Wall surface obstacle detection and shape estimation method and device

Abstract

The invention relates to a wall surface obstacle detection and shape estimation method and device. The method comprises the following steps: acquiring point cloud data of a plurality of laser radars of a vehicle; combining and splicing the point cloud data through time synchronization and coordinate conversion; based on sector division, straight lines are extracted from the spliced point cloud data, and the types of the straight lines are judged; segmenting the wall surface in the point cloud data according to each straight line and the straight line type; and clustering the segmented wall point cloud data, and performing state and attitude estimation according to a clustering result. According to the method, through time synchronization and coordinate transformation, the accuracy of point cloud wall segmentation is improved according to the straight line type, and then shape estimation is realized through clustering analysis, so that the accuracy and robustness of wall obstacle and shape estimation are improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of vehicle assisted driving, and particularly relates to a method and device for detecting and shape estimating wall obstacles, and more particularly to a method and device for detecting and shape estimating wall obstacles based on multi-source point cloud data. Background Art

[0002] The methods for segmenting wall and ground points in lidar point clouds mainly include methods based on geometric features, color features, deep learning, etc. However, the existing methods have some problems, such as poor segmentation effect for point cloud data in complex environments, inaccurate segmentation results, and slow segmentation speed. The existing methods for estimating the target shape of intelligent vehicles are affected by uncertain factors such as the sparsity and noise of lidar point clouds, resulting in poor accuracy of the size and heading of the three-dimensional bounding box fitted to the L-shaped point cloud. Moreover, there is currently no effective method for detecting wall obstacles. Therefore, it is necessary to propose a method for detecting wall obstacles and shape estimation from multi-lidar point cloud data. Summary of the Invention

[0003] To improve the accuracy of detecting wall obstacles in lidar point clouds, in a first aspect of the present invention, there is provided a method for detecting and shape estimating wall obstacles, including: acquiring point cloud data of multiple lidars of a vehicle; merging and splicing the point cloud data through time synchronization and coordinate transformation; extracting straight lines from the spliced point cloud data based on sector division and judging the type of the straight lines; segmenting the wall in the point cloud data according to each straight line and its type; clustering the segmented wall point cloud data and performing state and pose estimation according to the clustering result.

[0004] In some embodiments of the present invention, the merging and splicing the point cloud data through time synchronization and coordinate transformation includes: performing time synchronization on the point cloud data; converting the time-synchronized point cloud data to the same coordinate system through coordinate transformation; merging the transformed point cloud data and outputting the merged point cloud.

[0005] In some embodiments of the present invention, extracting straight lines from the spliced point cloud data and judging the type of the straight lines includes: uniformly dividing the point cloud data in the circumferential direction into multiple sectors; dividing each sector into multiple grids, traversing each grid and extracting straight lines according to the coordinate information of the grid; calculating the slope of the straight line according to the start point and end point of the extracted straight line; judging the type of the straight line by comparing the slope of the straight line with a preset wall slope threshold and a ground slope threshold respectively.

[0006] Further, the determination of the straight line type by comparing the slope of the straight line with the preset wall slope threshold and the ground slope threshold respectively includes: if the slope of the straight line is less than the ground slope threshold, the intercept is limited to a preset value, and the connected grids are not less than 2, then the straight line is determined to be a ground line; if the slope of the straight line is greater than the preset wall slope threshold and the connected grids are not less than 3, then the straight line is determined to be a wall line.

[0007] Furthermore, the segmentation of the wall surface in the point cloud data according to each straight line and its straight line type includes: traversing the point cloud in each sector grid, if the difference between the height value of the point cloud and the average value of the height of the point cloud in the upper sector is within the preset value, then the point cloud is marked as ground point cloud; if the distance between the point cloud and the wall straight line is less than the wall straight line threshold, then the point cloud is marked as wall point cloud.

[0008] In some embodiments of the present invention, the state and pose estimation according to the clustering result includes: traversing each point cloud data of the clustering result and calculating the centroid of the point cloud; determining the optimal fitting angle of each clustering result based on the rectangular fitting method of angle search; calculating the center and size of the bounding box of each clustering result through the optimal fitting angle.

[0009] In a second aspect of the present invention, there is provided a wall obstacle detection and shape estimation device, including: an acquisition module for acquiring point cloud data of multiple lidars of a vehicle; a merging module for merging and splicing the point cloud data through time synchronization and coordinate transformation; a segmentation module for extracting straight lines from the spliced point cloud data based on sector division and determining the straight line type; segmenting the wall surface in the point cloud data according to each straight line and its straight line type; an estimation module for clustering the segmented wall surface point cloud data and performing state and pose estimation according to the clustering result.

[0010] Further, the merging module includes: a synchronization unit for performing time synchronization on the point cloud data; a transformation unit for transforming the time-synchronized point cloud data into the same coordinate system through coordinate transformation; a merging unit for merging the transformed point cloud data and outputting the merged point cloud.

[0011] In a third aspect of the present invention, there is provided an electronic device, including: one or more processors; a storage device for storing one or more programs, when the one or more programs are executed by the one or more processors, enabling the one or more processors to implement the wall obstacle detection and shape estimation method provided by the present invention in the first aspect.

[0012] In a fourth aspect of the present invention, there is provided a computer-readable medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the method for detecting and shape estimating wall obstacles provided by the present invention in the first aspect is implemented.

[0013] The beneficial effects of the present invention are as follows: The present invention first performs preliminary processing on multi-source lidar data through time synchronization and coordinate transformation to achieve data consistency; then extracts straight lines by calculating coordinate points and pose information, and distinguishes walls and the ground by straight line types, thereby improving the accuracy of point cloud wall segmentation; finally, shape estimation is achieved through clustering analysis, further improving the accuracy and robustness of wall obstacle and shape estimation. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 is a schematic diagram of the basic process of the method for detecting and shape estimating wall obstacles in some embodiments of the present invention; Figure 2 is a schematic diagram of the specific process of merging and splicing point cloud data in some embodiments of the present invention; Figure 3 is a schematic diagram of the specific process of wall segmentation of point cloud data in some embodiments of the present invention; Figure 4 is a schematic diagram of the specific process of pose estimation of point cloud data in some embodiments of the present invention; Figure 5 is a schematic diagram of the processing effect of point cloud data in some embodiments of the present invention; Figure 6 is a schematic diagram of the specific process of the method for detecting and shape estimating wall obstacles in some embodiments of the present invention; Figure 7 is a schematic diagram of the structure of the device for detecting and shape estimating wall obstacles in some embodiments of the present invention; Figure 8 is a schematic diagram of the structure of an electronic device in some embodiments of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0015] The principles and features of the present invention are described below with reference to the accompanying drawings. The examples given are only for explaining the present invention and are not intended to limit the scope of the present invention.

[0016] Refer to Figure 1 、 Figure 4 and Figure 6, in the first aspect of the present invention, a method for detecting and estimating the shape of wall obstacles is provided, including: S100. Obtaining point cloud data of multiple lidars of a vehicle; S200. Merging and splicing the point cloud data through time synchronization and coordinate transformation; S300. Based on sector division, extracting straight lines from the spliced point cloud data and judging the type of straight lines; segmenting the wall in the point cloud data according to each straight line and its type; S400. Clustering the segmented wall point cloud data and performing state and pose estimation according to the clustering result.

[0017] Without loss of generality, in the present invention, the projection point of the center of the rear axle of the vehicle on the ground is taken as the origin, the positive front direction of the vehicle is taken as the positive x-axis direction, the right side is taken as the positive y-axis direction, and the vertically upward direction is taken as the positive z-axis direction; an example of an autonomous vehicle equipped with a 32-line lidar located directly above the origin and two 16-line lidars located symmetrically on the left and right sides behind the vehicle is used to illustrate the proposed method.

[0018] Reference Figure 2 , in step S200 of some embodiments of the present invention, the merging and splicing of the point cloud data through time synchronization and coordinate transformation includes: S201. Performing time synchronization on the point cloud data; Specifically, the PCL (Point Cloud Library) and the message_filters library of ROS are used to achieve approximate time synchronization, and then the point cloud data of multiple lidars are converted to a unified coordinate system. The approximate time synchronization strategy is used to synchronize multiple point cloud data. The main function of the synchronization strategy is to synchronize data from different sensors as much as possible when the timestamps cannot fully match. The synchronization strategy allows a certain time difference, and the allowable range of the time difference can be adjusted through the parameters of the synchronization strategy constructor. If the vehicle obtains point cloud data of different lidars at a certain moment, the synchronization strategy will find a set of data with the closest timestamp. If the time difference of the found set of data is within the allowable range, this set of messages is considered synchronized. The parameter used is the default parameter of 100 milliseconds, indicating that the maximum allowable time deviation is 100 milliseconds. Once a set of approximately synchronized data is found, this set of data will be passed to the next step.

[0019] S202. Through coordinate transformation, converting the time-synchronized point cloud data into the same coordinate system; The coordinate system transformation is divided into two steps: In the first step, the point cloud obtained by the 16-line lidar is first transformed from the original coordinate system to the 32-line lidar coordinate system, and then the transformed point cloud and the point cloud obtained by the 32-line lidar are jointly transformed to the vehicle coordinate system; the steps for transforming the point cloud from the 32-line lidar coordinate system to the vehicle coordinate system are as follows (the method for transforming the point cloud from the 16-line coordinate system to the 32-line coordinate system is the same as this step); specifically, as follows: Rotation matrix about the z-axis: , Rotation matrix about the y-axis: , Rotation matrix about the x-axis: , The combined rotation matrix is: , The translation vector is expressed as: , and the rotation matrix and the translation vector are combined into a transformation matrix of is: , where: ， ; , ; , ; Traverse each input point cloud, for each point , apply the affine transformation , and the transformed coordinates can be obtained.

[0020] S203. Merge the transformed point cloud data and output the merged point cloud.

[0021] Merge the point cloud data of the three lidars after coordinate transformation and store them uniformly in the output point cloud.

[0022] Reference Figure 3 , in step S300 of some embodiments of the present invention, extracting a straight line from the spliced point cloud data and determining the straight line type include: S301. Uniformly divide the point cloud data in the circumferential direction into multiple sectors; Specifically, according to the point cloud information (X, Y, Z, I, R, T) obtained by the lidar, where (X, Y, Z represent the coordinates of the point; I represents the intensity information returned by the laser; R is used to record which ring of the laser emitter collected this point; T represents the time when this point is collected by the lidar), all point clouds are registered in the corresponding fan-shaped area.

[0023] Rotate the lidar Divide the acquired frame of point cloud data into fan-shaped regions at a certain angle (for example, divide a frame of point cloud data into 360 fan-shaped regions, and the corresponding fan-shaped region indices are from 0 to 359). Then, according to the maximum number of lines in the spliced point cloud, divide each fan-shaped region into different grids (for example, 32 lines and 2 16-line splicings, the maximum number of lines is 32, so each fan-shaped region is further divided into 32 grids).

[0024] S302. Divide each sector into multiple grids, traverse each grid, and extract straight lines according to the coordinate information of the grid; Specifically, based on X and Y in the point cloud information, use piecewise trigonometric functions: , First, register the point cloud to the corresponding fan-shaped region, and then, according to the R value information of the point cloud, register the point cloud to the grids corresponding to each sector. At this time, each point cloud data has both sectorIdx (sector number) and binIdx (grid number).

[0025] Map the three-dimensional point cloud information to two-dimensional information

[0026] , represents the distance of the point cloud from the origin of the lidar, represents the height information of the point cloud. Calculate the mean point coordinates of each grid (n represents the total number of point clouds in the current grid): .

[0027] Then extract the straight line: Taking the sector as the processing unit, process each sector in turn. For each sector, first, according to the grid number, traverse each grid in the sector from small to large, and use the grid average value information to extract the straight line.

[0028] Take the first point traversed as the origin of the local coordinate system, and construct a local coordinate system using the first point and the second point. Assume that the two-dimensional information of the first point traversed is , and the two-dimensional information of the second point is , and the local coordinate system is defined as follows: ; ; ; ; ; ; Among them, and respectively represent the horizontal and vertical coordinate differences between two points. represents the distance between two points. , and respectively represent the sine value, cosine value and tangent value between two points. Predefined parameters , respectively represent the wall slope threshold and the ground slope threshold. If or , then a straight line can be extracted in this coordinate system. The average value point in the local coordinate system can be represented as : , Among them, and represent the two-dimensional information of the current point, represents the slope of the line formed by the current point and the previous point (abbreviation: current point slope). If the difference between the current point slope and the previous point slope is within a certain threshold range, the current point can be added to the currently extracted straight line. If the current point cannot be added to the currently extracted straight line, then judge the type of the current straight line and continue to traverse from the current position to extract a new straight line.

[0029] S303. Calculate the slope of the straight line according to the starting point and ending point of the extracted straight line; Specifically, assume that the starting point of the extracted straight line is , and the ending point is . Calculate the information of the straight line according to the following formula: , , , , Among them, is the slope of the extracted straight line, is its intercept.

[0030] S304. Judge the type of the straight line by comparing the slope of the straight line with the preset wall slope threshold and ground slope threshold respectively.

[0031] Specifically, if , (ground threshold), and the number of connected grids , then mark this line segment as the ground line; , and the number of connected grids , then mark this line segment as the wall line.

[0032] Traverse the original point cloud in the sector grid. If the height value of the original point cloud in the grid differs from the average height of all point clouds in the sector by (ground straight line deviation threshold), then mark this point as the ground point cloud; if the distance between the original point cloud in the grid and the wall straight line (wall straight line deviation threshold), then mark this point as the wall point cloud.

[0033] It can be understood that to accurately identify the wall point cloud, after obtaining the segmented wall point cloud data, it is necessary to perform format conversion and clustering on the incoming point cloud data. For example, use the EuclideanCluster Extraction algorithm in the PCL library to cluster the wall point cloud.

[0034] Reference Figure 4 and Figure 5 , in step S400 of some embodiments of the present invention, the state and pose estimation according to the clustering result includes: S401. Traverse each point cloud data of the clustering result and calculate the centroid of the point cloud; Specifically, traverse each point cloud data of the clustering result and calculate its centroid according to the following formula: , Obtain the minimum and maximum values of the Z axis in the point cloud and , for subsequent calculation of the height of the bounding box.

[0035] S402. Based on the rectangular fitting method of angle search, determine the optimal fitting angle for each clustering result; Specifically, in range, traverse with a step size of . For each traversed angle , let be the unit vector under , is the unit vector perpendicular to .

[0036] , Assume that there are A point cloud data, for each point cloud in the clustering result , where .

[0037] First, calculate the projection values of all point clouds on two basis vectors and respectively, and obtain the minimum and maximum projection values at the current angle and . The minimum and maximum projection values at the current angle are obtained.

[0038] , ; Then, calculate the distances of each point to the boundary and .

[0039]

[0040] Finally, calculate the reciprocals of the distances and , and accumulate them to obtain the total value , that is, obtain the compactness at the current angle .

[0041] , After all the angles in the range are traversed, the compactness corresponding to each angle can be obtained. Find the maximum value of all , and the corresponding angle is the optimal fitting angle .

[0042] S403. Calculate the center and size of the bounding box of each clustering result through the optimal fitting angle.

[0043] Recalculate the projection at the best angle , determine the minimum and maximum projection values of the bounding box, and calculate the center, size and orientation of the bounding box.

[0044] ; Projection value calculation, and obtain the minimum and maximum projection values:[[]] , ; For the four sides of the bounding box, they are represented as linear equations in the two-dimensional plane: ; ​

[0045] Calculate the coordinates of the diagonal vertices of the bounding box: , Calculate the coordinates of the center point of the bounding box: , Calculate the dimensions of the bounding box in the best-fitting direction:

[0046] Bounding box diagonal vector: , Bounding box length, width, height, and rotation angle:

[0047] , , .

[0048] Embodiment 2 Reference Figure 7 , the second aspect of the present invention provides a wall obstacle detection and shape estimation device 1, including: an acquisition module 11 for acquiring point cloud data of multiple lidars of a vehicle; a merging module 12 for merging and splicing the point cloud data through time synchronization and coordinate transformation; a segmentation module 13 for extracting lines and determining line types from the spliced point cloud data based on sector division; segmenting the wall in the point cloud data according to each line and its line type; an estimation module 14 for clustering the segmented wall point cloud data and performing state and pose estimation according to the clustering result.

[0049] Further, the merging module 12 includes: a synchronization unit for performing time synchronization on the point cloud data; a conversion unit for converting the time-synchronized point cloud data into the same coordinate system through coordinate transformation; a merging unit for merging the converted point cloud data and outputting the merged point cloud.

[0050] Embodiment 3 Reference Figure 8 , the third aspect of the present invention provides an electronic device, including: one or more processors; a storage device for storing one or more programs, when the one or more programs are executed by the one or more processors, enabling the one or more processors to implement the wall obstacle detection and shape estimation method of the first aspect of the present invention.

[0051] The electronic device 500 may include a processing device (such as a central processing unit, a graphics processing unit, etc.) 501, which may perform various appropriate actions and processes according to a program stored in the read-only memory (ROM) 502 or a program loaded from the storage device 508 into the random access memory (RAM) 503. In the RAM 503, various programs and data required for the operation of the electronic device 500 are also stored. The processing device 501, the ROM 502, and the RAM 503 are connected to each other through a bus 504. The input / output (I / O) interface 505 is also connected to the bus 504.

[0052] Generally, the following devices may be connected to the I / O interface 505: an input device 506 including, for example, a touch screen, a touchpad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; an output device 507 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; a storage device 508 including, for example, a hard disk, etc.; and a communication device 509. The communication device 509 may allow the electronic device 500 to communicate with other devices wirelessly or wirelessly to exchange data. Although Figure 8 an electronic device 500 with various devices is shown, it should be understood that it is not required to implement or have all the shown devices. Instead, more or fewer devices may be implemented or had. Figure 8 Each block shown in the figure may represent a device or, as needed, multiple devices.

[0053] In particular, according to embodiments of the present disclosure, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of the present disclosure include a computer program product that includes a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such an embodiment, the computer program can be downloaded and installed from a network via a communication device 509, or installed from a storage device 508, or installed from a ROM 502. When the computer program is executed by a processing device 501, the above-described functions defined in the methods of the embodiments of the present disclosure are performed. It should be noted that the computer-readable medium described in the embodiments of the present disclosure can be a computer-readable signal medium, a computer-readable storage medium, or any combination of the two. A computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the above. More specific examples of a computer-readable storage medium can include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In embodiments of the present disclosure, a computer-readable storage medium can be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In embodiments of the present disclosure, a computer-readable signal medium can include a data signal propagated in a baseband or as part of a carrier wave, in which computer-readable program code is carried. Such a propagated data signal can take many forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. A computer-readable signal medium can also be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device. The program code contained on a computer-readable medium can be transmitted using any appropriate medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination of the above.

[0054] The above computer-readable medium can be included in the above electronic device; or can exist separately without being assembled into the electronic device. The above computer-readable medium carries one or more computer programs, and when the above one or more programs are executed by the electronic device, the electronic device is caused to: Computer program code for performing the operations of the embodiments of the present disclosure may be written in one or more programming languages or combinations thereof. The programming languages include object-oriented programming languages such as Java, Smalltalk, C++, Python, and also include conventional procedural programming languages such as the "C" language or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, executed as a stand-alone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or it may be connected to an external computer (e.g., by connecting through the Internet using an Internet service provider).

[0055] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architectures, functions, and operations of systems, methods, and computer program products according to various embodiments of the present disclosure. In this regard, each block in the flowchart or block diagram may represent a module, a program segment, or a part of code that contains one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order than marked in the accompanying drawings. For example, two consecutive blocks shown may actually be executed substantially in parallel, and they may sometimes be executed in the reverse order, depending on the functions involved. It should be noted that each block in the block diagram and / or flowchart, and the combination of blocks in the block diagram and / or flowchart, may be implemented by a dedicated hardware-based system for performing the specified functions or operations, or may be implemented by a combination of dedicated hardware and computer instructions.

[0056] The above are only the preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for wall obstacle detection and shape estimation, characterized in that, Including: Obtain the point cloud data of multiple lidars of the vehicle; Through time synchronization and coordinate transformation, merge and splice the point cloud data; Based on sector division, extract straight lines from the spliced point cloud data and judge the straight line type; according to each straight line and its straight line type, segment the wall surface in the point cloud data; Cluster the segmented wall surface point cloud data, and perform state and pose estimation according to the clustering results.

2. The wall obstacle detection and shape estimation method according to claim 1, wherein The step of merging and splicing the point cloud data through time synchronization and coordinate transformation includes: Perform time synchronization on the point cloud data; Through coordinate transformation, transform the time-synchronized point cloud data into the same coordinate system; Merge the transformed point cloud data and output the merged point cloud.

3. The wall obstacle detection and shape estimation method according to claim 1, characterized in that, The step of extracting straight lines from the spliced point cloud data and judging the straight line type based on sector division includes: Uniformly divide the point cloud data in the circumferential direction into multiple sectors; Divide each sector into multiple grids, traverse each grid and extract straight lines according to the coordinate information of the grid; According to the starting point and ending point of the extracted straight line, calculate the slope of the straight line; compare the slope of the straight line with the preset wall slope threshold and ground slope threshold respectively to judge the straight line type.

4. The method for detecting and shape estimating wall obstacles according to claim 3, characterized in that The step of judging the straight line type by comparing the slope of the straight line with the preset wall slope threshold and ground slope threshold respectively includes: If the slope of the straight line is less than the ground slope threshold, and the intercept is limited to the preset value, and the connected grids are not less than 2, then judge the straight line as the ground line; If the slope of the straight line is greater than the preset wall slope threshold, and the connected grids are not less than 3, then judge the straight line as the wall line.

5. The wall obstacle detection and shape estimation method according to claim 4, characterized in that The step of segmenting the wall surface in the point cloud data according to each straight line and its straight line type includes: Traverse the point cloud in each sector grid. If the difference between the height value of the point cloud and the average height of the point cloud in the upper sector is within the preset value, then mark the point cloud as the ground point cloud; If the distance between the point cloud and the wall straight line is less than the wall straight line threshold, then mark the point cloud as the wall point cloud.

6. The wall obstacle detection and shape estimation method according to claim 1, characterized in that The step of performing state and pose estimation according to the clustering results includes: Traverse each point cloud data of the clustering results and calculate the centroid of the point cloud; Based on the rectangular fitting method of angle search, determine the optimal fitting angle of each clustering result; Through the optimal fitting angle, calculate the center and size of the bounding box of each clustering result.

7. A wall obstacle detection and shape estimation device, characterized in that, Including: An acquisition module for acquiring the point cloud data of multiple lidars of the vehicle; A merging module for merging and splicing the point cloud data through time synchronization and coordinate transformation; A segmentation module for extracting straight lines from the spliced point cloud data and judging the straight line type based on sector division; segmenting the wall surface in the point cloud data according to each straight line and its straight line type; An estimation module for clustering the segmented wall surface point cloud data and performing state and pose estimation according to the clustering results.

8. The wall obstacle detection and shape estimation device according to claim 7, wherein The merging module includes: A synchronization unit for performing time synchronization on the point cloud data; A transformation unit for transforming the time-synchronized point cloud data into the same coordinate system through coordinate transformation; A merging unit for merging the transformed point cloud data and outputting the merged point cloud.

9. An electronic device, comprising: One or more processors; A storage device for storing one or more programs, which, when executed by the one or more processors, cause the one or more processors to implement the wall obstacle detection and shape estimation method according to any one of claims 1 to 6.

10. A computer-readable medium having a computer program stored thereon, wherein, When the computer program is executed by a processor, it implements the wall obstacle detection and shape estimation method according to any one of claims 1 to 6.

Images