Passable map construction method for legged robot

Abstract

The invention discloses a passable map construction method for a leg-foot robot. The passable map construction method comprises the following steps: preprocessing acquired original point cloud information, inertia measurement information, latitude and longitude information and camera image information; performing error state Kalman filtering on the preprocessed point cloud information and inertial measurement information to obtain a rough pose of the robot; the precise pose of the robot is obtained through graph optimization; constructing a 2.5 D grid topographic map through extended Kalman filtering, and processing the 2.5 D grid topographic map to obtain a smooth and void-free grid topographic map; and extracting a normal vector in each unit grid in the grid topographic map, calculating the gradient, gradient change rate, roughness and passability in each unit grid, and maintaining a legged robot passable map in real time through extended Kalman filtering. According to the method, adaptive calculation of the passability of different terrains and materials in the real world is realized, so that the obtained passability graph is more credible.

Description

Technical Field

[0001] The present invention relates to the technical field of environment perception and map construction of legged robots, and in particular to a traversable map construction method for legged robots. Background Art

[0002] With the rapid development of intelligent mobile robot technology, the importance of traversable maps in robot autonomous navigation is increasing. In particular, legged robots that perform tasks in complex indoor and outdoor environments, such as humanoid robots and quadruped robots, face more complex and unpredictable terrain than wheeled robots and unmanned vehicles. They rely on real-time and accurate traversable terrain information to ensure the efficiency and safety of their operation. The traversable map of legged robots dynamically reflects the traversability of each fine area in the environment, enabling the robot to quickly identify obstacles and traversal costs, and select the best path, thereby avoiding being trapped in inaccessible areas and reducing energy consumption and time costs. Therefore, it is of great practical significance to develop a traversable map for legged robots that can be generated and updated in real time.

[0003] The mainstream methods for constructing traversable maps include extracting traversable areas from 3D maps and calculating grid traversability from 2.5D grid maps (mostly elevation maps). Both methods have their own advantages and disadvantages. 3D maps can provide rich environmental details and are suitable for accurate traversability judgment of complex terrains, but they require large computing resources, are complex to generate and maintain, and are inefficient in real-time applications. 2.5D grid maps have high computational efficiency, are easy to build and update, and are suitable for robots to make real-time traversability judgments in dynamic environments. However, they can only represent the height information of the terrain and cannot distinguish other characteristics of the terrain in detail, such as slope, ground material, etc., and the accuracy of the constructed traversable maps is low. To solve the above problems, relevant research has been carried out around the accurate and rapid construction of traversable maps. Kang Ke et al. (CN118392156A) calculated the height of each grid in the elevation map, the slope, curvature and roughness of the grid, obtained the initial passability cost map, and then expanded the obstacles in the initial passability cost map based on the edge of the rectangle approximated by the robot to construct the final passability cost map. Li Chao et al. (CN118329008A) improved the precision by integrating the depth camera data into the existing elevation map. The depth camera continuously collected the point cloud data of the terrain, which was pre-processed by the perception calculation unit and then registered with the previous frame point cloud to obtain the accurate relative pose between frames; then the elevation map was updated according to the relative pose to obtain the local terrain map; during the registration, the robot's positioning information was used to obtain the initial value of the relative pose of the two registered point clouds. Li Zhijun et al. (CN116147642A) provided a method and system for constructing a reachability map of a quadruped robot that integrates terrain and force, and established a two-dimensional grid map containing height information by integrating visual information and inertial information and set a height threshold to judge the terrain accessibility. Li Chuanpeng (CN117830991A) and others proposed a complex scene perception method and system for a quadruped robot based on multi-mode fusion. The terrain ahead was reconstructed through a deep neural network, and super-pixel images were obtained by clustering RGB images. Then, a multi-layer perceptron network was used to generate the passable area in the image coordinate system, and then a cost map was generated.

[0004] However, the above methods still do not solve the problem of holes and burrs in the actual use of elevation maps, and the calculation methods of the traversability of different terrains are not differentiated, resulting in the low practicality of the traversability map finally constructed. Summary of the invention

[0005] In view of the deficiencies in the prior art, the present invention proposes a method for constructing a traversable map for a legged robot.

[0006] The technical solution of the present invention includes the following contents:

[0007] A method for constructing a traversable map for a legged robot comprises the following steps:

[0008] S1. Obtain the original point cloud information around the robot through the laser radar, obtain the inertial measurement information of the robot body through the inertial measurement unit, obtain the longitude and latitude information of the robot body through the satellite signal receiver, and obtain the image information in front of the robot through the camera;

[0009] S2, preprocessing the original point cloud information, inertial measurement information, longitude and latitude information and camera image information obtained in step S1; wherein the preprocessing includes original point cloud information preprocessing, IMU pre-integration processing, UTM coordinate system projection of longitude and latitude information and image semantic information extraction;

[0010] S3, performing error state Kalman filtering on the point cloud information and inertial measurement information preprocessed in step S2 to obtain a rough posture of the robot; using the projected longitude and latitude information as a priori constraints and the rough posture of the robot as a factor to be optimized, obtaining the precise posture of the robot through a graph optimization method;

[0011] S4, constructing a 2.5D grid topographic map by using an extended Kalman filter method for the point cloud information preprocessed in step S2 and the precise position and posture of the robot obtained in step S3, and processing the topographic map by using a convolutional neural network to obtain a smooth grid topographic map without holes;

[0012] S5. For the grid topographic map obtained in step S4, extract the normal vector in each unit grid, and calculate the slope, slope change rate and roughness in each unit grid according to the normal vector and the precise position and posture of the robot obtained in step S3; calculate the passability in each unit grid in combination with the image semantic information obtained in step S2, and maintain a legged robot passable map in real time through extended Kalman filtering.

[0013] Furthermore, the step S2 includes the following sub-steps:

[0014] S21, preprocessing of original point cloud information: performing outlier filtering and motion distortion correction on the original point cloud information, and removing dynamic point clouds to obtain credible point clouds;

[0015] S22, IMU pre-integration processing: according to the IMU angular velocity and acceleration data, the relative motion between two frames during measurement is calculated to obtain the robot posture change;

[0016] S23, UTM coordinate system projection of longitude and latitude information: The satellite receiver interprets the message sent by the global satellite navigation and positioning system GNSS as longitude and latitude information based on the WGS84 geocentric coordinate system, and then converts the longitude and latitude information into the UTM coordinate system, that is, the XYZ three-axis coordinate information in the robot's local coordinate system;

[0017] S24, Image semantic information extraction: The features of the original image information are extracted through a convolutional neural network, and the fully connected layer is used to output the category and bounding box of the object to achieve semantic information extraction.

[0018] Furthermore, the step S3 includes the following sub-steps:

[0019] S31, inferring the initial posture of the current frame according to the credible point cloud obtained in step S2 and the posture change between two frames of the robot obtained by IMU pre-integration; transforming the initial posture to the coordinate system of the previous frame, and constructing an error model through the geometric registration relationship between the two frames of point clouds to obtain the optimal robot posture state, that is, the rough posture of the robot;

[0020] S32. Using a graph optimization method, with the rough posture of the robot as the vertex and the longitude and latitude information obtained in step S2 as the constraint edge, a graph model is constructed to solve the graph optimization problem and obtain the precise posture state of the robot.

[0021] Furthermore, the step S4 includes the following sub-steps:

[0022] S41, according to the credible point cloud obtained in step S2 and the precise posture state of the robot obtained in step S3, a 2.5D grid terrain map is obtained by extending the Kalman filter; wherein each grid contains a number of points, an average height value and a variance estimate obtained by the Kalman filter;

[0023] S42, the raster topographic map obtained in step S41 is vectorized and input into a GAN network, global features of the image are extracted by downsampling, the spatial resolution of the image is gradually restored by upsampling, and a vectorized smoothed and completed raster topographic map is output.

[0024] Furthermore, the step S5 includes the following sub-steps:

[0025] S51, performing plane fitting on a number of point clouds in each unit grid according to the grid topographic map obtained in step S4, and taking the normal vector of the fitted plane as the normal vector of the unit grid;

[0026] S52, calculating the slope and roughness of each grid according to the normal vector of each grid extracted in step S51 and the normal vectors in its neighboring grids;

[0027] S53, weighted calculation of the passability of each cell according to the parameters of each unit grid and its surrounding grids obtained in step S52.

[0028] Furthermore, in step S53, the weight parameters in the weighted calculation of the passability are determined by the image semantic information obtained in step S2, and the weights of the parameters corresponding to different semantic areas are different.

[0029] In summary, the beneficial effects of the present invention are as follows: for robots that perform tasks in complex indoor and outdoor environments, a method for generating and updating a 2.5D legged robot traversable map in real time using multi-sensor information is invented. On the basis of retaining information related to traversability, the present invention chooses to construct a 2.5D grid map with a smaller amount of information. Compared with a three-dimensional traversable map, it is lighter, consumes less computing and storage costs, and can be deployed on a robot for real-time operation. The present invention innovatively uses an adversarial neural network to solve the problem of holes and burrs in elevation maps in practical applications, and provides more accurate and realistic robot surrounding information for subsequent calculations of traversability. When calculating traversability, the present invention innovatively proposes to determine the weights of each information parameter in the traversability calculation formula in different semantic areas based on semantic information, and realizes adaptive calculation of traversability for different terrains and materials in the real world, making the obtained traversable map more credible. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 A framework diagram of a method for constructing a traversable map for a legged robot according to the present invention;

[0031] Figure 2 A diagram summarizing the coordinate relationship of an embodiment of the present invention;

[0032] Figure 3 The flowchart of the algorithm of the adversarial neural network used to complete the elevation map of the present invention. DETAILED DESCRIPTION

[0033] The present invention is described in detail below in conjunction with specific embodiments. The embodiments described below are exemplary and are only used to explain the present invention, and should not be construed as limiting the present invention. It should be noted that, for those of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present invention. These all belong to the protection scope of the present invention.

[0034] Robots that perform tasks in complex indoor and outdoor environments rely on real-time and accurate traversability information to ensure the efficiency and safety of their operations. This paper proposes a traversability map construction method for legged robots to address the problems of holes and burrs in elevation maps and the inability to distinguish between traversability calculation methods for different terrains during actual use. Figure 1 The present invention mainly includes five steps: obtaining multi-sensor information, preprocessing the original information, calculating the robot posture, constructing a grid terrain map and calculating the passability, which will be described in detail below:

[0035] S1. Obtain multi-sensor information. Obtain the original point cloud information around the robot through the laser radar, obtain the inertial measurement information of the robot body through the inertial measurement unit, obtain the longitude and latitude information of the robot body through the satellite signal receiver, and obtain the image information in front of the robot through the camera. In this embodiment, Figure 2 As shown in the figure, the relationship between multiple coordinate systems is described, including the robot body coordinate system B, the IMU (Inertial Measurement Unit) center coordinate system I, the laser radar coordinate system L, the camera coordinate system C, the satellite positioning receiver coordinate system G, the UTM (Universal Transverse Mercator) grid world coordinate system W and the robot odometer coordinate system O. Among them, the conversion matrix between the coordinate systems is given, such as the conversion matrix from the UTM world coordinate system to the robot odometer coordinate system O is Generally, the inertial measurement unit IMU is installed at the center of mass of the robot, that is, the robot body coordinate system coincides with the origin of the IMU center coordinate system I, and the directions of each axis are the same.

[0036] S2. Preprocessing the original information. Preprocessing the original point cloud information, inertial measurement information, longitude and latitude information, and camera image information obtained in step S1, wherein the preprocessing includes preprocessing of the original point cloud information, IMU pre-integration processing, UTM coordinate system projection of the longitude and latitude information, and extraction of image semantic information.

[0037] S21. Preprocessing method for original point cloud information: first, outlier filtering is performed, that is, a fixed search radius is set, and the number of neighboring points within the radius of each point in the point cloud is checked. If the number of neighboring points is lower than the preset threshold, the point is identified as an outlier and filtered out. Then motion distortion correction is performed, that is, by calculating the position change of the laser radar each time it scans, the point cloud data is converted to the same coordinate system (robot mileage coordinate system O) to correct motion distortion; and the dynamic point cloud is removed to obtain a credible point cloud.

[0038] The specific formula for motion distortion correction is:

[0039]

[0040] Among them, t i ,t j ,t k Respectively represent the time corresponding to the three frames, and Represents the robot's posture change from time k to time i and from time j to time i.

[0041] S22. Preprocessing method for inertial unit information: Calculate the relative position change of the robot between two adjacent frames according to the IMU pre-integration formula:

[0042]

[0043] Among them, i, j, and k represent three frames respectively, a k and ω k is the acceleration and angular velocity of the kth frame in the I coordinate system, and is the bias of the accelerometer and gyroscope, n a and n g represents the noise of the accelerometer and gyroscope. ij is the change in robot position from frame i to frame j, Δv ij is the speed change of the robot from the i-th frame to the j-th frame, Δq ij is the posture change of the robot from the i-th frame to the j-th frame.

[0044] S23. Preprocessing method for satellite positioning information: The satellite receiver interprets the message sent by the global satellite navigation positioning system (GNSS) as longitude and latitude information based on the WGS84 geocentric coordinate system. This method converts the longitude and latitude information into the UTM coordinate system, that is, the XYZ three-axis coordinate information in the robot's local coordinate system.

[0045] Specifically, the satellite positioning information gives the longitude and latitude and altitude h, converted to Cartesian coordinates (X GNSS ,Y GNSS ,Z GNSS ):

[0046]

[0047] Among them, the positive direction of the X axis is due east, the positive direction of the Y axis is due north, and the positive direction of the Z axis is vertically upward.

[0048] S24. Method for preprocessing image information: Method for extracting image semantic information: Extract the features of the original image information through a convolutional neural network, and use a fully connected layer to output the category and bounding box of the object to achieve semantic information extraction.

[0049] Specifically, the acquired image information is segmented and detected using the YOLOv5 visual 2D target detection algorithm to identify semantic information such as cement roads, grass, and steps, and its detection frame is projected to the point cloud coordinate system L according to the pyramid rule and BEV (Bird's-Eye-View) rule.

[0050] S3, calculate the robot posture. Perform error state Kalman filtering on the point cloud information and inertial measurement information preprocessed in step S2 to obtain the rough posture of the robot. Take the projected longitude and latitude information as the prior constraints, and the rough posture of the robot as the factor to be optimized, and obtain the precise posture of the robot through the graph optimization method. The following will be explained in detail:

[0051] S31. Method for obtaining the rough posture of the robot: first, the nominal true value of the current frame is inferred based on the credible point cloud in step S2 and the posture change between two frames of the robot obtained by IMU pre-integration. That is, use the pose x of the previous frame k Add the relative posture change δx of the robot between two adjacent frames calculated in step S22.

[0052]

[0053] The nominal truth value is completely obtained by the IMU, and drift may occur when the robot moves a longer distance. The next step is to multiply the nominal truth value by the transformation matrix to transform it to the coordinate system of the previous frame, and construct an error model through the geometric registration relationship between the two frames of point clouds to obtain the optimal robot posture state. This step is also called error state Kalman filtering, which can be expressed as considering the nominal truth value obtained from the IMU state propagation model. The error state δx to be corrected and the error model of point cloud registration. The optimization problem can be expressed as follows:

[0054]

[0055] Among them, ||·|| represents the Mahalanobis distance norm, J k is the Jacobian matrix of the laser radar measurement noise, which is generally a unit matrix, M k is the measurement noise of the lidar, which is set by the manual parameters of the lidar. The specific form of f(·) is as follows:

[0056]

[0057] By solving this optimization problem, we can get the optimal error state δx, and further get the real position x of the robot in the current frame. k+1 .

[0058] S32. Method for obtaining the precise posture of the robot: The above process uses the optimization method of tightly coupled LiDAR and IMU to obtain the rough posture of the robot, and then integrates the satellite positioning information through the graph optimization method. Specifically, a graph model is constructed, whose vertices are the rough postures of the robot at different times, and the edges are the constraints formed by the UTM coordinates obtained in step S2. The goal of graph optimization is to optimize the structure of the entire graph by minimizing the errors between these vertices, and then obtain the most accurate posture of the robot. The specific formula can be expressed as:

[0059]

[0060] Among them, X i and X j is the robot pose, the initial value is the rough pose obtained in the previous step, Z ij is the relative observation value between the two frames obtained from the UTM coordinate system, f(X i ,X j ) is the predicted relative pose calculated based on the pose vertices, Ω i,j is the covariance matrix, which is used to represent the confidence of the sensor measurement and is usually a symmetric positive definite matrix.

[0061] S4, constructing a grid topographic map. The preprocessed point cloud information obtained in step S2 and the precise position of the robot obtained in step S3 are used to construct a 2.5D grid topographic map by using an extended Kalman filter method, and the topographic map is processed by a convolutional neural network to obtain a smooth grid topographic map without holes. The following will be described in detail:

[0062] S41. Construct a 2.5D grid terrain map by using the extended Kalman filter method: According to the coordinate system transformation relationship described in step S1 and the precise position and posture of the robot obtained in step S3, the height value of the surrounding terrain centered on the robot body can be obtained from the trusted point cloud information in step S2. After the trusted point cloud is rasterized, the height of each unit grid takes the average value of the point cloud height belonging to the grid as the initial value, and is updated after receiving new point cloud information. During the update process, the height data in the newly received trusted point cloud is fused with the existing elevation estimate through the extended Kalman filter. The introduction of this filtering method can improve the measurement accuracy while processing measurement noise. After each measurement update, each grid point of the terrain contains an estimated value and a covariance matrix to represent its uncertainty. When the robot moves, the height estimates of all grid points will be updated according to the changes in the robot state to ensure the consistency of the elevation map with the actual terrain. The update formula is:

[0063]

[0064] Where h is the height of the grid cell, is the variance of the height of this grid cell, p is the average height of this grid cell obtained from the new credible point cloud, is the variance of the height of the grid cell obtained from the new credible point cloud. Generally, the height value obtained from the credible point cloud is considered to follow a Gaussian distribution, that is:

[0065]

[0066] The present invention uses a model that is proportional to the square of the measured distance to represent the variance of the height value. Among them, α is a custom scale parameter, and d is the distance from the midpoint of the grid to the lidar.

[0067] S42, for a method of obtaining a smooth raster topographic map without holes: the raster topographic map obtained in step S41 is vectorized and input into a GAN network, the global features of the raster map are extracted by downsampling, the spatial resolution of the raster map is gradually restored by upsampling, and a vectorized smoothed and completed raster topographic map is output. Figure 3 As shown, the following is a detailed description:

[0068] First, the input grid passes through a convolutional layer with 64 filters of size 5x5 and stride 1. The convolution operation slides the filters over the grid to extract low-level features. After that, a batch normalization layer normalizes the convolution output to ensure numerical stability during training. Next, the ReLU activation function sets all negative values ​​to 0 to introduce nonlinear features.

[0069] The next two convolutional layers gradually deepen the depth of the network and increase the complexity of feature extraction. Convolutional layer 2 downsamples the spatial dimension through 128 3x3 filters and convolution operations with a stride of 2, halving the size of the feature map. Convolutional layer 3 continues to extract and process high-level features of the grid through 128 3x3 filters and convolutions with a stride of 1. Each layer is standardized and nonlinearly mapped through batch normalization and ReLU activation functions.

[0070] After multiple layers of convolution, the network enters the dilated convolution layer. The purpose of dilated convolution is to expand the receptive field of the convolution kernel without increasing the computational complexity, thereby capturing a wider range of contextual information. This is crucial for the global structure of the grid map. Dilated convolution is also equipped with batch normalization and ReLU activation.

[0071] After the dilated convolution, the network recovers the details and texture information of the grid through convolution layer 4. The convolution operation of this layer continues to extract features at a higher level, and the number of filters is kept at 256 to capture enough details.

[0072] In order to restore the feature map to the same spatial resolution as the input map, the network uses deconvolution operations. Deconvolution layers 1 and 2 use 128 and 64 filters respectively, and upsample the feature map by deconvolution operations with a stride of 2, doubling its size.

[0073] The core task of the network is to extract and generate high-resolution, reliable raster images from the low-level features of the input raster image. Through multi-layer convolution, dilated convolution and deconvolution operations, the network can capture global context information while retaining the details of the original terrain map, and finally generate a smoothed and completed raster terrain map similar to the input raster.

[0074] S5. Calculate the passability. For the grid terrain map obtained in step S4, extract the normal vector in each unit grid, and calculate the slope, slope change rate and roughness in each unit grid according to the normal vector and the precise position of the robot obtained in step S3. Combined with the image semantic information obtained in step S2, calculate the passability in each unit grid, and maintain a legged robot passable map in real time through the extended Kalman filter. The following will be explained in detail:

[0075] S51, extract the normal vector of the unit grid. In the constructed grid map, there are several point clouds in each grid, and a plane is fitted through these point clouds. Specifically, the present invention uses the least squares method to fit a plane. The normal vector of the fitted plane is the normal vector of the grid. Finally, the normal vector is normalized to a unit normal vector n, which is convenient for subsequent analysis and processing.

[0076] S52: Calculate the slope and roughness of each grid according to the normal vector obtained in step S51. The specific calculation formula is as follows:

[0077] s=arccos(n,n z )

[0078]

[0079] Among them, s is the slope, r is the roughness, n is the grid normal vector, and n z is the ground normal vector, n i is the neighborhood grid normal vector, and m is the custom neighborhood size.

[0080] S53, according to the normal vector slope and roughness of each unit grid and its surrounding grids obtained in step S52, weighted calculation of each unit cell passability. The specific calculation formula is as follows:

[0081]

[0082] Among them, trav is the traversability, and its value range is [0,1]. a and b are custom scale parameters, whose values ​​are determined by the semantic information obtained in step S24, and different semantic areas correspond to different parameters. This design enables the robot to autonomously distinguish different scenes such as stairs, grass, and concrete when calculating traversability, which is more valuable in practical applications.

[0083] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for constructing a traversable map for a legged robot, characterized in that: The following steps are involved: S1. Obtain the original point cloud information around the robot through the laser radar, obtain the inertial measurement information of the robot body through the inertial measurement unit, obtain the longitude and latitude information of the robot body through the satellite signal receiver, and obtain the image information in front of the robot through the camera; S2, preprocessing the original point cloud information, inertial measurement information, longitude and latitude information and camera image information obtained in step S1; wherein the preprocessing includes original point cloud information preprocessing, IMU pre-integration processing, UTM coordinate system projection of longitude and latitude information and image semantic information extraction; S3, performing error state Kalman filtering on the point cloud information and inertial measurement information preprocessed in step S2 to obtain a rough posture of the robot; using the projected longitude and latitude information as a priori constraints and the rough posture of the robot as a factor to be optimized, obtaining the precise posture of the robot through a graph optimization method; S4, constructing a 2.5D grid topographic map by using an extended Kalman filter method for the point cloud information preprocessed in step S2 and the precise position and posture of the robot obtained in step S3, and processing the topographic map by using a convolutional neural network to obtain a smooth grid topographic map without holes; S5. For the grid topographic map obtained in step S4, extract the normal vector in each unit grid, and calculate the slope, slope change rate and roughness in each unit grid according to the normal vector and the precise position and posture of the robot obtained in step S3; calculate the passability in each unit grid in combination with the image semantic information obtained in step S2, and maintain a legged robot passable map in real time through extended Kalman filtering.

2. The method for constructing a traversable map for a legged robot according to claim 1, characterized in that: The step S2 comprises the following sub-steps: S21, preprocessing of original point cloud information: performing outlier filtering and motion distortion correction on the original point cloud information, and removing dynamic point clouds to obtain credible point clouds; S22, IMU pre-integration processing: according to the IMU angular velocity and acceleration data, the relative motion between two frames during measurement is calculated to obtain the robot posture change; S23, UTM coordinate system projection of longitude and latitude information: The satellite receiver interprets the message sent by the global satellite navigation and positioning system GNSS as longitude and latitude information based on the WGS84 geocentric coordinate system, and then converts the longitude and latitude information into the UTM coordinate system, that is, the XYZ three-axis coordinate information in the robot's local coordinate system; S24, Image semantic information extraction: The features of the original image information are extracted through a convolutional neural network, and the fully connected layer is used to output the category and bounding box of the object to achieve semantic information extraction.

3. The method for constructing a traversable map for a legged robot as claimed in claim 2, characterized in that: The step S3 comprises the following sub-steps: S31, inferring the initial posture of the current frame according to the credible point cloud obtained in step S2 and the posture change between two frames of the robot obtained by IMU pre-integration; transforming the initial posture to the coordinate system of the previous frame, and constructing an error model through the geometric registration relationship between the two frames of point clouds to obtain the optimal robot posture state, that is, the rough posture of the robot; S32. Using a graph optimization method, with the rough posture of the robot as the vertex and the longitude and latitude information obtained in step S2 as the constraint edge, a graph model is constructed to solve the graph optimization problem and obtain the precise posture state of the robot.

4. The method for constructing a traversable map for a legged robot according to claim 1, characterized in that: The step S4 comprises the following sub-steps: S41, according to the credible point cloud obtained in step S2 and the precise posture state of the robot obtained in step S3, a 2.5D grid terrain map is obtained by extending the Kalman filter; wherein each grid contains a number of points, an average height value and a variance estimate obtained by the Kalman filter; S42, the raster topographic map obtained in step S41 is vectorized and input into a GAN network, global features of the image are extracted by downsampling, the spatial resolution of the image is gradually restored by upsampling, and a vectorized smoothed and completed raster topographic map is output.

5. The method for constructing a traversable map for a legged robot according to claim 1, characterized in that: The step S5 comprises the following sub-steps: S51, performing plane fitting on a number of point clouds in each unit grid according to the grid topographic map obtained in step S4, and taking the normal vector of the fitted plane as the normal vector of the unit grid; S52, calculating the slope and roughness of each grid according to the normal vector of each grid extracted in step S51 and the normal vectors in its neighboring grids; S53, weighted calculation of the passability of each cell according to the parameters of each unit grid and its surrounding grids obtained in step S52.

6. The method for constructing a traversable map for a legged robot as claimed in claim 5, characterized in that: In step S53, the weight parameters in the weighted calculation of the passability are determined by the image semantic information obtained in step S2, and the weights of the parameters corresponding to different semantic areas are different.

Images