where the center point of the sliding window is located is the center grid, and there are 3 grids with sliding window search radii on each side.

[0063] Step S12: If the number of point clouds in a grid is greater than the preset number of point clouds and the difference between the maximum height and the minimum height is less than the preset difference threshold, the grid is determined to be a planar grid.

[0064] Here, the preset difference threshold is mainly used to determine whether the point cloud heights within a grid are concentrated in a small range. This threshold is independent of the carrier height but related to the preset grid size. When the grid size is large, the ground area corresponding to the grid may have a large tilt, thus the point cloud height difference within the grid will be greater. For example, and not as a limitation, the preset difference threshold is 0.2.

[0065] When a grid cell is determined to be a planar grid cell, it can be determined that a planar grid cell exists within the sliding window coverage area. Therefore, it can be determined that the operating environment corresponding to the sliding window coverage area may have a plane, and steps S13 to S15 are continued. Conversely, if it is determined that the operating environment corresponding to the sliding window coverage area cannot have a plane, that is, there is no passable area, then the message "initial ground height acquisition failed" is returned, and the program ends.

[0066] Step S13: Statistically analyze the frequency of minimum height of the point cloud of the planar grid in the sliding window coverage area.

[0067] The minimum height can be statistically determined by quantizing the point cloud height of the raster, setting a quantization interval, and recording the frequency of occurrence of that minimum height. As an example, and not a limitation, let's assume the point cloud height is represented by h, and the quantization interval can be 0.05 meters. The specific division scheme is as follows:

[0068] Given point clouds with heights of 0.2, 0.23, and 0.24, the points belong to interval e; points with heights of 0.15, 0.15, 0.16, 0.17, and 0.19 belong to interval d. Therefore, whether two point clouds belong to the same quantization interval depends not only on the size of the quantization interval but also on the boundaries of the quantization partitions.

[0069] Step S14: Based on the frequency of minimum height and preset rules, obtain a preset number of minimum heights for the sliding window coverage area.

[0070] Specifically, the preset rule can be to select the minimum height that appears most frequently. In a specific example, the preset quantity could be 3.

[0071] Step S15: Determine the first plane height and the second plane height of the sliding window coverage area according to the preset number of minimum heights.

[0072] For example, traverse all the grids within the sliding window's coverage area to obtain three minimum heights, and sort the three heights in descending order of frequency of occurrence, labeling them as Z1, Z2, and Z3 respectively.

[0073] When only Z1 exists, the heights of the two planes are considered to be equal, that is, the height of the first plane = the height of the second plane = Z1.

[0074] When only Z1 and Z2 exist, the height of the first plane = Z1 and the height of the second plane = Z2.

[0075] When Z1, Z2, and Z3 exist simultaneously, there are two cases. First, if Z1 and Z2 share a quantization interval, and Z3 and Z1 exceed a quantization interval, then Z2 is discarded, and the height of the first plane = Z1, and the height of the second plane = Z3. Second, for cases other than the first, the height of the first plane is determined to be Z1, and the height of the second plane = Z2.

[0076] Using the same quantization interval divisions as an example, let's say a: [0≤h<0.05], b: [0.05≤h<0.1], c: [0.1≤h<0.15], d: [0.15≤h<0.2], and e: [0.2≤h<0.25]. Assuming the statistical results are: interval ade has no point clouds, interval b has 5 point clouds, and interval c has 3 point clouds, then the interval with the most point clouds is b, followed by c. The others have no point clouds and are not calculated. Therefore, Z1 is the minimum height of 0.05 corresponding to interval b, Z2 is the minimum height of 0.1 corresponding to interval c, and Z3 does not exist.

[0077] Step 20: When there is a planar grid in the area covered by the sliding window, and the height of the first plane meets the preset difference threshold or the height of the second plane meets the preset difference threshold, determine the initial ground height corresponding to the sliding window based on the height of the first plane, the height of the second plane, and the preset difference threshold.

[0078] Specifically, the absolute value of the difference between the height of the first plane and the height of the ground is taken and marked as the first absolute value; the absolute value of the difference between the height of the second plane and the height of the ground is taken and marked as the second absolute value; and the first absolute value and the second absolute value are compared with preset difference thresholds respectively.

[0079] If the first absolute value is greater than the preset difference threshold and the second absolute value is greater than the preset difference threshold, then return the initial ground height acquisition failure information.

[0080] If the first absolute value is not greater than a preset difference threshold or the second absolute value is not greater than a preset difference threshold, then the first absolute value and the second absolute value are compared, and the plane height corresponding to the smaller absolute value is taken as the initial ground height of the sliding window. In this example, the initial ground height is represented by ground_int.

[0081] For example, if the first absolute value is 0.15 and the second absolute value is 0.17, then the first plane height corresponding to the first absolute value is taken as the initial ground height corresponding to the sliding window.

[0082] Step 103: Initialize the search parameters of the carrier according to the initial ground height, and mark the grid attributes of the sliding window coverage area with positive and negative labels.

[0083] Specifically, the search parameters include the center ground height of the sliding window coverage area, the ground height in the preset first direction, the ground height in the preset second direction, the horizontal coordinate of the center of the sliding window coverage area, the horizontal coordinate of the preset first direction, and the horizontal coordinate of the preset second direction.

[0084] In this example, the first preset direction can be understood as the left side of the carrier, and the second preset direction can be understood as the right side of the carrier. Therefore, the center ground height of the sliding window coverage area is represented by `ground_middle`, the ground height of the first preset direction is represented by `ground_left`, and the ground height of the second preset direction is represented by `ground_right`. The x-coordinate of the center of the sliding window coverage area is represented by `middle_y`, the x-coordinate of the first preset direction is represented by `left_y`, and the x-coordinate of the second preset direction is represented by `right_y`.

[0085] After initialization: ground_middle=ground_left=ground_right=ground_int. middle_y=left_y=right_y=0.

[0086] After initialization settings,

[0087] First, the height difference of each grid is determined based on the maximum and minimum height of the point cloud in the carrier coordinate system.

[0088] Specifically, the height difference is obtained by subtracting the maximum height from the minimum height.

[0089] Secondly, based on the height difference and the initial ground height corresponding to the sliding window, the grid attributes of each grid cell are marked as positive or negative.

[0090] Specifically, positive and negative markings can be understood as positive and negative obstacle markings.

[0091] More specifically, such as Figure 8 As shown,

[0092] Step 1031: Compare the height difference with a preset positive obstacle height threshold. If the height difference is greater than the preset positive obstacle height threshold, proceed to step 1032; if the height difference is not greater than the preset positive obstacle height threshold, proceed to steps 1033-1034. Afterwards, the grid attributes are positively marked in the following two ways:

[0093] Step 1032: Add positive tags to the raster attributes;

[0094] Step 1033: Generate the first height difference based on the maximum height of the point cloud of the raster and the initial ground height corresponding to the sliding window;

[0095] Step 1034: When the first height difference is greater than the preset positive obstacle height threshold, the grid attribute is positively marked.

[0096] Therefore, the above two types of grid attributes are positive markers, and the area corresponding to the grid is the area where positive obstacles exist. Among them, the latter type of obstacle is a low obstacle.

[0097] Furthermore, negative labeling of raster attributes includes:

[0098] When the first height difference is less than a preset difference threshold and the height difference is greater than a preset negative obstacle threshold, the grid attribute is negatively marked. In other words, when the first height difference is less than the preset difference threshold, the maximum height is approximately the same as the initial ground height corresponding to the sliding window.

[0099] Furthermore, to reduce computational load, as a preferred approach, after a raster attribute is negatively labeled, the method also includes, as... Figure 9 The steps shown are as follows:

[0100] Step S1: Determine the grid cells whose distance is negatively marked and which satisfy a preset first distance.

[0101] Specifically, when the grid meets the preset first distance, step S2 or steps S3 to S5 are executed.

[0102] Step S2: Compare the minimum height of the point cloud of each grid cell that meets the preset first distance with the initial ground height. When the minimum height of the point cloud of a grid cell is less than the initial ground height, mark the grid cell attribute negatively.

[0103] Step S3: Determine whether the distance between the grid that meets the preset first distance and the origin of the carrier coordinate system meets the preset second distance. If the preset second distance is met, proceed to step S4.

[0104] Step S4: Determine if the number of point clouds in the raster is zero. If the number of point clouds in the raster is zero, proceed to step S5.

[0105] Step S5: Negatively mark the raster attributes.

[0106] Step 104: Based on the initialized search parameters, perform a traversal search in the preset first direction and the preset second direction respectively to determine whether the ground height corresponding to the sliding window in the preset first direction and the ground height corresponding to the sliding window in the preset second direction have been successfully obtained.

[0107] Specifically, firstly, with the center y-coordinate (x) of the sliding window unchanged, the system iterates leftward sequentially with a preset step size based on the center x-coordinate (middle_y) of the initially initialized sliding window coverage area, then obtains the heights of the two planes within the sliding window. The method for obtaining the two plane heights is the same as the process for obtaining the first and second plane heights, and will not be repeated here. If the two plane heights are successfully obtained, they are compared with the initialized preset ground height (ground_left) in the first direction, respectively, and two differences are obtained. The plane height with a difference less than a preset difference threshold and the smaller difference is selected as the new ground height of the sliding window in the preset first direction, which is the new ground height on the left side. Secondly, the current x-coordinate (left_y) of the preset first direction is updated to the center x-coordinate of the current sliding window coverage area.

[0108] If both differences are not less than a preset difference threshold, the search continues in the preset first direction. If the heights of the two planes are not successfully obtained, the program terminates.

[0109] Following the same steps, a traversal search is performed in the preset second direction. However, during the comparison process, it is compared with the ground height (ground_right) of the preset second direction, and the current x-coordinate (right_y) of the preset second direction is updated to the center x-coordinate of the current sliding window coverage area. It should be noted that this application does not specifically limit whether the traversal search in the preset first direction or the preset second direction is performed first.

[0110] For example Figure 4 As shown, the direction the front of the car faces is longitudinal, where both longitudinal and lateral have preset maximum distances, i.e., the farthest longitudinal movement and the farthest lateral movement. During the sliding window movement, there are preset step lengths, i.e., longitudinal step length and lateral step length.

[0111] Step 105: If the ground height corresponding to the sliding window in the preset first direction and / or the ground height corresponding to the sliding window in the preset second direction are successfully obtained, the search parameters are updated and the grid attributes of the sliding window coverage area are marked positively or negatively.

[0112] Specifically, the method for positive and negative marking is the same as described above and will not be repeated here. If left_y = right_y = middle_y, it means that the ground height corresponding to the sliding window has not been successfully obtained in either the preset first direction or the preset second direction, and the program ends. As long as the ground height corresponding to the sliding window in one of the directions is successfully obtained, the search parameters can be updated. That is, middle_y = (right_y + left_y) / 2, which serves as the center x-coordinate of the current sliding window coverage area.

[0113] Step 106: Based on the updated search parameters, perform a traversal search in a direction parallel to the preset initial direction.

[0114] Specifically, the direction that is parallel to the initial direction can be understood as the direction that is parallel to the X-axis, but not directly in front of the car.

[0115] More specifically, firstly, the new starting point coordinates are (x, middle), and the heights of the two planes corresponding to the sliding window at the new starting point are obtained.

[0116] If the ground height corresponding to the sliding window in the direction parallel to the preset initial direction is successfully obtained, the two plane heights are subtracted from the ground heights obtained by the preset first direction traversal search and the preset second direction traversal search respectively. If the difference is not greater than the preset difference threshold, the plane height corresponding to the smaller difference is taken as the new ground_middle.

[0117] If two plane heights are not successfully obtained, or the difference between them is greater than the preset difference threshold, then ground_middle is not updated with plane heights. Instead, the center ground height of the sliding window coverage area obtained from the previous traversal search is used as the new ground_middle, and the search parameters are updated again, ground_left = ground_right = ground_middle.

[0118] Simultaneously initialize middle_y = left_y = right_y = 0. Repeat steps 103 to 106 until the vertical coordinate, i.e., x, reaches the preset maximum value, then terminate the program.

[0119] This invention provides a method for detecting passable areas. This method processes lidar point cloud data and then sets a sliding window to scan according to preset rules to obtain the ground height around the carrier. At the same time, it detects positive and negative obstacles in the carrier's operating environment, thus solving the problem of safe passage of the carrier without a high-precision map.

[0120] Example 2

[0121] Figure 10 This is a module structure diagram of a passable area detection device provided in Embodiment 2 of the present invention. The device includes:

[0122] The point cloud raster map generation module 10 is used to process the lidar point cloud data and generate a point cloud raster map of the target detection area.

[0123] Ground height determination module 20 is used to determine the initial ground height corresponding to the sliding window with a preset initial direction;

[0124] The grid attribute marking module 30 is used to initialize the search parameters of the carrier according to the initial ground height, and to mark the grid attributes of the sliding window coverage area with positive and negative labels.

[0125] The ground height determination module 20 is also used to perform traversal searches in the preset first direction and the preset second direction according to the initialized search parameters, and to determine whether the ground height corresponding to the sliding window in the preset first direction and the ground height corresponding to the sliding window in the preset second direction have been successfully obtained.

[0126] The grid attribute marking module 30 is also used to update the search parameters and mark the grid attributes of the sliding window coverage area positive and negative if the ground height corresponding to the sliding window in the preset first direction and / or the ground height corresponding to the sliding window in the preset second direction are successfully obtained.

[0127] The ground height determination module 20 is also used to perform a traversal search in a direction parallel to the preset initial direction based on the updated search parameters.

[0128] The second embodiment of the present invention provides a passable area detection device, which can execute the method steps in the above method embodiment. The point cloud raster map generation module 10 implements step 101, the ground height determination module 20 implements steps 102, 104, and 106, and the raster attribute marking module 30 implements steps 103 and 105.

[0129] The specific implementation principles and technical effects are similar, and will not be elaborated here.

[0130] It should be noted that the division of the various modules in the above device is merely a logical functional division. In actual implementation, they can be fully or partially integrated into a single physical entity, or they can be physically separated. Furthermore, these modules can be implemented entirely in software via processing element calls; they can be fully implemented in hardware; or some modules can be implemented by processing element calls to software, while others are implemented in hardware. For example, a determination module can be a separate processing element, or it can be integrated into a chip within the above device. Alternatively, it can be stored as program code in the memory of the above device, and its function can be called and executed by a processing element of the device. The implementation of other modules is similar. Moreover, these modules can be fully or partially integrated together, or they can be implemented independently. The processing element described here can be an integrated circuit with signal processing capabilities. In the implementation process, each step of the above method or each of the above modules can be completed through integrated logic circuits in the hardware of the processor element or through software instructions.

[0131] For example, these modules can be one or more integrated circuits configured to implement the above methods, such as one or more Application Specific Integrated Circuits (ASICs), one or more Digital Signal Processors (DSPs), or one or more Field Programmable Gate Arrays (FPGAs). As another example, when a module is implemented using processing element scheduler code, the processing element can be a general-purpose processor, such as a Central Processing Unit (CPU) or other processor capable of calling program code. Furthermore, these modules can be integrated together as a System-on-a-Chip (SoC).

[0132] In the above embodiments, implementation can be achieved, in whole or in part, through software, hardware, firmware, or any combination thereof. When implemented in software, it can be implemented, in whole or in part, as a computer program product. This computer program product includes one or more computer instructions. When these computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The aforementioned computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The aforementioned computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the aforementioned computer instructions can be transmitted from one website, computer, server, or data center to another via wired (e.g., coaxial cable, fiber optic, Digital Subscriber Line (DSL)) or wireless (e.g., infrared, wireless, Bluetooth, microwave, etc.) means. The aforementioned computer-readable storage medium can be any available medium that a computer can access, or a data storage device such as a server or data center that integrates one or more available media. The aforementioned available media can be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., DVDs), or semiconductor media (e.g., solid-state disks (SSDs)).

[0133] Example 3

[0134] A computer server provided in Embodiment 3 of the present invention includes: a memory, a processor, and a transceiver;

[0135] The processor is used to couple with the memory, read and execute instructions in the memory to implement the passable area detection method of any one of the above embodiments;

[0136] The transceiver is coupled to the processor, and the processor controls the transceiver to send and receive messages.

[0137] The computer server provided in Embodiment 3 of the present invention may be, for example, an autonomous driving domain controller applied to an autonomous vehicle. It is communicatively connected to various sensors installed on the autonomous vehicle, acquires environmental perception data and vehicle speed information through these sensors, extracts obstacle information based on the acquired environmental perception data, generates the location information of the autonomous vehicle based on the vehicle speed information, and performs path planning based on the location information and obstacle information.

[0138] Example 4

[0139] The present invention provides a chip system in embodiment four, including a processor and a memory coupled together. The memory stores program instructions. When the program instructions stored in the memory are executed by the processor, the passable area detection method of any one of the above embodiments is implemented.

[0140] Example 5

[0141] A computer system provided in Embodiment 5 of the present invention includes a memory and one or more processors communicatively connected to the memory;

[0142] The memory stores instructions that can be executed by one or more processors to enable one or more processors to implement the passable area detection method as described in any of the embodiments above.

[0143] Example 6

[0144] The sixth embodiment of the present invention provides a mobile tool, which includes the computer server described in the third embodiment above.

[0145] The mobile vehicle includes, but is not limited to, vehicles that meet one of the six autonomous driving technology levels (L0-L5) as defined by the Society of Automotive Engineers International (SAE International) or the Chinese national standard "Classification of Driving Automation for Automobiles". For example, it can be vehicle equipment or robotic equipment with the following functions:

[0146] (1) Passenger transport function, such as family cars, buses, etc.;

[0147] (2) Cargo carrying function, such as ordinary trucks, box trucks, trailer trucks, enclosed trucks, tank trucks, flatbed trucks, container trucks, dump trucks, special structure trucks, etc.;

[0148] (3) Tool functions, such as logistics delivery vehicles, automated guided vehicles (AGVs), patrol vehicles, cranes, hoists, excavators, bulldozers, loaders, road rollers, loaders, off-road engineering vehicles, armored engineering vehicles, sewage treatment vehicles, sanitation vehicles, vacuum trucks, floor scrubbers, water sprinkler trucks, sweeping robots, food delivery robots, shopping guide robots, lawnmowers, golf carts, etc.

[0149] (4) Entertainment functions, such as recreational vehicles, amusement park automatic driving devices, balance bikes, etc.;

[0150] (5) Special rescue functions, such as fire trucks, ambulances, power repair vehicles, and engineering emergency rescue vehicles.

[0151] This application does not strictly limit the types of mobile tools, and will not list them all here.

[0152] Example 7

[0153] Embodiment 7 of the present invention further illustrates the technical solution of this application by taking the application of this method in an autonomous vehicle as an example. Figure 11 As shown.

[0154] Step 1: Input the point cloud and rasterize it;

[0155] Step 2: Obtain the ground height in the initial sliding window in front of the vehicle to get the initial height, and initialize the left and right search parameters and mark the initial sliding window.

[0156] Step 3: Traverse to the left, update the ground height and mark the sliding window;

[0157] Step 4: Traverse to the right, update the ground height and mark the sliding window;

[0158] Step 5: Summarize the first two steps and move forward one step to update the ground height and mark the sliding window.

[0159] In steps 2 to 4, if the ground height update fails, the entire process ends; in step 5, if the ground height update fails or the preset maximum distance is reached in the longitudinal direction, the entire process ends; otherwise, the process returns to step 3 to continue iterating.

[0160] Those skilled in the art will further recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0161] The steps of the methods or algorithms described in conjunction with the embodiments disclosed herein can be implemented in hardware, software modules executed by a processor, or a combination of both. The software modules can be located in random access memory (RAM), main memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disks, removable disks, CD-ROM power system control methods, or any other form of storage medium known in the art.

[0162] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for detecting passable areas, characterized in that, The method includes: Process the lidar point cloud data to generate a point cloud raster map of the target detection area; Determine the initial ground height corresponding to the sliding window with the preset initial orientation; Based on the initial ground height, the search parameters of the carrier are initialized, and the grid attributes of the sliding window coverage area are marked with positive and negative signs; the search parameters include the center ground height of the sliding window coverage area, the ground height in a preset first direction, the ground height in a preset second direction, the center horizontal coordinate of the sliding window coverage area, the horizontal coordinate in the preset first direction, and the horizontal coordinate in the preset second direction; Based on the initialized search parameters, the system traverses and searches in the preset first direction and the preset second direction respectively to determine whether the ground height corresponding to the sliding window in the preset first direction and the ground height corresponding to the sliding window in the preset second direction have been successfully obtained. If the ground height corresponding to the sliding window in the preset first direction and / or the ground height corresponding to the sliding window in the preset second direction are successfully obtained, the search parameters are updated, and the grid attributes of the sliding window coverage area are marked positively or negatively. Based on the updated search parameters, perform a traversal search in a direction parallel to the preset initial direction; Specifically, determining the initial ground height corresponding to the sliding window with the preset initial direction includes: determining whether there are planar grids in the sliding window coverage area based on the number of point clouds, maximum height, and minimum height of each grid, and determining the first and second planar heights of the sliding window coverage area; The steps of determining whether a planar grid exists in the sliding window coverage area and determining the first and second planar heights of the sliding window coverage area based on the point cloud quantity, maximum height, and minimum height of each grid cell specifically include: Determine the number of grids in the area covered by the sliding window based on the preset sliding window specifications and grid size; If the number of point clouds in a grid is greater than a preset number of point clouds and the difference between the maximum height and the minimum height is less than a preset difference threshold, the grid is determined to be a planar grid. The frequency of minimum height of the point cloud of the planar grid in the area covered by the sliding window is statistically analyzed; Based on the frequency of minimum height and preset rules, obtain a preset number of minimum heights for the sliding window coverage area; Based on the preset number of minimum heights, determine the first plane height and the second plane height of the sliding window coverage area.

2. The method for detecting passable areas according to claim 1, characterized in that, After processing the lidar point cloud data to generate a point cloud raster map of the target detection area, the process also includes: The number of point clouds falling into each grid cell, the maximum height of the point cloud in the carrier coordinate system, and the minimum height are counted.

3. The method for detecting passable areas according to claim 2, characterized in that, The determination of the initial ground height corresponding to the sliding window with the preset initial orientation specifically includes: When the sliding window covers an area with a planar grid, and the height of the first plane meets a preset difference threshold or the height of the second plane meets the preset difference threshold, the initial ground height corresponding to the sliding window is determined based on the height of the first plane, the height of the second plane, and the preset difference threshold.

4. The method for detecting passable areas according to claim 2, characterized in that, The step of marking the grid attributes of the sliding window coverage area with positive and negative labels specifically includes: The height difference of each grid cell is determined based on the maximum and minimum height of the point cloud in the carrier coordinate system. Based on the height difference and the initial ground height corresponding to the sliding window, the grid attributes of each grid cell are marked positively or negatively.

5. The method for detecting passable areas according to claim 4, characterized in that, The step of assigning positive and negative labels to the grid attributes of each grid cell based on the height difference and the initial ground height corresponding to the sliding window specifically includes: The height difference is compared with a preset positive obstacle height threshold. When the height difference is greater than the preset positive obstacle height threshold, the grid attribute is positively marked; or, When the height difference is not greater than the preset positive obstacle height, a first height difference is generated based on the maximum height of the grid's point cloud and the initial ground height corresponding to the sliding window; When the first height difference is greater than the preset positive obstacle height threshold, the grid attribute is positively marked.

6. The method for detecting passable areas according to claim 5, characterized in that, The step of assigning positive and negative labels to the grid attributes of each grid cell based on the height difference and the initial ground height of the sliding window further includes: When the first height difference is less than a preset difference threshold and the height difference is greater than a preset negative obstacle threshold, the grid attribute is negatively marked.

7. The method for detecting passable areas according to claim 6, characterized in that, After negatively marking the raster attribute, the method further includes: Identify the grid cells whose distance to the negative marker satisfies a preset first distance; The minimum height of the point cloud of each grid that meets the preset first distance is compared with the initial ground height. When the minimum height of the point cloud of a grid is less than the initial ground height, the grid attribute is negatively marked.

8. The method for detecting passable areas according to claim 7, characterized in that, After determining that the grid with the negative distance mark satisfies the preset first distance, the process further includes: Determine whether the distance between the grid that satisfies the preset first distance and the origin of the carrier coordinate system satisfies the preset second distance; When the condition is met, determine whether the number of point clouds in the grid is zero; When the number of point clouds in the raster is zero, the raster attribute is negatively marked.

9. The method for detecting passable areas according to claim 1, characterized in that, After determining whether the ground height corresponding to the sliding window in the direction parallel to the preset initial direction has been successfully obtained, the method further includes: If the search is not successfully completed, the search parameters are updated again, and the search is performed again in the preset first direction and the preset second direction.

10. A passable area detection device, characterized in that, include: The point cloud raster map generation module is used to process lidar point cloud data and generate a point cloud raster map of the target detection area. The ground height determination module is used to determine the initial ground height corresponding to the sliding window with a preset initial direction; The grid attribute marking module is used to initialize the search parameters of the carrier according to the initial ground height, and to mark the grid attributes of the sliding window coverage area with positive and negative labels; the search parameters include the center ground height of the sliding window coverage area, the ground height in a preset first direction, the ground height in a preset second direction, the center horizontal coordinate of the sliding window coverage area, the horizontal coordinate in the preset first direction, and the horizontal coordinate in the preset second direction. The ground height determination module is also used to perform traversal searches in the preset first direction and the preset second direction according to the initialized search parameters, and to determine whether the ground height corresponding to the sliding window in the preset first direction and the ground height corresponding to the sliding window in the preset second direction have been successfully obtained. The grid attribute marking module is also used to update the search parameters and mark the grid attributes of the sliding window coverage area positive and negative if the ground height corresponding to the sliding window in the preset first direction and / or the ground height corresponding to the sliding window in the preset second direction are successfully obtained. The ground height determination module is further configured to perform a traversal search in a direction parallel to the preset initial direction based on the updated search parameters; wherein, determining the initial ground height corresponding to the sliding window in the preset initial direction specifically includes: determining whether a planar grid exists in the sliding window coverage area and determining the first and second planar heights of the sliding window coverage area based on the point cloud count, maximum height, and minimum height of each grid; the determination of whether a planar grid exists in the sliding window coverage area and determining the first and second planar heights of the sliding window coverage area based on the point cloud count, maximum height, and minimum height of each grid specifically includes: determining the number of grids in the sliding window coverage area based on preset sliding window specifications and grid size; if the point cloud count of a grid is greater than the preset point cloud count and the difference between the maximum and minimum heights is less than a preset difference threshold, the grid is determined to be a planar grid; the frequency of the minimum height of the point cloud of the planar grid in the sliding window coverage area is statistically analyzed; a preset number of minimum heights in the sliding window coverage area are obtained based on the frequency of the minimum heights and preset rules; and the first and second planar heights of the sliding window coverage area are determined based on the preset number of minimum heights.

11. A computer server, characterized in that, include: Memory, processor, and transceiver; The processor is used to couple with the memory, read and execute instructions in the memory to implement the passable area detection method according to any one of claims 1-9; The transceiver is coupled to the processor, and the processor controls the transceiver to send and receive messages.

12. A chip system, characterized in that, The method includes a processor coupled to a memory storing program instructions, which, when executed by the processor, implement the passable area detection method according to any one of claims 1-9.

13. A computer system, characterized in that, Includes a memory, and one or more processors communicatively connected to the memory; The memory stores instructions that can be executed by the one or more processors to enable the one or more processors to implement the passable area detection method as described in any one of claims 1-9.

14. A mobile tool, characterized in that, Includes the computer server described in claim 11 above.

Images