Road understanding method and device under conditions of satellite denial and no prior map

Abstract

The invention relates to a road understanding method and device under the conditions of satellite denial and no prior map. The method comprises the following steps: generating global guide information by using a spline curve according to satellite map marking points; fusing the laser radar point cloud data to construct a three-dimensional terrain model, extracting roughness and a normal vector to obtain a corresponding graph, and constructing an environment trafficability graph; then, defining a credibility value for each graph, fusing a roughness graph and a normal vector graph evidence, and carrying out basic probability distribution fusion with a trafficability graph to obtain final data; determining a passable area based on the global guidance information and the final data, estimating the width of the passable area, and calculating the current position, angle expectation and standard deviation in combination with the local guidance information; and finally, calculating posteriori estimation by taking the distance as a measured value, and analyzing a boundary, so as to obtain a corrected guide path and a passable area. By adopting the method, the path generation accuracy can be improved.

Description

Technical Field

[0001] The present application relates to the field of autonomous driving technology, and more particularly to a method and device for understanding a road under satellite denial and no prior map conditions. Background Art

[0002] In the field of robotics and autonomous driving, the environment can be roughly divided into two types: structured and unstructured. Compared with structured environments, autonomous navigation in unstructured environments is like a difficult adventure and faces many difficult challenges. First of all, the unstructured environment itself is highly uncertain, which makes the scene in dynamic change at all times. Obstacles that may appear suddenly and the ever-changing road conditions all add great difficulty to navigation. Secondly, the complex state space also greatly increases the difficulty of accurate modeling and decision-making of the navigation system. In such an environment, various interference factors are intertwined, making it difficult for the system to accurately define its own state. At the same time, limited prior knowledge and the lack of high-precision positioning information further aggravate the complexity of navigation. When performing long-distance autonomous navigation tasks, in unfamiliar unstructured environments, the global planning module plays a vital role. This module needs to generate a path that can guide the vehicle based on the prior information of the target environment and the specific task requirements of the vehicle, and provide macro-guidance for subsequent local path planning and vehicle control. However, when faced with a complex and unknown unstructured environment, the lack of prior information is like putting shackles on the global path planning module, which seriously limits its function. In unknown unstructured environments, due to the lack of high-precision prior maps, satellite images become the main source of prior information on target scenes. Therefore, the global path planning module can only rely on satellite maps to generate a global guidance path. However, this path generated by satellite maps has great limitations and poor accuracy. Satellite maps are restricted by factors such as resolution and update frequency, and it is difficult to reflect the real-time details of the target environment, resulting in path accuracy far below the requirements of subsequent local path planning modules. With only satellite maps as the only prior information, achieving long-distance autonomous driving is undoubtedly a very challenging task, which places extremely high demands on the reliability, adaptability and intelligence level of autonomous driving technology. Summary of the invention

[0003] Based on this, it is necessary to provide a road understanding method and device under satellite denial and no prior map conditions that can improve the accuracy of path generation in response to the above technical problems.

[0004] A method for understanding a road under satellite denial and no prior map conditions, the method comprising: Based on the acquired satellite map, the necessary points on the marked path are used to generate dense and smooth global guidance information using spline curves; Effectively fuse and accurately infer the lidar point cloud data obtained by the ground unmanned platform at different times and positions to construct a three-dimensional terrain model; extract terrain roughness and normal vectors based on the three-dimensional terrain model to obtain a roughness map and a normal vector map; calculate an environmental traversability map based on the BEV constructed by multi-frame fusion of lidar and odometer; Define corresponding confidence values according to the roughness map, the normal vector map, and the environmental traversability map, fuse the evidence of the homologous roughness map and normal vector map to obtain the initially fused evidence; fuse the basic probability assignment corresponding to the environmental traversability map with the fused evidence to obtain the finally fused data; Determine the current traversable area based on the global guidance information and the finally fused data; estimate the width of the current traversable area, and calculate the expected value and standard deviation of the current position and angle according to the estimated width and the local guidance information received by the unmanned platform; Use the expected value and standard deviation of the current position and angle as measurement values, use the historical frame as a prior value, calculate the posterior distribution of the current frame, analyze the boundary of the traversable area according to the posterior distribution of the current frame, and calculate the corrected guidance path and the traversable area using the boundary of the traversable area.

[0005] A road understanding device under satellite denial and without prior map conditions, the device includes: A global guidance information generation module, configured to generate dense and smooth global guidance information using a spline curve according to the obtained set of path passing points marked on the satellite map; An environmental traversability map calculation module, configured to effectively fuse and accurately infer the lidar point cloud data obtained by the ground unmanned platform at different times and positions to construct a three-dimensional terrain model; extract terrain roughness and normal vectors based on the three-dimensional terrain model to obtain a roughness map and a normal vector map; calculate an environmental traversability map based on the BEV constructed by multi-frame fusion of lidar and odometer; A data fusion module, configured to define corresponding confidence values according to the roughness map, the normal vector map, and the environmental traversability map, fuse the evidence of the homologous roughness map and normal vector map to obtain the initially fused evidence; fuse the basic probability assignment corresponding to the environmental traversability map with the fused evidence to obtain the finally fused data; The passable area correction module is used to determine the current passable area based on the global guidance information and the finally fused data; estimate the width of the current passable area, calculate the expected values and standard deviations of the current position and angle according to the estimated width and the local guidance information received by the unmanned platform; use the expected values and standard deviations of the current position and angle as measurement values, use the historical frame as the prior value, calculate the posterior distribution of the current frame, analyze the boundary of the passable area according to the posterior distribution of the current frame, and calculate the corrected guidance path and the passable area using the boundary of the passable area.

[0006] The above-mentioned road understanding method under satellite denial and no prior map conditions. In this application, a three-dimensional terrain model is constructed by fusing lidar point cloud data, and the roughness map and normal vector map are extracted. Combining the environmental passability map constructed by multi-frame fusion of lidar and odometer, multi-source data is used to describe the environment from different angles, making up for the lack of satellite map information and reducing the impact of positioning uncertainty and inaccurate global guidance information. Define confidence values for each map and perform evidence fusion. Fuse the evidence of the homologous roughness map and normal vector map, and then fuse it with the basic probability assignment of the environmental passability map to obtain more comprehensive and accurate finally fused data, making the understanding of the environment more reliable and providing a solid basis for determining the passable area. Finally, generate global guidance information using spline curves based on the marked points on the satellite map, determine the passable area in combination with the finally fused data, and at the same time consider the local guidance information received by the unmanned platform, calculate the expected values and standard deviations of the position and angle by integrating global and local information, and then analyze the boundary of the passable area to obtain the corrected guidance path and passable area, effectively improving the accuracy of path generation and making it more adaptable to complex unstructured environments. BRIEF DESCRIPTION OF THE DRAWINGS

[0007] Figure 1 It is a schematic flowchart of a road understanding method under satellite denial and no prior map conditions in an embodiment; Figure 2 It is a schematic flowchart of the BEV generation process in an embodiment; Figure 3 It is a schematic diagram of the passability calculation process in an embodiment; Figure 4 It is a schematic diagram of manually drawn guidance information in another embodiment; Figure 5 It is a schematic diagram of the trajectory traveled on the passable map in an embodiment; Figure 6 It is a structural block diagram of a road understanding device under satellite denial and no prior map conditions in an embodiment. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0008] To make the objectives, technical solutions, and advantages of this application more clear and understandable, the following further details this application in combination with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely used to explain this application and are not used to limit this application.

[0009] In one embodiment, as Figure 1 shown, a road understanding method under satellite denial and without prior maps is provided, including the following steps: Step 102, generate dense and smooth global guidance information using spline curves according to the obtained satellite map marked path necessary point set.

[0010] When driving in an unfamiliar environment, humans usually rely on the global route planned in public software, the surrounding environment, and the general road structure to maintain an overall sense of direction, and then quickly identify the key areas in the local environment by combining local environmental information. The same can be analogized for ground unmanned platforms. In any scenario, assuming that the satellite map of the unfamiliar area where the ground unmanned platform is about to perform a task can be obtained, global guidance information is obtained through the satellite map. A human-machine interaction module is constructed, and the operator marks the path necessary point set based on the satellite map , and the program will automatically generate dense and smooth global guidance information using spline curves , as Figure 4 shown. Load this guidance information into the ground unmanned platform as the initial reference information for its autonomous navigation. The ground unmanned platform retrieves the nearby guidance information according to the real-time positioning information as the current local subtask and executes it.

[0011] Step 104, effectively fuse and accurately infer the lidar point cloud data obtained by the ground unmanned platform at different times and different positions to construct a three-dimensional terrain model; extract terrain roughness and normal vectors based on the three-dimensional terrain model to obtain a roughness map and a normal vector map; calculate the environmental traversability map based on the BEV constructed by multi-frame fusion of lidar and odometer.

[0012] In a complex off-road environment, the terrain has a high degree of complexity and uncertainty. To effectively cope with the strong vibrations and large fluctuations in pose (position and attitude) caused by terrain changes, it is necessary to analyze the environmental traversability using geometric information. By effectively fusing and accurately inferring the lidar point cloud data obtained by the ground unmanned platform at different times and different positions, a three-dimensional terrain model that can objectively, completely, and robustly represent the local scene space characteristics of the ground unmanned platform is generated .

[0013] Based on extract terrain roughness and normal vector Features, and the two are key parameters for evaluating passability.

[0014] For a certain grid cell , let represent the elevation value of this grid cell, and calculate the maximum elevation difference between the point cloud in this grid cell and the elevation value . Take this as the measure of the terrain roughness : (1) Among them, represents the elevation value of the point cloud . To reduce the influence of noise, normalize the elevation difference, and then apply Gaussian blur to the elevation difference grid to reduce the influence of noise on the passability analysis. Thus, an environmental roughness map can be obtained.

[0015] Similarly, for a certain grid cell , this application can use to represent its three-dimensional coordinates. By calculating the cross product of two orthogonal vectors formed by four adjacent points in the grid neighborhood, the normal vector is obtained: (2) Among them, , , and respectively represent the three-dimensional coordinates of the upper, lower, left, and right four spatial neighborhood grids of the grid cell , represents the cross product operation. To facilitate the evaluation of passability, this application calculates the cosine similarity between the normal vector and the unit vector perpendicular to the upward direction, and takes this as the evaluation result of the normal vector. Thus, an environmental normal vector map is obtained. Let represent the cosine similarity between the normal vector corresponding to the grid and .

[0016] Finally, for the roughness and the normal vector corresponding to any grid cell , this application uses the artificial potential field (APF) to characterize the influence of these two geometric factors on the environmental passability: (3) Among them, is the natural exponential function, are the roughness coefficient and the normal vector coefficient, respectively, and are the maximum and minimum thresholds for the roughness to affect passability. Similarly, and are the maximum and minimum thresholds for the normal vector to affect passability, respectively.

[0017] For the environmental apparent information, the environmental bird's-eye view (BEV) is used for environmental passability analysis. The BEV projects the surrounding environment onto a top-down representation, which can filter out information unrelated to vehicle driving, such as the sky, and is more suitable for downstream decision-making and planning tasks. As Figure 2 shown in the process, a high-density environmental bird's-eye view is constructed through point cloud projection and multi-frame fusion: First, a spatio-temporally aligned data stream is obtained through multi-modal fusion, including calibrated LiDAR point clouds, synchronized camera images (visible light or infrared), and a tightly coupled LiDAR-inertial odometry pose sequence. Then, using the pose sequence provided by the odometry data, the accumulated multi-frame radar point cloud data with color information is used to generate the environmental bird's-eye view. From the environmental bird's-eye view collected during driving, self-supervised labels are generated by an automatic process. During the data collection process, the areas passed by the vehicle and the LiDAR perception results are recorded. The areas passed by the vehicle can be considered safe, and the areas determined by the perception module to be obstacles are considered impassable, serving as self-supervised signals for learning environmental passability. An environmental passability feature learning method that only uses self-supervised labels is constructed. For the bird's-eye view input at time a passability feature extraction network is defined to predict the passability feature map , where D represents the dimension of the passability feature. The model consists of an encoder and a decoder . Through this model, a feature map of the same size as the original image is obtained.

[0018] The unstructured environmental scene is complex, with multiple feasible road surfaces, and it is difficult to determine the passability vector benchmark through experience. Implement online clustering of the passability prototype vectors. As Figure 3 shown, a dynamically adaptive prototype vector queue is maintained to objectively describe the environmental passability. By calculating the average similarity between the feature F and the passability prototype vector queue as a measure of the environmental apparent passability, the appearance-based environmental passability map is obtained through normalization. Through probabilistic environmental representation, it is easy to integrate with various planning paradigms in the form of a cost map.

[0019] Step 106: Define the corresponding confidence values according to the roughness map, the normal vector map, and the environmental passability map, fuse the evidence of the homologous roughness map and normal vector map to obtain the initially fused evidence; fuse the basic probability assignment corresponding to the environmental passability map with the fused evidence to obtain the finally fused data.

[0020] A single sensor alone is not sufficient for a complete analysis of the environmental passability. The color and texture information of the environment is analyzed using a camera, and the geometric information of the environment is analyzed using a lidar. The robustness of the system can be effectively improved through multi-sensor information fusion. Based on the environmental bird's-eye view constructed by multi-frame fusion of lidar and odometer, the calculated from it is completely aligned with the roughness map in space and time. Therefore, an environmental passability map of the same size is constructed For any grid cell in, the D-S evidence theory is used to evaluate its passability.

[0021] First, define the frame of discernment , and the evidence sources include , and their corresponding basic probability assignments (BPAs) are respectively . For any grid cell , the confidence value corresponding to each evidence source is: (4) Next, for the homologous results , evidence fusion: (5) Among them, is the conflict degree between evidence and : (6) The synthesized represents the comprehensive evaluation of roughness and normal vector. Next, fuse this result with the basic probability assignment of the appearance information further: (7) Among them, is the conflict degree between the comprehensive evidences, and the calculation method is as shown in equation (6).

[0022] The application of LiDAR point cloud data makes it possible to build a high-precision three-dimensional terrain model. The LiDAR point cloud data obtained at different times and locations can depict the contours and features of the terrain in detail after effective fusion and accurate inference. By extracting the terrain roughness and normal vector, the roughness map and normal vector map obtained quantitatively describe the environment from the perspective of terrain characteristics. The roughness map can reflect the undulation and flatness of the ground, and the normal vector map can reflect the inclination direction and angle of the terrain. This information is crucial for judging the passability of the environment. The bird's-eye view (BEV) environmental passability map constructed by multi-frame fusion of LiDAR and odometer analyzes the environment from another dimension. Odometer data can provide motion information of the unmanned platform. After fusion with LiDAR data, the scope of the passable area can be more accurately determined. The comprehensive use of multi-source data describes the environment from multiple levels, greatly enriches the understanding of the environment, reduces the uncertainty caused by a single data source, and provides more comprehensive and accurate basic data for subsequent path planning. Traditional global path planning that relies only on satellite maps cannot accurately reflect the real-time status of the environment due to the limitations of satellite map information. This application effectively makes up for this shortcoming by integrating multi-source data.

[0023] Step 108, determine the current passable area based on the global guidance information and the final fused data; estimate the width of the current passable area, and calculate the expected value and standard deviation of the current position and angle based on the estimated width and the local guidance information received by the unmanned platform.

[0024] The path directly obtained from the satellite image usually has a large uncertainty, and the accuracy level cannot meet the requirements of local planning. Secondly, local planning is often highly sensitive to positioning errors. If the pose estimation module fails or produces inaccurate results, the planned local path will deviate significantly, increasing the risk of incorrect routes and improper driving behavior, posing a serious threat to the safety of autonomous driving. Therefore, this application determines the current passable area based on global guidance information and the final fused data, giving full play to the synergy of global guidance information and local guidance information. Based on the satellite map, the spline curve is used to generate dense and smooth global guidance information, which provides macro-directional guidance for the driving of the unmanned platform. Combined with the final fused data, the current passable area is determined, and the width of the passable area is estimated. On this basis, the local guidance information received by the unmanned platform is fully considered to calculate the expected value and standard deviation of the current position and angle. Local guidance information can reflect the real-time environmental information around the unmanned platform, such as the location and dynamic changes of nearby obstacles. By combining global and local information, the position and driving direction of the unmanned platform in the current environment can be more accurately determined, avoiding path planning errors caused by the inaccuracy of global information or the one-sidedness of local information.

[0025] Assume that there is only one passable area in the local environment , and its geometric structure is consistent with the geometric structure of the local guidance information , and it is the area with the highest passing probability in the current environment. The passable area is an area that meets the current environmental observations with a certain width. First, estimate its width. Assuming that there is no significant mutation in the environment during short-term driving, use the width information of the passable area observed in the short term as the benchmark for predicting the width of the future driving area. To achieve this prediction, project the short-term driving trajectory of the vehicle onto the current environmental passable map to establish a sample set , where is the set of grids covered by short-term driving, as shown in Figure 5 . Assume that the passability index follows a Gaussian distribution , and obtain the parameters through maximum likelihood estimation: (8) To estimate the width of the current passable area, perform two-way area expansion along the normal direction of the trajectory driven in the short term, and use the principle for confidence detection: For the expansion point , if it satisfies , it is determined as a passable area, otherwise it is marked as a boundary point. Through the iterative expansion and detection process, finally calculate the mean Euclidean distance of the bilateral boundary point set: (9) where is the estimate of the width of the current passable area, and are the left and right boundary points respectively

[0026] Due to the influence of factors such as positioning errors and satellite map distortion, with as the width, and the local guidance information received by the unmanned platform as the structure and the area of the initial position, there are position and angle deviations from the true passable area . Assume that the deviations , , and follow Gaussian distributions and are independent of each other. Use the random sampling method. In , , respectively represent the sampling ranges. With as the width hypothesis and as the geometric structure, randomly generate a series of hypotheses of passable areas , where is a coordinate transformation operator, represents the expansion along the path normal , generating the th candidate region.

[0027] For any sampling , calculate the average traversable cost within this region , and based on this, calculate the weight of this sampling deviation : (10) Then, by calculating the weighted average and weighted variance, obtain the expected values and standard deviations of the position and angle: (11) where are the means obtained by observing the current environment for the deviation respectively, are the standard deviations obtained by observing the current environment for the deviation respectively.

[0028] Step 110, calculate using the expected values and standard deviations of the current position and angle as measurement values, use the historical frame as the prior value, calculate the posterior distribution of the current frame, analyze the traversable region boundary based on the posterior distribution of the current frame, and calculate the corrected guidance path and traversable region using the traversable region boundary.

[0029] Assume that the current is the th frame. Through the above calculations, the measurement obtained from the current environment can be obtained, where represents the measurement value, represents the state. Using the posterior distribution of the th frame as the prior distribution of the current frame. Since the positioning information comes from the result of the previous frame and the current motion estimation and environment matching, the positioning result of the current frame is only related to the positioning result of the previous frame. Therefore, assume that the deviation caused by positioning follows the Markov assumption. Under this premise, estimate the posterior distribution of the th frame, and there is: (12) According to the chain rule: (13) where (14) Therefore, it can be obtained: (15) Among them, represents the normalization factor. Thus, the posterior estimate of the current frame is obtained. Based on this, the boundary of the passable area is analyzed. The calculation process is the same as (9), and the width of the passable area can be obtained. Thus, the corrected guidance path and the passable area can be obtained: (16) Taking the expected values and standard deviations of the current position and angle as measurement values, and using the historical frame information as prior values, the posterior estimate of the current frame is calculated. This process fully considers the uncertainty of the measurement data. By analyzing the posterior estimate probability of the current frame, the boundary of the passable area can be determined more accurately. As the unmanned platform moves and the environment changes, the method based on posterior estimation can update the understanding of the passable area in real time, and timely discover and correct the errors in path planning. According to the boundary of the passable area, the corrected guidance path and the passable area are calculated, enabling the path planning to dynamically adapt to the changes in the environment, and further improving the accuracy and adaptability of path generation.

[0030] The above method for road understanding under satellite denial and without prior map conditions. In this application, a three-dimensional terrain model is constructed by fusing lidar point cloud data, and the roughness map and normal vector map are extracted. Combining the environmental passability map constructed by multi-frame fusion of lidar and odometer, multi-source data are used to describe the environment from different angles, making up for the deficiency of satellite map information and reducing the impact of positioning uncertainty and inaccurate global guidance information. Trust values are defined for each map and evidence fusion is performed. The evidence of the homologous roughness map and normal vector map is fused, and then fused with the basic probability assignment of the environmental passability map to obtain more comprehensive and accurate final fusion data, making the understanding of the environment more reliable and providing a solid basis for determining the passable area. Finally, global guidance information is generated using a spline curve based on the marked points on the satellite map. Combining the final fusion data to determine the passable area, and considering the local guidance information received by the unmanned platform at the same time, the expected values and standard deviations of the position and angle are calculated by integrating global and local information, and then the boundary of the passable area is analyzed to obtain the corrected guidance path and the passable area, effectively improving the accuracy of path generation and making it more adaptable to complex unstructured environments.

[0031] In one of the embodiments, extracting terrain roughness and normal vectors based on the three-dimensional terrain model includes: For a certain grid cell of the three-dimensional terrain model , let represent the elevation value of the grid cell, and calculate the point cloud within the grid cell and the elevation value The maximum elevation difference between them is used as the measurement of terrain roughness: ; where represents the elevation value of the point cloud ; For a certain grid cell of the three-dimensional terrain model , use to represent its three-dimensional coordinates. By calculating the cross product of two orthogonal vectors formed by four adjacent points in the grid neighborhood, the normal vector is obtained: : ; where , , and respectively represent the three-dimensional coordinates of the upper, lower, left, and right four spatial neighborhood grids of the grid cell , represents the cross product operation.

[0032] In one of the embodiments, the roughness map and the normal vector map are respectively: ; where is the natural exponential function, are respectively the roughness coefficient and the normal vector coefficient, and are respectively the maximum and minimum thresholds for the roughness to affect passability, and are respectively the maximum and minimum thresholds for the normal vector to affect passability, is the roughness corresponding to any grid cell , represents the cosine similarity between the normal vector corresponding to the grid cell and .

[0033] Fuse the evidence of the homologous roughness map and normal vector map to obtain the initially fused evidence as: ; where is the conflict degree of the evidence of the homologous roughness map and normal vector map, is the evidence of the roughness map, is the evidence of the normal vector map, represents a custom fusion operator.

[0034] In one embodiment, the basic probability assignment corresponding to the environmental passability map is fused with the fused evidence to obtain the finally fused data, including: The basic probability assignment corresponding to the environmental passability map is fused with the fused evidence to obtain the finally fused data as: ; Wherein, is the conflict degree between comprehensive evidences, is the evidence after initial fusion, is the basic probability assignment corresponding to the environmental passability map.

[0035] In one embodiment, the width of the passable area is estimated, including: Project the short-term driving trajectory of the vehicle onto the current environmental passability map to establish a sample set wherein, is the set of grids covered by short-term driving; Assume that the passability index follows a Gaussian distribution , and the parameters are obtained through maximum likelihood estimation as: ; Wherein, n represents the total number of candidate areas, represents the th candidate area; Perform two-way area expansion along the normal direction of the trajectory traveled in the short term , and use the principle for confidence detection. For the expansion point , if it satisfies , it is determined as a passable area, otherwise it is marked as a boundary point. Through the iterative expansion and detection process, the average Euclidean distance of the bilateral boundary point set is finally calculated as the width of the current passable area.

[0036] In one embodiment, the expected value and standard deviation of the current position and angle are calculated according to the estimated width and the local guidance information received by the unmanned platform, including: Taking the estimated width as the width hypothesis and the local guidance information received by the unmanned platform as the geometric structure, a series of hypotheses of passable areas are randomly generated wherein is the coordinate transformation operator, represents expansion along the path normal direction , generating the th candidate area, represents the position deviation, Indicates the angular deviation; For any sampling Calculate the average traversable cost within this area , and based on this, calculate the weight of this sampling deviation, and then obtain the expected values and standard deviations of the position and angle by calculating the weighted average and weighted variance.

[0037] In one embodiment, obtaining the expected values and standard deviations of the position and angle by calculating the weighted average and weighted variance includes: The expected values and standard deviations of the position and angle obtained by calculating the weighted average and weighted variance are: ; Wherein, Are respectively the expected values obtained by observing the deviation Through the current environment, Are respectively the standard deviations obtained by observing the deviation Through the current environment, Represents the weight of the sampling deviation, n Represents the total number of candidate areas.

[0038] In one embodiment, taking the expected values and standard deviations of the current position and angle as measurement values, using the historical frame as the prior value, and calculating the posterior distribution of the current frame includes: Taking the expected values and standard deviations of the current position and angle as measurement values , where Represents the measurement value, Represents the state, with the posterior distribution Of the Frame as the prior distribution of the current frame, assuming that the deviation caused by positioning follows the Markov assumption. On this premise, estimate the posterior distribution Of the Frame, there is: ; According to the chain rule: ; Wherein, ; Therefore, it can be obtained that: ; Wherein, Represents the normalization factor, from which the posterior estimate Of the current frame is obtained.

[0039] In one embodiment, using the boundary of the traversable area to calculate the corrected guiding path and the traversable area includes: The corrected guidance path and the passable area calculated using the passable area boundary are as follows: ; Among them, represents the posterior estimation of the current frame, represents the corrected guidance path, represents the corrected passable area, represents the passable area boundary.

[0040] In another embodiment, under the conditions of GNSS denial and lack of prior maps, satellite maps are used as the positioning reference. Due to the update delay of satellite images, there is a time-varying deviation between the obtained satellite images and the real environment, resulting in significant differences in the observation results of the same area. To reduce the influence of factors such as seasonal changes and lighting changes, the satellite maps are analyzed, and an appearance-based environmental passability analysis method is used to slice the satellite maps along the global guidance information, extract prototype vectors through the global guidance information, conduct environmental passability analysis, map the satellite maps to the passability space, and obtain the environmental passability map based on the satellite maps , realizing the mapping of environmental features to the passability space. By projecting the real-time perceived BEV features and satellite images into a unified feature space, the matching problem that the regional features are easily affected by meteorological conditions (rain, fog, smoke, and dust) and spatio-temporal variations (seasonal lighting) is effectively overcome. After obtaining the representation in the same feature space, relevant area slices of the satellite map passability map are intercepted according to the initial pose, and the similarity between and is calculated based on the normalized cross-correlation (NCC) as the particle weight. The particle pose is updated by odometer motion prediction + Gaussian noise, and the pose probability distribution is iteratively optimized to achieve global positioning.

[0041] It should be understood that although each step in the flowchart of Figure 1 is shown in sequence according to the arrow indication, these steps are not necessarily executed in the order indicated by the arrow. Unless otherwise clearly stated in this article, the execution of these steps has no strict order limit, and these steps can be executed in other orders. Moreover, Figure 1 at least a part of the steps in

[0042] In one embodiment, as shown in Figure 6As shown, a road understanding device under satellite denial and without prior map conditions is provided, including: a global guidance information generation module 602, an environmental passability map calculation module 604, a data fusion module 606, and a passable area correction module 608, where: The global guidance information generation module 602 is configured to generate dense and smooth global guidance information using spline curves based on the obtained satellite map marked path necessary point set; The environmental passability map calculation module 604 is configured to effectively fuse and accurately infer the lidar point cloud data obtained by the ground unmanned platform at different times and positions to construct a three-dimensional terrain model; extract terrain roughness and normal vectors based on the three-dimensional terrain model to obtain a roughness map and a normal vector map; calculate an environmental passability map based on the BEV constructed by multi-frame fusion of lidar and odometer; The data fusion module 606 is configured to define corresponding confidence values according to the roughness map, the normal vector map, and the environmental passability map, fuse the evidence of the homologous roughness map and normal vector map to obtain the initially fused evidence; fuse the basic probability assignment corresponding to the environmental passability map with the fused evidence to obtain the finally fused data; The passable area correction module 608 is configured to determine the current passable area based on the global guidance information and the finally fused data; estimate the width of the current passable area, calculate the expected value and standard deviation of the current position and angle according to the estimated width and the local guidance information received by the unmanned platform; use the expected value and standard deviation of the current position and angle as measurement values, use the historical frame as a prior value, calculate the posterior distribution of the current frame, analyze the passable area boundary according to the posterior distribution of the current frame, and calculate the corrected guidance path and the passable area using the passable area boundary.

[0043] For the specific limitations of the road understanding device under satellite denial and without prior map conditions, reference can be made to the limitations of the road understanding method under satellite denial and without prior map conditions in the above text, which will not be elaborated here. Each module in the above road understanding device under satellite denial and without prior map conditions can be implemented in whole or in part by software, hardware, and their combination. The above modules can be embedded in the processor of the computer device in hardware form or independent of it, or stored in the memory of the computer device in software form, so that the processor can call and execute the operations corresponding to the above modules.

[0044] The technical features of the above embodiments can be combined arbitrarily. For the sake of concise description, not all possible combinations of the technical features in the above embodiments are described. However, as long as the combinations of these technical features do not conflict, they should be considered to be within the scope described in this specification.

[0045] The above-described embodiments merely represent several implementation manners of the present application. The description thereof is relatively specific and detailed, but it should not be construed as a limitation on the scope of the invention. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present application, several modifications and improvements can still be made, and these all fall within the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the appended claims.

Claims

1. A method for understanding roads under satellite denial and no prior map conditions, characterized in that: The method comprises: Based on the acquired satellite map, the necessary points on the marked path are used to generate dense and smooth global guidance information using spline curves; The laser radar point cloud data acquired by the ground unmanned platform at different times and locations are effectively fused and accurately inferred to construct a three-dimensional terrain model; the terrain roughness and normal vector are extracted based on the three-dimensional terrain model to obtain a roughness map and a normal vector map; the BEV calculation constructed based on the laser radar and odometer multi-frame fusion obtains an environmental passability map; According to the corresponding trust values ​​defined by the roughness map, the normal vector map and the environment passability map, evidence of the same source roughness map and normal vector map is fused to obtain initial fused evidence; the basic probability distribution corresponding to the environment passability map is fused with the fused evidence to obtain final fused data; Determine the current passable area based on the global guidance information and the final fused data; estimate the width of the current passable area, and calculate the expected value and standard deviation of the current position and angle according to the estimated width and the local guidance information received by the unmanned platform; The expected value and standard deviation of the current position and angle are used as measurement values, and the historical frame is used as a priori value to calculate the posterior distribution of the current frame. The boundary of the passable area is analyzed according to the posterior distribution of the current frame, and the corrected guidance path and the passable area are calculated using the boundary of the passable area.

2. The method according to claim 1, characterized in that Extracting terrain roughness and normal vector based on the three-dimensional terrain model includes: For 3D terrain models A grid cell ,make Indicates the elevation value of the grid cell and calculates the point cloud within the grid cell With elevation value The maximum elevation difference between Metrics: ； in, Representing point clouds The elevation value of For 3D terrain models A grid cell ,use Represents its three-dimensional coordinates, and the normal vector is obtained by calculating the cross product of two orthogonal vectors formed by four adjacent points in the grid neighborhood. : ； in, , , and Represents grid cells The three-dimensional coordinates of the upper, lower, left and right spatial neighborhood grids, Represents the cross product operation.

3. The method according to claim 1, characterized in that The roughness map and normal vector map are respectively: ； in, is the natural exponential function, are the roughness coefficient and the normal vector coefficient, respectively. and are the maximum and minimum thresholds of roughness affecting passability, and are the maximum and minimum thresholds of the normal vector affecting the passability, For any grid cell The corresponding roughness, Represents a grid cell The corresponding normal vector is The cosine similarity of The evidence of the homologous roughness map and normal vector map is fused, and the initial fused evidence is obtained as follows: ； in, is the conflict degree of evidence of the same roughness map and normal map, is the evidence of the roughness map, is the evidence of the normal vector graph, Indicates a custom fusion operator.

4. The method according to claim 3, characterized in that: The basic probability distribution corresponding to the environmental passability graph is fused with the fused evidence to obtain final fused data, including: The basic probability distribution corresponding to the environmental passability graph is fused with the fused evidence to obtain the final fused data: ； in, is the degree of conflict between the comprehensive evidence, is the evidence after the initial fusion, It is the basic probability distribution corresponding to the environment traversability graph.

5. The method according to claim 1, characterized in that The width of the traversable area is estimated, including: Project the short-term driving trajectory of the vehicle onto the traversable map of the current environment On the top, create a sample set ,in, A collection of grids covered by short-term driving; Assume that the accessibility index follows a Gaussian distribution , the parameters obtained by maximum likelihood estimation are: ； in, n represents the total number of candidate regions, Indicates candidate regions; Along the track traveled in the short term The normal direction of the two-way area expansion is carried out using Principle for confidence detection, for extension points , if satisfied , it is determined as a passable area, otherwise it is marked as a boundary point. Through iterative expansion and detection process, the mean Euclidean distance of the boundary point set on both sides is finally calculated as the width of the current passable area.

6. The method according to claim 1, characterized in that The expected value and standard deviation of the current position and angle are calculated based on the estimated width and the local guidance information received by the unmanned platform, including: The estimated width Assume that the width is 100,000. A random sequence of traversable regions is generated for the geometric structure ,in is the coordinate transformation operator, Indicates expansion along the path normal , generating candidate regions, Indicates position deviation, Indicates the angle deviation; For any sample Calculate the average traversable cost in the area , and based on this, the weight of the sampling deviation is calculated, and then the expected value and standard deviation of the position and angle are obtained by calculating the weighted mean and weighted variance.

7. The method according to claim 6, characterized in that By calculating the weighted mean and weighted variance, the expected value and standard deviation of the position and angle are obtained, including: By calculating the weighted mean and weighted variance, the expected value and standard deviation of position and angle are obtained as follows: ； in, The deviation The expected value obtained by observing the current environment, The deviation The standard deviation obtained by observing the current environment, represents the weight of sampling bias, n Represents the total number of candidate regions.

8. The method according to claim 1, characterized in that The expected value and standard deviation of the current position and angle are used as measurement values, and the historical frame is used as a priori value to calculate the posterior distribution of the current frame, including: The expected value and standard deviation of the current position and angle are taken as the measured values ,in Indicates the measured value, Indicates the status, Posterior distribution of frames Assume that the deviation caused by positioning follows the Markov assumption as the prior distribution of the current frame. Under this premise, estimate Posterior distribution of frames ,have: ； According to the chain rule: ； in, ； Therefore, we can get: ； in, Represents the normalization factor, thus, the posterior estimate of the current frame is obtained .

9. The method according to claim 1, characterized in that: The corrected guidance path and the traversable area are calculated using the traversable area boundary, including: The corrected guidance path and the passable area are calculated using the passable area boundary as follows: ； in, represents the current frame posterior estimate, represents the corrected guidance path, represents the corrected traversable area, Indicates the boundary of the traversable area.

10. A road understanding device under satellite denial and no prior map conditions, characterized in that: The device comprises: A global guidance information generation module is used to generate dense and smooth global guidance information using a spline curve based on a set of necessary points on the path marked on the acquired satellite map; The environment accessibility map calculation module is used to effectively fuse and accurately infer the laser radar point cloud data obtained by the ground unmanned platform at different times and different positions to build a three-dimensional terrain model; extract the terrain roughness and normal vector based on the three-dimensional terrain model to obtain a roughness map and a normal vector map; calculate the environment accessibility map based on the BEV constructed by multi-frame fusion of the laser radar and the odometer; A data fusion module is used to define corresponding trust values ​​according to the roughness map, the normal vector map and the environmental passability map, fuse the evidence of the same source roughness map and the normal vector map to obtain initial fused evidence; fuse the basic probability distribution corresponding to the environmental passability map with the fused evidence to obtain final fused data; A traversable area correction module is used to determine the current traversable area based on the global guidance information and the final fused data; estimate the width of the current traversable area, and calculate the expected value and standard deviation of the current position and angle according to the estimated width and the local guidance information received by the unmanned platform; use the expected value and standard deviation of the current position and angle as measurement values, use historical frames as prior values, calculate the posterior distribution of the current frame, analyze the traversable area boundary according to the posterior distribution of the current frame, and use the traversable area boundary to calculate the corrected guidance path and traversable area.

Images