Unstructured terrain environment-oriented trafficability prediction method

Abstract

The invention discloses a trafficability prediction method for an unstructured terrain environment, and is applied to the technical field of vehicle traffic capacity prediction. According to the method, a three-dimensional grid map is constructed based on a multi-source sensor, and geometric and semantic combined representation of the environment is formed by combining semantic terrain classification and elevation structure feature extraction. By comparing a control instruction with the motion response of the vehicle in the actual terrain, a traffic capacity coefficient is constructed as a supervision signal, and a conditional variational auto-encoder is trained to model traffic capacity coefficient distribution and uncertainty thereof under different terrain conditions. And further extracting intermediate layer features from the semantic recognition network, and establishing a category statistical model for recognizing terrain categories which do not appear in the training set. And finally, fusing the traffic capacity confidence and the terrain credibility, outputting a grid-level traffic probability, and realizing trafficability prediction in the unstructured environment.

Description

Technical Field

[0001] This invention relates to the field of vehicle capacity prediction technology, and more specifically to a method for predicting accessibility in unstructured terrain environments. Background Technology

[0002] In structured road environments, road boundaries are clear, pavement materials are uniform, and geometric properties are relatively stable, allowing vehicles to make highly deterministic and predictable judgments about passability. However, in tasks such as field operations, disaster response, and complex terrain surveying, special vehicles must operate in unstructured environments. These scenarios feature diverse terrains, complex geometric structures, and a lack of available prior map information, resulting in high uncertainty and posing significant challenges to passability judgment. Existing mainstream methods largely rely on semantic segmentation and rule mapping strategies to directly convert identified terrain categories into passability labels. While this method has some practicality, it ignores the differences in physical response between different terrains within the same semantic category during actual driving, making it difficult to comprehensively reflect the true passability of the ground. Furthermore, existing models generally lack the ability to discriminate their own cognitive boundaries. When faced with novel terrain categories not seen in the training set, they are prone to high-confidence misjudgments, leading to path planning failures or vehicles entering dangerous areas. Therefore, providing a passability prediction method for unstructured terrain environments is a problem that urgently needs to be solved by those skilled in the art. Summary of the Invention

[0003] In view of this, the present invention provides a accessibility prediction method for unstructured terrain environments, which can characterize the physical accessibility of the terrain and the cognitive boundary of the model, and improve the reliability of special vehicles' passage decisions in complex and highly uncertain environments and their adaptability to unknown terrain.

[0004] To achieve the above objectives, the present invention provides the following technical solution: A accessibility prediction method for unstructured terrain environments includes the following steps: S1. Collect terrain information using multi-source sensors, fuse the terrain information to estimate the pose transformation matrix at each time step, construct a grid map, and predict the terrain category of each grid in the grid map based on a semantic segmentation network that is based on image point cloud collaboration. S2. Establish a three-dimensional semantic map and calculate the elevation feature vector of each two-dimensional grid based on the point cloud height information of each grid in the grid map; S3. During the multi-source sensor acquisition process, record the actual speed vector of the vehicle in each grid, define and calculate the traffic capacity coefficient to characterize the vehicle's ability to execute control commands in complex terrain; S4. Model the vehicle capacity coefficient as a conditional random variable under terrain conditions, and use a conditional variational autoencoder to model the conditional distribution. S5. During the actual operation of the vehicle, a Mahalanobis distance recognition mechanism is introduced to determine whether the currently observed terrain belongs to the learned distribution support domain. S6, combined with the terrain accessibility distribution of vehicles S4 and the terrain perception credibility of vehicle S5, comprehensively assesses the accessibility of the terrain at the grid.

[0005] Optionally, terrain information acquired from multiple sensors includes: lidar scan point cloud sequences. , Indicates the first t Frame point cloud; camera image sequence , For the first t Frame images; IMU measurement sequences , Indicates the first t Acceleration and angular velocity at frame time.

[0006] Optionally, S1 is as follows: The pose transformation matrix at each time step is estimated by fusing information from LiDAR, camera vision, and IMU using a factor graph-based nonlinear optimization method. : ; In the formula, Represents the rotation matrix. Representing the three-dimensional Euclidean group, Represents the three-dimensional orthogonal group. Represents the translation vector, which will be used to register the point cloud. Projecting the map onto a world coordinate system, a dense map is constructed; a voxel-based 3D occupancy raster mapping method is used to generate a raster map, OctoMap. A semantic segmentation network based on image point cloud collaboration is introduced to segment the image... With the corresponding projected point cloud The input is fed into a semantic segmentation network to predict the terrain category for each raster: ; in, It is a grid semantic prediction labels, This represents a predefined set of terrain semantic categories. This represents a semantic segmentation network.

[0007] Optionally, S2 is as follows: Based on the camera's intrinsic parameter matrix and pose transformation matrix The semantically predicted labels are back-projected onto the 3D map coordinate system to assign semantic attributes to the voxel raster, thereby establishing a 3D semantic map. : ; In the formula, x represents the coordinate vector. Represents depth values, based on the OctoMap raster map. The point cloud height information of each cell in the two-dimensional grid is used to calculate the grid. Elevation eigenvectors: ; Among them, is the grid. The elevation eigenvector represents the geometric features of the Earth's surface. For all falling into the grid The collection of point clouds, This represents the maximum height of all point clouds at the corresponding location. The average of all heights. For the standard deviation of all heights, n This represents the number of point clouds.

[0008] Optionally, S3 specifically refers to: Record vehicle movement within the grid during multi-source sensor acquisition. actual velocity vector : ; In the formula, Indicates linear velocity. Represents angular velocity, and records the speed at which the control system issues the desired control command. : ; In the formula, To achieve the desired control command linear velocity, Define the traffic capacity coefficient to represent the desired control command angular velocity. To characterize the vehicle's ability to execute control commands in complex terrain: ; Traffic capacity coefficient This indicates the strength of a vehicle's response to desired control under certain terrain conditions. A larger value indicates that the terrain does not significantly hinder movement, while a smaller value indicates that movement is restricted.

[0009] Optionally, S4 specifically refers to: Vehicle capacity coefficient Modeling for terrain conditions To characterize this uncertainty, conditional random variables are used to learn conditional probability distributions. A conditional variational autoencoder is used to model the conditional distribution. The conditional variational autoencoder consists of three parts: a conditional encoder module, an encoder network, and a decoder network. A training process sample tripartite dataset is constructed using the collected data. ; The conditional encoding module will input the condition. Joint encoding as a conditional representation vector The encoder network is defined as a posterior distribution: ; In the formula, As latent variables, This represents the encoder network parameters, which map observations to terrain conditions using Gaussian distribution parameters. To construct an approximate posterior, during the generation process, latent variables are first sampled from the standard normal distribution. Then with the condition vector The data is input together into the decoder network to predict throughput. The conditional probability distribution is obtained by using a Gaussian distribution as the basis. The modeling and decoding methods are as follows: ; in, and For the output of the decoder network, These are the decoder network parameters.

[0010] Optional, S5 specifically includes: Extracting deep feature representations from the last stage of the encoder in a semantic segmentation network. High-dimensional semantic feature embeddings used to describe the current image region: ; In the formula, For feature extraction function, For semantic segmentation networks, The input is a raster parameter vector; during the training phase, for each known semantic terrain category... Collect the corresponding deep feature set from the training set. The mean vector of the statistical distribution With covariance matrix : ; In the formula, To determine the number of semantic terrain categories, construct a Gaussian model for the features of each terrain category: : During the reasoning phase, for the grid Extracting the intermediate layer feature vector Calculate the Mahalanobis distance with each semantic class Gaussian model: ; The category with the smallest distance among all categories is taken as the discrimination result: ; The value of the discrimination result is used to assess whether the current terrain falls within the known distribution range of the training set.

[0011] Optional, S6 specifically includes: Gaussian distribution modeled as a conditional distribution Mapping to pass probability, setting a safety threshold To distinguish whether a vehicle can respond to control commands, the probability of a vehicle successfully passing through is assessed by calculating the right-tail cumulative distribution function of the capacity distribution: ; In the formula, Traffic capacity coefficient Exceeding the safety threshold The probability; defining the credibility decay factor. Markovnikov distance Integrate into a unified scoring system: ; in, As a moderating term, it controls for the impact of the degree of cognitive deviation on the final score. It is the Mahalanobis distance between the current terrain feature and the mean of its nearest semantic category, reflecting its typicality in the training distribution; finally, combining the probability of the accessibility distribution with the cognitive credibility of the terrain, the terrain accessibility probability is defined. : ; Terrain accessibility probability Simultaneously, the actual impact of terrain on vehicle responsiveness and the credibility of the perception model are considered.

[0012] As can be seen from the above technical solution, compared with the prior art, the present invention provides a accessibility prediction method for unstructured terrain environments, which has the following beneficial effects: 1. This invention introduces the actual motion response information of vehicles under different terrain conditions, constructs a traffic capacity coefficient as a supervision signal, combines elevation features and semantic labels to train a traffic capacity prediction model, introduces a probabilistic modeling mechanism to express the uncertainty of the prediction results, and improves the accuracy of traffic capacity assessment. 2. This invention extracts features from the intermediate layer of the semantic recognition network and combines them with the feature distribution model of terrain categories to determine whether the current input belongs to the known terrain in the training set, thereby reducing the risk of the model misjudging unfamiliar terrain and improving the robustness and security of the system in unknown environments. 3. This invention constructs a joint evaluation mechanism for accessibility confidence and terrain recognition credibility, integrating the consistency between vehicle motion response information and model's current terrain perception into accessibility judgment, outputting grid-level accessibility probability, and improving the system's risk perception and path evaluation capabilities in complex environments. Attached Figure Description

[0013] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0014] Figure 1 This is a flowchart of the accessibility prediction method for unstructured terrain environments according to the present invention. Figure 2 Flowchart for constructing the multi-source sensor information-driven accessibility dataset of this invention; Figure 3 This is a flowchart of the probability modeling of traffic capacity based on the conditional variational autoencoder of the present invention. Figure 4 This is a flowchart of the terrain recognition matching judgment based on Mahalanobis distance according to the present invention. Figure 5 This is a flowchart of the probability calculation for considering terrain uncertainty in this invention. Detailed Implementation

[0015] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0016] This invention discloses a accessibility prediction method for unstructured terrain environments, such as... Figure 1 As shown, it includes the following steps: S1. Collect terrain information using multi-source sensors, fuse the terrain information to estimate the pose transformation matrix at each time step, construct a grid map, and predict the terrain category of each grid in the grid map based on a semantic segmentation network that is based on image point cloud collaboration. S2. Establish a three-dimensional semantic map and calculate the elevation feature vector of each two-dimensional grid based on the point cloud height information of each grid in the grid map; S3. During the multi-source sensor acquisition process, record the actual speed vector of the vehicle in each grid, define and calculate the traffic capacity coefficient to characterize the vehicle's ability to execute control commands in complex terrain; S4. Model the vehicle capacity coefficient as a conditional random variable under terrain conditions, and use a conditional variational autoencoder to model the conditional distribution. S5. During the actual operation of the vehicle, a Mahalanobis distance recognition mechanism is introduced to determine whether the currently observed terrain belongs to the learned distribution support domain. S6, combined with the terrain accessibility distribution of vehicles S4 and the terrain perception credibility of vehicle S5, comprehensively assesses the accessibility of the terrain at the grid.

[0017] Furthermore, the terrain information acquired by the multi-source sensors includes: lidar scan point cloud sequences. , Indicates the first t Frame point cloud; camera image sequence , For the first t Frame images; IMU measurement sequences , Indicates the first t Acceleration and angular velocity at frame time.

[0018] Furthermore, such as Figure 2 As shown, S1 is specifically: The pose transformation matrix at each time step is estimated by fusing information from LiDAR, camera vision, and IMU using a factor graph-based nonlinear optimization method. : ; In the formula, Represents the rotation matrix. Representing the three-dimensional Euclidean group, Represents the three-dimensional orthogonal group. Represents the translation vector, which will be used to register the point cloud. Projecting the map onto a world coordinate system, a dense map is constructed; a voxel-based 3D occupancy raster mapping method is used to generate a raster map, OctoMap. A semantic segmentation network based on image point cloud collaboration is introduced to segment the image... With the corresponding projected point cloud The input is fed into a semantic segmentation network to predict the terrain category for each raster: ; in, It is a grid semantic prediction labels, This represents a predefined set of terrain semantic categories. This represents a semantic segmentation network.

[0019] Furthermore, such as Figure 2 As shown, S2 is specifically: Based on the camera's intrinsic parameter matrix and pose transformation matrix The semantically predicted labels are back-projected onto the 3D map coordinate system to assign semantic attributes to the voxel raster, thereby establishing a 3D semantic map. : ; In the formula, x represents the coordinate vector. Represents depth values, based on the OctoMap raster map. The point cloud height information of each cell in the two-dimensional grid is used to calculate the grid. Elevation eigenvectors: ; Among them, is the grid. The elevation eigenvector represents the geometric features of the Earth's surface. For all falling into the grid The collection of point clouds, This represents the maximum height of all point clouds at the corresponding location. The average of all heights. For the standard deviation of all heights, n This represents the number of point clouds.

[0020] Furthermore, such as Figure 2 As shown, S3 specifically refers to: Record vehicle movement within the grid during multi-source sensor acquisition. actual velocity vector : ; In the formula, Indicates linear velocity. This represents the angular velocity, which can be derived from the pose transformation between two frames, recording the velocity of the desired control command given by the control system. : ; In the formula, To achieve the desired control command linear velocity, Define the traffic capacity coefficient to represent the desired control command angular velocity. To characterize the vehicle's ability to execute control commands in complex terrain: ; Traffic capacity coefficient This indicates the strength of a vehicle's response to desired control under certain terrain conditions. A larger value indicates that the terrain does not significantly hinder movement, while a smaller value indicates that movement is restricted.

[0021] In this embodiment of the invention, considering the instability of low-speed division in practical systems, a robust form is introduced: ; In the formula, To ensure stability of the denominator parameter, it takes the value of a tiny integer.

[0022] Furthermore, such as Figure 3 As shown, S4 specifically refers to: In complex terrain environments, even with similar terrain conditions, the actual driving performance of vehicles can vary, and this difference manifests in the traffic capacity coefficient. Unstable fluctuations will affect the vehicle capacity coefficient. Modeling for terrain conditions To characterize this uncertainty, conditional random variables are used to learn conditional probability distributions. A conditional variational autoencoder is used to model the conditional distribution. The conditional variational autoencoder consists of three parts: a conditional encoder module, an encoder network, and a decoder network. A training process sample tripartite dataset is constructed using the collected data. ; The conditional encoding module will input the condition. Joint encoding as a conditional representation vector The encoder network is defined as a posterior distribution: ; In the formula, As latent variables, This represents the encoder network parameters, which map observations to terrain conditions using Gaussian distribution parameters. To construct an approximate posterior, during the generation process, latent variables are first sampled from the standard normal distribution. Then with the condition vector The data is input together into the decoder network to predict throughput. The conditional probability distribution is obtained by using a Gaussian distribution as the basis. The modeling and decoding methods are as follows: ; in, and For the output of the decoder network, These are the decoder network parameters.

[0023] In this embodiment of the invention, the training objective of the conditional variational autoencoder is to maximize the variational lower bound (ELBO) of the conditional marginal likelihood, which is in the form of: ; in, KL divergence measures the latent distribution predicted by the encoder network. Compared with the standard Gaussian prior distribution The distance between them The reconstruction loss is used to measure the Gaussian distribution generated by the decoder relative to the observations. The degree of fit is specifically expressed as: .

[0024] Furthermore, such as Figure 4 As shown, S5 specifically refers to: Extracting deep feature representations from the last stage of the encoder in a semantic segmentation network. High-dimensional semantic feature embeddings used to describe the current image region: ; In the formula, For feature extraction function, For semantic segmentation networks, The input is a raster parameter vector; during the training phase, for each known semantic terrain category... Collect the corresponding deep feature set from the training set. The mean vector of the statistical distribution With covariance matrix : ; In the formula, To determine the number of semantic terrain categories, construct a Gaussian model for the features of each terrain category: ; During the reasoning phase, for the grid Extracting the intermediate layer feature vector Calculate the Mahalanobis distance with each semantic class Gaussian model: ; The category with the smallest distance among all categories is taken as the discrimination result: ; The result is used to assess whether the current terrain falls within the known distribution range of the training set. A higher value indicates that the terrain feature has deviated from the training distribution, and the accessibility model has the risk of extrapolation.

[0025] Furthermore, such as Figure 5 As shown, S6 specifically refers to: The conditional variational autoencoder outputs a Gaussian distribution. This represents the range of possible accessibility under the current grid terrain conditions, modeled by a Gaussian distribution. Mapping to pass probability, setting a safety threshold To distinguish whether a vehicle can respond to control commands, the probability of a vehicle successfully passing through is assessed by calculating the right-tail cumulative distribution function of the capacity distribution: ; In the formula, Traffic capacity coefficient Exceeding the safety threshold The probability when; Larger or When smaller, A value approaching 1 indicates high traffic stability; while when Smaller or When the probability is large, the right tail probability will decrease significantly, indicating that the terrain has high accessibility uncertainty; define a confidence decay factor. Markovnikov distance Integrate into a unified scoring system: ; in, As a moderating term, it controls for the impact of the degree of cognitive deviation on the final score. The confidence level is the Mahalanobis distance between the current terrain feature and the mean of its nearest semantic category, reflecting its typicality in the training distribution; the confidence level decreases exponentially when the feature distribution deviates significantly; the more the terrain conforms to the training experience, the higher the confidence level. The closer it is to 1; ultimately, combining the probability of the accessibility distribution with the credibility of terrain perception, the terrain accessibility probability is defined. : ; Terrain accessibility probability Simultaneously, the actual impact of terrain on vehicle responsiveness and the credibility of the perception model are considered.

[0026] In this embodiment of the invention, based on the probability of passability To perform binarized passability judgment, a probability threshold can be set. ,when The terrain is considered passable at certain times; otherwise, it is considered a high-risk area.

[0027] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The same or similar parts between the various embodiments can be referred to each other.

[0028] Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A accessibility prediction method for unstructured terrain environments, characterized in that, Includes the following steps: S1. Collect terrain information using multi-source sensors, fuse the terrain information to estimate the pose transformation matrix at each time step, construct a grid map, and predict the terrain category of each grid in the grid map based on a semantic segmentation network that is based on image point cloud collaboration. S2. Establish a three-dimensional semantic map and calculate the elevation feature vector of each two-dimensional grid based on the point cloud height information of each grid in the grid map; S3. During the multi-source sensor acquisition process, record the actual speed vector of the vehicle in each grid, define and calculate the traffic capacity coefficient to characterize the vehicle's ability to execute control commands in complex terrain; S4. Model the vehicle capacity coefficient as a conditional random variable under terrain conditions, and use a conditional variational autoencoder to model the conditional distribution. S5. During the actual operation of the vehicle, a Mahalanobis distance recognition mechanism is introduced to determine whether the currently observed terrain belongs to the learned distribution support domain. S6, combined with the terrain accessibility distribution of vehicles S4 and the terrain perception credibility of vehicle S5, comprehensively assesses the accessibility of the terrain at the grid.

2. The accessibility prediction method for unstructured terrain environments according to claim 1, characterized in that, Terrain information acquired by multiple sensors includes: lidar scan point cloud sequences. , Indicates the first t Frame point cloud; camera image sequence , For the first t Frame images; IMU measurement sequences , Indicates the first t Acceleration and angular velocity at frame time.

3. The accessibility prediction method for unstructured terrain environments according to claim 2, characterized in that, S1 specifically refers to: The pose transformation matrix at each time step is estimated by fusing information from LiDAR, camera vision, and IMU using a factor graph-based nonlinear optimization method. : ； In the formula, Represents the rotation matrix. Representing the three-dimensional Euclidean group, Represents the three-dimensional orthogonal group. Represents the translation vector, which will be used to register the point cloud. Projecting the map onto a world coordinate system, a dense map is constructed; a voxel-based 3D occupancy raster mapping method is used to generate a raster map, OctoMap. A semantic segmentation network based on image point cloud collaboration is introduced to segment the image... With the corresponding projected point cloud The input is fed into a semantic segmentation network to predict the terrain category for each raster: ； in, It is a grid semantic prediction labels, This represents a predefined set of terrain semantic categories. This represents a semantic segmentation network.

4. The accessibility prediction method for unstructured terrain environments according to claim 3, characterized in that, S2 specifically refers to: Based on the camera's intrinsic parameter matrix and pose transformation matrix The semantically predicted labels are back-projected onto the 3D map coordinate system to assign semantic attributes to the voxel raster, thereby establishing a 3D semantic map. : ； In the formula, x represents the coordinate vector. Represents depth values, based on the OctoMap raster map. The point cloud height information of each cell in the two-dimensional grid is used to calculate the grid. Elevation eigenvectors: ； Among them, is the grid. The elevation eigenvector represents the geometric features of the Earth's surface. For all falling into the grid The collection of point clouds, This represents the maximum height of all point clouds at the corresponding location. The average of all heights. For the standard deviation of all heights, n This represents the number of point clouds.

5. The accessibility prediction method for unstructured terrain environments according to claim 1, characterized in that, S3 specifically refers to: Record vehicle movement within the grid during multi-source sensor acquisition. actual velocity vector : ； In the formula, Indicates linear velocity. Represents angular velocity, and records the speed at which the control system issues the desired control command. : ； In the formula, To achieve the desired control command linear velocity, Define the traffic capacity coefficient to represent the desired control command angular velocity. To characterize the vehicle's ability to execute control commands in complex terrain: ； Traffic capacity coefficient This indicates the strength of a vehicle's response to desired control under certain terrain conditions. A larger value indicates that the terrain does not significantly hinder movement, while a smaller value indicates that movement is restricted.

6. The accessibility prediction method for unstructured terrain environments according to claim 1, characterized in that, S4 specifically refers to: Vehicle capacity coefficient Modeling for terrain conditions To characterize this uncertainty, conditional random variables are used to learn conditional probability distributions. ; A conditional variational autoencoder (CVA) is used to model the conditional distribution. The CVA consists of three parts: a conditional encoder module, an encoder network, and a decoder network. A training process sample triplet dataset is constructed using the collected data. ； The conditional encoding module will input the condition. Joint encoding as a conditional representation vector The encoder network is defined as a posterior distribution: ； In the formula, As latent variables, This represents the encoder network parameters, which map observations to terrain conditions using Gaussian distribution parameters. To construct an approximate posterior, during the generation process, latent variables are first sampled from the standard normal distribution. Then with the condition vector The data is input together into the decoder network to predict throughput. The conditional probability distribution is obtained by using a Gaussian distribution as the basis. The modeling and decoding methods are as follows: ； in, and For the output of the decoder network, These are the decoder network parameters.

7. The accessibility prediction method for unstructured terrain environments according to claim 1, characterized in that, S5 specifically refers to: Extracting deep feature representations from the last stage of the encoder in a semantic segmentation network. High-dimensional semantic feature embeddings used to describe the current image region: ； In the formula, For feature extraction function, For semantic segmentation networks, The input is a raster parameter vector; during the training phase, for each known semantic terrain category... Collect the corresponding deep feature set from the training set. The mean vector of the statistical distribution With covariance matrix : ； In the formula, To determine the number of semantic terrain categories, construct a Gaussian model for the features of each terrain category: ； During the reasoning phase, for the grid Extracting intermediate layer feature vectors Calculate the Mahalanobis distance with each semantic class Gaussian model: ； The category with the smallest distance among all categories is taken as the discrimination result: ； The value of the discrimination result is used to assess whether the current terrain falls within the known distribution range of the training set.

8. The accessibility prediction method for unstructured terrain environments according to claim 1, characterized in that, S6 specifically refers to: Gaussian distribution modeled as a conditional distribution Mapping to pass probability, setting a safety threshold To distinguish whether a vehicle can respond to control commands, the probability of a vehicle successfully passing through is assessed by calculating the right-tail cumulative distribution function of the capacity distribution: ； In the formula, Traffic capacity coefficient Exceeding the safety threshold The probability; defining the credibility decay factor. Markov distance Integrate into a unified scoring system: ； in, As a moderating term, it controls for the impact of the degree of cognitive deviation on the final score. It is the Mahalanobis distance between the current terrain feature and the mean of its nearest semantic category, reflecting its typicality in the training distribution; finally, combining the probability of the accessibility distribution with the cognitive credibility of the terrain, the terrain accessibility probability is defined. : ； Terrain accessibility probability Simultaneously, the actual impact of terrain on vehicle responsiveness and the credibility of the perception model are considered.

Images