Route determination method and device, storage medium and vehicle

Abstract

The application discloses a kind of determination method, device, storage medium and vehicle of driving route.Therein, the method comprises: obtaining the drivable area of vehicle in map;Determine the ground point of grid that constitutes drivable area, and determine the local ground height and ground undulation of grid based on ground point;Determine target drivable area in drivable area based on local ground height and ground undulation;In target drivable area, the driving route that vehicle is to be driven is planned.The present application solves the technical problem that the rationality of vehicle route planning is poor.

Description

Technical Field

[0001] This invention relates to the field of vehicles, and more specifically, to a method, apparatus, storage medium, and vehicle for determining a driving route. Background Technology

[0002] Roads can be divided into structured roads (such as highways or urban arterial roads) and unstructured roads (such as rural roads).

[0003] In related technologies, high-precision maps can be used to assist in the perception and identification of obstacles for structured roads. However, for unstructured roads, due to the unclear lane and road boundaries and the difficulty in collecting relevant data in real time to update the high-precision map, there is still a technical problem of poor rationality in vehicle route planning.

[0004] There is currently no effective solution to the technical problem of poor rationality in vehicle route planning in the aforementioned related technologies. Summary of the Invention

[0005] This invention provides a method, apparatus, storage medium, and vehicle for determining a driving route, in order to at least solve the technical problem of poor rationality in vehicle route planning.

[0006] According to one aspect of the present invention, a method for determining a driving route is provided. The method may include: acquiring a drivable area of ​​a vehicle in a map; determining ground points of a grid constituting the drivable area, and determining the local ground elevation and ground undulation of the grid based on the ground points; determining a target drivable area within the drivable area based on the local ground elevation and ground undulation; and planning a driving route for the vehicle to travel within the target drivable area.

[0007] Optionally, determining the ground points of the grid that makes up the drivable area includes: dividing the drivable area into grids; performing clustering on each grid and determining the local ground equation for each grid based on the clustering results; calculating the point cloud in each grid based on the local ground equation to obtain the calculation result; and determining the point cloud as a ground point in response to the calculation result being less than a distance threshold.

[0008] Optionally, determining the local ground height of a grid based on ground points includes: determining the calculation results of the ground points in each grid in the local ground equation corresponding to the grid; and determining the average of the sum of all calculation results as the local ground height of the grid.

[0009] Optionally, determining the ground relief of the grid based on ground points includes: determining the calculation results of the point cloud (excluding ground points) in each grid in the local ground equation corresponding to the grid; and determining the maximum value in the calculation results as the ground relief of the grid.

[0010] Optionally, grids with ground relief greater than the structured road ground relief threshold and less than the unstructured road ground relief threshold are identified and designated as unstructured sub-regions.

[0011] Optionally, determining the target drivable area within the drivable area based on local ground height and ground undulation includes: performing obstacle perception on obstacles within the drivable area to obtain perception results; determining the center point position and height information of the obstacle based on the perception results; determining the target obstacle among the obstacles based on the center point position and height information, local ground height and ground undulation; and determining the target drivable area based on the target obstacle.

[0012] Optionally, based on the center point location, height location, local ground height, and ground undulation, the target obstacle in the obstacle is determined, including: in response to a first height in the height information being greater than the local ground height, the obstacle is identified as a target obstacle, resulting in a first obstacle set containing at least one obstacle; the point cloud in the drivable area, excluding ground points, is clustered and segmented to obtain an initial obstacle set; obstacles in the initial obstacle set whose center point location is located in an unstructured sub-region, and whose second height in the height information is between the ground undulation threshold of the structured sub-region and the ground undulation threshold of the unstructured sub-region are filtered out, resulting in... The system proceeds to the second obstacle set, where the first height of the same obstacle is less than the second height; the drivable area is expanded equidistantly to determine obstacle information within the target range, resulting in the third obstacle set; obstacles intersecting the second and third obstacle sets are identified as interfering obstacles, resulting in interfering obstacles, where interfering obstacles are those that do not obstruct the vehicle's path; the second and third obstacle sets are merged, and the interfering obstacle set is removed from the merged obstacle set to obtain the fourth obstacle set; the fourth obstacle set is merged with the first obstacle set to obtain the target obstacle set with at least one target obstacle.

[0013] According to another aspect of the present invention, a route determination device is also provided. The device may include: an acquisition unit for acquiring a drivable area of ​​a vehicle in a map; a first determination unit for determining ground points of a grid constituting the drivable area, and determining the local ground elevation and ground undulation of the grid based on the ground points; a second determination unit for determining a target drivable area within the drivable area based on the local ground elevation and ground undulation; and a planning unit for planning a route for the vehicle to travel within the target drivable area.

[0014] According to another aspect of the present invention, a computer-readable storage medium is also provided. The computer-readable storage medium includes a stored program, wherein, when the program is executed, it controls the device where the computer-readable storage medium is located to perform the method for determining a driving route according to the embodiments of the present invention.

[0015] According to another aspect of the present invention, a processor is also provided. The processor is used to run a program, wherein the program executes the method for determining a driving route according to the embodiments of the present invention.

[0016] According to another aspect of the present invention, a vehicle is also provided. This vehicle is used to perform the method for determining a travel route according to the embodiments of the present invention.

[0017] In this embodiment of the invention, the drivable area of ​​the vehicle in the map is obtained; the ground points of the grids that make up the drivable area are determined, and the local ground height and ground undulation of the grids are determined based on the ground points; the target drivable area in the drivable area is determined based on the local ground height and ground undulation; and the driving route of the vehicle to be driven is planned in the target drivable area. In other words, this embodiment of the invention determines the drivable area of ​​the vehicle from the map, divides the area into multiple grids, determines the local ground height and ground undulation of each grid, and processes the local ground height and ground undulation to determine the target drivable area of ​​the vehicle from the drivable area. Because the drivable area is divided into multiple grids for analysis, the problem of inaccurate determination of the target drivable area is avoided, thereby achieving the goal of more accurate planning of the vehicle's driving route. This solves the technical problem of poor rationality in vehicle route planning and achieves the technical effect of improving the rationality of vehicle route planning. Attached Figure Description

[0018] The accompanying drawings, which are included to provide a further understanding of the invention and form part of this application, illustrate exemplary embodiments of the invention and, together with their description, serve to explain the invention and do not constitute an undue limitation thereof. In the drawings:

[0019] Figure 1 This is a flowchart of a method for determining a driving route according to an embodiment of the present invention;

[0020] Figure 2 This is a flowchart of a method for obtaining the perception range of a structured road using high-precision map data according to an embodiment of the present invention;

[0021] Figure 3 This is a schematic diagram of roadside weeds extending onto the road according to an embodiment of the present invention;

[0022] Figure 4This is a schematic diagram of a region of interest containing unstructured roads according to an embodiment of the present invention;

[0023] Figure 5(a) is a schematic diagram of the scanning point distribution of a mechanical rotating lidar according to an embodiment of the present invention;

[0024] Figure 5(b) is a schematic diagram of the scanning point distribution of a solid-state lidar according to an embodiment of the present invention;

[0025] Figure 6 This is a flowchart of a method for determining road type based on local ground height and ground undulation according to an embodiment of the present invention;

[0026] Figure 7 This is a schematic diagram of different road areas within a region of interest according to an embodiment of the present invention;

[0027] Figure 8 This is a schematic diagram of fallen leaves extending from a road boundary onto the road according to an embodiment of the present invention;

[0028] Figure 9 This is a schematic diagram of a route determination device according to an embodiment of the present invention. Detailed Implementation

[0029] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. 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 should fall within the scope of protection of the present invention.

[0030] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0031] Example 1

[0032] According to an embodiment of the present invention, an embodiment of a method for determining a driving route is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0033] Figure 1 This is a flowchart of a method for determining a driving route according to an embodiment of the present invention, such as... Figure 1 As shown, the method includes the following steps:

[0034] Step S102: Obtain the drivable area of ​​the vehicle on the map.

[0035] In the technical solution provided in step S102 of the present invention, the drivable area of ​​a vehicle on the current road can be determined from a map. The map can be a high-precision map, a satellite map, etc., which is only an example and not a specific limitation. The drivable area can include structured sub-regions and unstructured sub-regions. Structured sub-regions can be used to characterize the drivable range on structured roads surrounding the vehicle in the map; unstructured sub-regions can be used to characterize the drivable range on unstructured roads surrounding the vehicle in the map. Structured roads can be well-structured and relatively orderly roads such as highways or urban arterial roads; unstructured roads can be less structured roads such as urban non-main roads or rural streets.

[0036] Optionally, the vehicle's geographical coordinates can be obtained through the vehicle's Global Positioning System (GPS) information. These coordinates can be projected onto a high-precision map, and a Region of Interest (ROI) can be defined on the map. Based on the ROI, the vehicle's drivable area can be obtained. For example, the drivable area can be determined by removing road boundaries and obstacles from the ROI.

[0037] For example, the region of interest of a vehicle can contain information on all roads within the target range surrounding the vehicle's geographical location. Based on the information of each road, it can be determined whether the road is a structured road or an unstructured road. The road boundaries of all roads within the region of interest can be obtained, and the area within each road boundary can be set as a region of interest for the vehicle. By connecting the regions of interest in all regions of interest and taking the union, the drivable area of ​​the vehicle on the map can be determined. It should be noted that no specific restrictions are placed on the target range here.

[0038] Optionally, information on all roads within the vehicle's region of interest is analyzed. If it is determined that there are structured roads within the region, the structured drivable sub-regions of the vehicle can be determined based on the road boundaries of the structured roads. By taking the union of the structured drivable sub-regions of all structured roads, the drivable area of ​​the vehicle's structured roads can be determined.

[0039] Optionally, information on all roads within the vehicle's region of interest is analyzed. If unstructured roads are found within this region, the region of interest can be temporarily marked for subsequent secondary determination of road boundaries to further ascertain the existence of unstructured roads. If so, the marking can be retained. The area within the road boundaries of unstructured roads can be defined as the vehicle's unstructured sub-region.

[0040] In this embodiment of the invention, it is possible to identify whether there are unstructured roads in the region of interest. If it is determined that there are, considering the possibility of misjudging unstructured roads, a second judgment can be made, thereby achieving the technical effect of improving the accuracy of unstructured road identification.

[0041] Step S104: Determine the ground points of the grid that makes up the drivable area, and determine the local ground height and ground undulation of the grid based on the ground points.

[0042] In the technical solution provided by step S104 of the present invention, after determining the drivable area of ​​the vehicle from the map, the drivable area can be divided into multiple grids, and the ground point of each grid can be determined. The local ground height and ground undulation of each grid can be determined based on the ground point. The local ground height can be used to characterize the height of each grid in the drivable area, and the ground undulation can be used to characterize the degree of undulation of the ground point.

[0043] For example, all drivable areas can be divided into multiple grids according to resolution. Clustering methods can be used to determine the set of points with the most points in each grid. Parts of the point cloud can be randomly extracted from the point cloud. Based on the partial point cloud, candidate ground points can be determined. The candidate ground points can be fitted to obtain the local ground equation for each grid.

[0044] Optionally, the point cloud can be filtered based on the local ground equation to determine the ground points in the point cloud. Based on the ground points and the distance between the ground points and the local ground equation, the local ground height and ground relief can be determined.

[0045] Step S106: Determine the target drivable area within the drivable area based on local ground height and ground undulation.

[0046] In the technical solution of step S106 of the present invention, after determining the local ground height and ground undulation based on ground points, the target drivable area can be determined from the drivable area by processing the local ground height and ground undulation. The target drivable area can be a drivable area where all obstacles affecting vehicle driving have been removed.

[0047] Optionally, acquiring all obstacles within the driving area allows for the determination of information such as the obstacle's height and location. Based on the obstacle's height, local ground elevation, and ground undulation, obstacles can be filtered, removing areas containing obstacles from the drivable area to obtain the target drivable area. For example, the minimum height of obstacles within the driving area can be obtained, and the relationship between the minimum height of all obstacles and the local ground elevation can be determined. If the minimum height is lower than the local ground elevation, the area containing that obstacle can be removed from the drivable area. The remaining obstacles can be clustered and segmented to filter out some obstacles. The drivable area obtained through the above process can be expanded equidistantly, and further clustering and segmentation steps can be performed to remove more obstacles, thus obtaining the target drivable area. It should be noted that this is only an illustrative example and does not impose specific limitations on the methods and processes for determining the areas where obstacles are located or the target drivable area.

[0048] Because structured roads are relatively flat, the driving route of a vehicle can be planned based on the drivable area of ​​the structured road. However, unstructured roads are more uneven, which may lead to inaccurate identification of obstacles on unstructured roads. Therefore, it is necessary to determine the target drivable area of ​​the road based on the local ground height and ground undulation. Thus, the embodiments of the present invention can make a secondary judgment on the drivable area based on the local ground height and ground undulation of ground points, so as to achieve more accurate identification of obstacles on the road and more accurate determination of the target drivable area. This solves the technical problem of inaccurate identification of unstructured roads and achieves the technical effect of improving the accuracy of unstructured road identification.

[0049] Step S108: Plan the driving route of the vehicle in the target drivable area.

[0050] In the above-described step S108 of the present invention, the driving route of the vehicle can be determined from the target drivable area, and the vehicle can drive based on the driving route.

[0051] For example, the vehicle's geographical coordinates can be obtained using its GPS information. These coordinates can be projected onto a high-definition map, and the vehicle's region of interest (ROI) can be defined. All roads within the ROI can be analyzed to determine whether they are structured or unstructured. The region enclosed by the boundaries of structured roads can be identified as the vehicle's ROI. Taking the union of all ROIs yields a structured drivable sub-region. ROIs containing unstructured roads can be marked and further discrimination can be performed to obtain... The unstructured drivable sub-region divides the drivable area into multiple grids. Ground candidate points are determined from all point clouds in the grid. By fitting all ground candidate points in a grid, the local ground equation of each grid can be obtained. The distance between the point cloud and the local ground equation of each grid can be determined. If the distance between the two is less than a preset threshold, the point cloud at this time can be determined as a ground point. Furthermore, the local ground height and ground undulation of all ground points can be determined, thereby removing the areas where obstacles are located in the drivable area and obtaining the target drivable area. In the target drivable area, the vehicle's driving route can be planned.

[0052] In steps S102 to S108 of this application, the drivable area of ​​the vehicle in the map is obtained; the ground points of the grids that make up the drivable area are determined, and the local ground height and ground undulation of the grids are determined based on the ground points; the target drivable area in the drivable area is determined based on the local ground height and ground undulation; and the driving route of the vehicle to be driven is planned in the target drivable area. In other words, by determining the drivable area of ​​the vehicle from the map, this embodiment of the invention can divide the area into multiple grids, determine the local ground height and ground undulation of each grid, and process the local ground height and ground undulation to determine the target drivable area of ​​the vehicle from the drivable area. Since the drivable area is divided into multiple grids for analysis, the problem of inaccurate determination of the target drivable area is avoided, thereby achieving the goal of more accurately planning the vehicle's driving route. This solves the technical problem of poor rationality in vehicle route planning and achieves the technical effect of improving the rationality of vehicle route planning.

[0053] The method described in this embodiment will be further described below.

[0054] As an optional embodiment, step S104, determining the ground points of the grid that makes up the drivable area, includes: dividing the drivable area into grids; performing clustering processing on each grid, and determining the local ground equation of each grid based on the obtained clustering processing results; calculating the point cloud in each grid based on the local ground equation to obtain the calculation result; and determining the point cloud as a ground point in response to the calculation result being less than a distance threshold.

[0055] In this embodiment, the drivable area of ​​the vehicle can be divided into multiple grids. Clustering can be performed on each grid to obtain clustering results. Based on the clustering results of each grid, the corresponding local ground equation can be determined. The local ground equation can be calculated on the point cloud within the grid, and the calculation result can be compared with a distance threshold. If the calculation result is less than the distance threshold, the point cloud can be identified as a ground point. The clustering result can serve as a candidate ground point. The local ground equation can be a local ground plane equation. The calculation result can be used to characterize the distance between each point cloud in the grid and its corresponding local ground equation. The distance threshold can be a preset value or a user-defined value; this is only an example and no specific limitation is imposed on the distance threshold.

[0056] Optionally, the entire drivable area of ​​the vehicle can be divided into m*n grids according to resolution. Since the drivable area is a three-dimensional range, the resulting grids are three-dimensional cylindrical. Each grid can be clustered, for example, using density clustering to find the set of points with the most points in the point cloud. A portion of the point cloud in the set can be randomly extracted as a seed point cloud. The seed point cloud can be fitted using a Random Sample Consensus (PANSAC) algorithm to obtain multiple ground candidate points. The ground candidate points can be fitted to determine the local ground plane equation within each grid. It should be noted that the clustering process and the algorithm and process for determining candidate points here are only illustrative examples and are not specifically limited. Any method and process for clustering each grid to obtain the local ground plane equation is within the protection scope of this invention.

[0057] Optionally, by using the local ground plane equation based on each grid, the distance from each point cloud in the grid to the local ground plane equation can be calculated. The result can be used to determine the relationship between the distance from all point clouds to the local ground screen equation and the distance threshold. If the distance between the two is less than the distance threshold, the point cloud can be identified as a ground point; if the distance between the two is greater than or equal to the distance threshold, the distance between the two can be used as an indicator to measure the ground undulation.

[0058] As an optional embodiment, step S104, determining the local ground height of the grid based on ground points, includes: determining the calculation result of the ground point in each grid in the local ground equation corresponding to the grid; and determining the average of the sum of all calculation results as the local ground height of the grid.

[0059] In this embodiment, after determining the point cloud with a calculation result less than the distance threshold as a ground point, the ground point in each grid and the corresponding local ground equation can be calculated to obtain the calculation result. The average value of all calculation results can be taken, and the average value can be determined as the local ground height of the grid.

[0060] Optionally, the distance between all point clouds in each grid and the corresponding local ground equation is determined, and the calculation results of all distances less than the distance threshold in a grid are summed and averaged. This average value can be determined as the local ground height of the grid.

[0061] For example, the local ground plane equation of the k-th grid can be set as F. k This allows us to determine the equation F of the local ground plane for each point cloud within the grid. k The distance between them can be used to map the i-th point cloud to F. k The distance between them is determined as d i,k For raster k, d can be extracted from that raster. i,k Points in the point cloud smaller than the distance threshold R0 are considered ground points, and the set of these ground points can be defined as {P}. i,k d i,k >R0}, the average of all calculation results in this set can be used as the local ground height H. k .

[0062] As an optional embodiment, step S104, determining the ground relief of the grid based on ground points, includes: determining the calculation results of the point cloud in each grid other than the ground points in the local ground equation corresponding to the grid; and determining the maximum value in the calculation results as the ground relief of the grid.

[0063] In this embodiment, after determining that the point cloud whose calculation result is less than the distance threshold is a ground point, the point cloud other than the ground point in each grid can be determined. The maximum value of the point cloud other than the ground point in the calculation result of the local ground equation corresponding to the grid can be determined as the ground undulation of the grid.

[0064] Optionally, the distance between all point clouds in each grid and the corresponding local ground equation can be determined. The maximum value among all calculated results where the distance within a grid is greater than or equal to a distance threshold can be determined as the ground relief of the grid.

[0065] For example, the local ground plane equation of the k-th grid can be F k , and the distance from each point cloud in the grid to the local ground plane equation F k can be determined. The distance from the i-th point cloud to F k can be determined as d i,k . For grid k, the point clouds in the grid with d i,k greater than or equal to the distance threshold R0 can be extracted. The maximum value of all distances d i,k less than the distance threshold in the grid can be determined as the ground undulation of the grid. Among them, the ground undulation of grid k can be expressed as R k = max{d i,k , d i,k > R0}.

[0066] For another example, the ground points and ground undulations {R k , 0 ≤ k < m*n} in all grids can be determined. For structured roads, the threshold of the ground undulation can be set as R1, and for unstructured roads, the threshold of the ground undulation can be set as R2. Since the ground undulation of unstructured roads is greater than that of structured roads, it can be determined that R1 < R2.

[0067] As an optional embodiment, in step S104, grids with a ground undulation greater than the ground undulation threshold of structured roads and less than the ground undulation threshold of unstructured roads are determined, and the grids are determined as unstructured sub-regions.

[0068] In this embodiment, the magnitude relationship between the ground undulation and the ground undulation thresholds of structured roads and unstructured roads can be judged. If the ground undulation of the grid is less than or equal to the ground undulation threshold of structured roads, the grid can be determined as a structured sub-region; if the ground undulation of the grid is greater than the ground undulation threshold of structured roads and less than the ground undulation threshold of unstructured roads, then the grid can be determined as an unstructured sub-region.

[0069] For example, since the grids of the drivable area are divided into m*n, the ground undulations of the m*n grids can be determined. For structured roads, the threshold of the ground undulation can be set as R1, and for unstructured roads, the threshold of the ground undulation can be set as R2. The magnitude relationship between R k and R1, R2 can be judged. If the ground undulation is less than the ground undulation threshold of structured roads, that is, R k < R1, then it can be determined that grid k is a structured sub-region; if the ground undulation is between the ground undulation threshold of structured roads and the ground undulation threshold of unstructured roads, that is, R1 < R kIf <R2, it can be determined that the grid k is an unstructured sub-region.

[0070] In the embodiment of the present invention, considering that the road is determined as an unstructured road or a structured road based on the ground undulation, the structured region and the unstructured sub-region in the drivable area can be discriminated again, thereby achieving the technical effect of improving the accuracy of identifying unstructured roads.

[0071] As an optional embodiment, step S106 of determining the target drivable area in the drivable area based on the local ground height and the ground undulation includes: performing obstacle perception on the obstacles in the drivable area to obtain a perception result; determining the center point position and height information of the obstacles based on the perception result; determining the target obstacle in the obstacles based on the center point position, height information, local ground height and ground undulation; and determining the target drivable area based on the target obstacle.

[0072] In this embodiment, when the local ground height and the ground undulation of all grids in the drivable area are determined, obstacle perception can be performed on the obstacles in the drivable area to obtain a perception result. The height information and the center point position of the obstacles can be determined from the perception result. By analyzing the center point position, height information, local ground height and ground undulation, the target obstacle in the obstacles can be determined, and thus the target driving area can be determined. Here, the perception result can be the bounding box of the obstacle. The height information can include the lowest height and the highest height of the perception result.

[0073] Optionally, obstacle perception can be performed on all obstacles in the drivable area. For example, perception can be performed through an artificial intelligence obstacle perception model to obtain the three-dimensional bounding box of all obstacles, and the center point of the three-dimensional bounding box, the lowest height and the highest height of the three-dimensional bounding box can be determined.

[0074] Optionally, through the local ground height and the ground undulation, the height information of the perception result can be compared with the local ground height to screen all obstacles in the drivable area. For example, the lowest height of the perception result can be compared with the local ground height. If the lowest height is less than the local ground height, the obstacle at this time can be removed; if the lowest height is greater than or equal to the local ground height, the obstacle at this time can be retained and determined as the target obstacle.

[0075] Considering that obstacles may fall into two categories: those that affect vehicle driving and those that do not, such as leaves or similar objects that do not affect vehicle movement, this invention allows for the filtering of obstacles by comparing the height information in the obstacle perception results with the local ground height. Obstacles that do not affect vehicle movement can be removed, eliminating the need to remove such obstacles when determining the vehicle's target drivable area. This improves the accuracy of obstacle identification around the vehicle.

[0076] In this embodiment of the invention, since each grid corresponds to a local ground height, when determining the target obstacle from among the obstacles, the minimum height of the obstacle in each grid in the drivable area can be compared with the local ground height of the grid. Obstacles with a minimum height less than the local ground height are eliminated, and the remaining obstacles are retained, thereby achieving the technical effect of improving the accuracy of obstacle identification.

[0077] As an optional embodiment, step S106, based on the center point location, height location, local ground height, and ground undulation, determines the target obstacle among the obstacles, including: in response to a first height in the height information being greater than the local ground height, determining the obstacle as a target obstacle, obtaining a first obstacle set containing at least one obstacle; clustering and separating the point cloud in the drivable area excluding ground points to obtain an initial obstacle set; filtering out ground undulation thresholds and unstructured sub-region ground undulation thresholds from the initial obstacle set whose center point location is located in an unstructured sub-region and whose second height in the height information is located in a structured sub-region. The obstacles between the two sets are used to obtain a second set of obstacles, wherein the first height of the same obstacle is less than the second height; the drivable area is expanded outward at equal intervals to determine the obstacle information within the target range, resulting in a third set of obstacles; obstacles intersecting the second and third sets of obstacles are identified as interfering obstacles, resulting in interfering obstacles, wherein interfering obstacles are obstacles that do not obstruct the vehicle's driving path; the second and third sets of obstacles are merged, and the interfering obstacle set is removed from the merged set of obstacles, resulting in a fourth set of obstacles; the fourth set of obstacles is merged with the first set of obstacles to obtain a target obstacle set with at least one target obstacle.

[0078] In this embodiment, the relationship between the first height information and the local ground height can be determined. If the first height information is greater than the local ground height, the obstacle can be identified as a target obstacle, and all target obstacles at this time can form a first obstacle set. The point cloud in the drivable area, excluding ground points, can be clustered to obtain an initial obstacle set. It can be determined whether the center point of the initial obstacle set is within an unstructured sub-region, and also the relationship between the second height information in the initial obstacle set and the ground undulation thresholds of the structured and unstructured sub-regions can be determined. If the center point of an obstacle in the initial obstacle set is located in an unstructured sub-region, and the second height information is between the ground undulation thresholds of the structured and unstructured sub-regions, the obstacle can be filtered out, and the remaining obstacles can form a second obstacle set. The drivable area can be expanded equidistantly to determine the obstacle information within the expanded target range, and the obstacles within the target range can form a third obstacle set. The intersection of the second and third obstacle sets can be identified, and the obstacles in this intersection are defined as interfering obstacles. All interfering obstacles can be grouped into an interfering obstacle set. The second and third obstacle sets can be merged, and the interfering obstacle set can be removed after merging to obtain a fourth obstacle set. Merging the fourth obstacle set with the first obstacle set yields the target obstacle set, where the first height information can be the minimum height, and the second height information can be the maximum height. The first height information of the same obstacle can be less than the second height information. Interfering obstacles can be obstacles that do not obstruct the vehicle's path, including leaves, rainwater, and snowflakes. It should be noted that this is only an example and does not impose specific restrictions on interfering obstacles. The ground undulation threshold for structured sub-regions and the ground undulation threshold for unstructured sub-regions can be preset values ​​or user-defined values; this is only an example and does not impose specific restrictions.

[0079] For example, the perception result of the i-th obstacle or the center point of the perception result can be represented as C. i (x i ,y i ,z i The lowest height in the perceived result can be represented as z. i,min The maximum height can be represented as z. i,max Local ground height H based on grid kIt can filter obstacles in the drivable area. For example, it can filter out obstacles whose minimum height is less than or equal to the local ground height, retain the remaining obstacles, and identify them as target obstacles. All target obstacles are defined as the first obstacle set. The first obstacle set can include obstacles whose minimum height according to the perception results is greater than the local ground height. The first obstacle set can be defined as S1 = {C} i , z i,min >H k}

[0080] For another example, clustering and isolating the remaining point cloud after removing ground points in the drivable area yields an initial obstacle set, which can be represented as S2. Alternatively, obstacles in the initial obstacle set can be filtered out whose center points lie in unstructured sub-regions of the marked unstructured road, and whose maximum height falls between the ground undulation thresholds of the structured and unstructured sub-regions. Obstacles meeting these conditions can then be grouped into a second obstacle set. This second obstacle set can include obstacles from the initial obstacle set whose perceived maximum height falls between the ground undulation thresholds of the structured and unstructured roads. This second obstacle set can also be represented as S2. ^ ={C i ∈S2,R1 <z i,max <R2}。

[0081] As an optional example, the drivable area obtained based on the above steps can be expanded outward at equal intervals to obtain the drivable area after the equal-distance expansion. The remaining point cloud after removing ground points in the drivable area can be clustered and isolated to obtain a third obstacle set, which can be represented as S3. The intersection between S2 and S3 can be determined as the interference obstacle set.

[0082] In this embodiment of the invention, the second obstacle set and the third obstacle set can be fused. Since there is an overlap between the obstacles in S2 and S3, if the overlap meets a certain threshold requirement and the grid occupied includes both the grid inside and outside the drivable area, then the obstacles in the overlap can be obstacles such as vegetation and rocks extending from outside the road boundary into the road boundary, i.e., interference obstacles. The interference obstacle set in the set obtained after the two obstacle sets are fused can be removed to obtain the fourth obstacle set, which can be represented as S4. S1 and S4 can be fused to obtain the target obstacle set, thereby achieving the technical effect of improving the accuracy of obstacle recognition on unstructured roads.

[0083] In this embodiment of the invention, the drivable area of ​​the vehicle in the map is obtained; the ground points of the grids that make up the drivable area are determined, and the local ground height and ground undulation of the grids are determined based on the ground points; the target drivable area in the drivable area is determined based on the local ground height and ground undulation; and the driving route of the vehicle to be driven is planned in the target drivable area. In other words, this embodiment of the invention determines the drivable area of ​​the vehicle from the map, divides the area into multiple grids, determines the local ground height and ground undulation of each grid, and processes the local ground height and ground undulation to determine the target drivable area of ​​the vehicle from the drivable area. Because the drivable area is divided into multiple grids for analysis, the problem of inaccurate determination of the target drivable area is avoided, thereby achieving the goal of more accurate planning of the vehicle's driving route. This solves the technical problem of poor rationality in vehicle route planning and achieves the technical effect of improving the rationality of vehicle route planning.

[0084] Example 2

[0085] The technical solutions of the embodiments of the present invention will be illustrated below with reference to preferred embodiments.

[0086] Autonomous driving scenarios are divided into structured roads and unstructured roads (including fully unstructured and semi-structured roads).

[0087] Currently, there are many mature solutions for structured roads, such as highways with lane edge lines and lane center lines. However, for unstructured roads, there are still problems in ensuring the completeness of the methods and quickly planning drivable routes that meet vehicle constraints.

[0088] Among related technologies, route planning includes curve-based methods, sampling and search-based methods, and optimization-based methods. These methods utilize the geometric rules of curves to quickly find collision-free feasible solutions on unstructured roads, but they cannot solve the technical problem of poor rationality in vehicle route planning.

[0089] In one related technology, a method for planning autonomous driving routes on unstructured roads is proposed. The method includes the following steps: receiving route planning input information; acquiring obstacle information, extracting the centerlines of gaps between obstacles using a dynamic step size module and processing them to obtain a discretized route, and determining whether the processing time exceeds the allowed system time; if the processing time does not exceed the set allowed system time, then performing non-uniform sampling route planning based on the dynamic step size to obtain an autonomous driving route; if the dynamic step size module times out, determining whether there is a route to be driven and its safety, to decide whether to execute the route to be driven or to apply braking; if the dynamic step size module exits successfully, then executing the autonomous driving route. This method for planning autonomous driving routes on unstructured roads has strong planning capabilities and a fast response speed.

[0090] In another related technology, an unstructured road vehicle route planning method is proposed, including the following steps: inputting a map containing obstacles and road boundaries, setting the initial configuration Ns, the target configuration Ng, and the target region Ag; setting the initial set to contain only the initial configuration Ns; selecting the node with the minimum cost from the initial set; determining whether node Ni already exists in the initial set; determining whether node Ni has reached the target region Ag, and generating a parking route that accurately reaches the target configuration using a "straight line-circular arc-straight line" parking model; determining whether any of the generated parking routes collide with obstacles; and route backtracking. This unstructured road vehicle route planning method can effectively realize vehicle route planning.

[0091] However, all of the aforementioned related technologies suffer from poor rationality in vehicle route planning.

[0092] To address the aforementioned problems, this invention provides an autonomous driving LiDAR obstacle perception method for simple rural roads. This method includes the following steps: acquiring potential rural road areas through map data and marking the approximate locations of the rural roads; determining the starting and ending distances of the rural road ends relative to the vehicle by identifying the degree of ground undulation, i.e., secondary confirmation of the rural road areas on the map; further processing the road boundaries on the map and fusing the obstacle perception results to remove the influence of vegetation, rocks, etc., on the road boundaries—that is, perception targets that autonomous driving does not need to focus on—while retaining the perception targets that truly require driving decisions, planning the autonomous driving route, and avoiding unnecessary obstacle avoidance or braking. This solves the technical problem of poor rationality in vehicle route planning and achieves the technical effect of improving the rationality of vehicle route planning.

[0093] The embodiments of the present invention will be further described below.

[0094] In an embodiment of the present invention, Figure 2This is a flowchart illustrating a method for obtaining the perception range of structured roads using high-precision map data according to an embodiment of the present invention. Figure 2 As shown, the method may include the following steps:

[0095] Step S202: Obtain the vehicle's geographical coordinates based on the vehicle's GPS information.

[0096] In the technical solution provided by step S202 of the present invention, the geographical coordinates of the vehicle can be obtained through the vehicle's global positioning system information.

[0097] Step S204: Project the geographic location coordinates onto the high-precision map to obtain the local map location of the vehicle.

[0098] In the technical solution provided by step S204 of the present invention, the geographical coordinates of the vehicle can be projected onto a high-precision map, a local map position of the vehicle can be divided on the high-precision map, and the region of interest of the vehicle can be determined at the local map position. The region of interest can contain information on all roads within the target range surrounding the geographical location of the vehicle.

[0099] Step S206: Based on road and vehicle information, the drivable area of ​​the vehicle can be determined.

[0100] In the technical solution provided by step S206 of the present invention, based on the information of each road, it can be determined whether the road is a structured road or an unstructured road, the road boundaries of all roads in the region of interest can be obtained, the area within each road boundary can be set as a region of interest for the vehicle, and by connecting the regions of interest in all regions of interest and taking the union, the drivable area of ​​the vehicle in the map can be determined.

[0101] Optionally, if the information of all roads within the vehicle's region of interest is analyzed and it is determined that there are structured roads in the region, the vehicle's region of interest can be determined based on the road boundaries of the structured roads. By taking the union of the regions of interest of all structured roads, the vehicle's structured drivable sub-region can be determined.

[0102] For example, Figure 3 This is a schematic diagram illustrating roadside weeds extending onto the road according to an embodiment of the present invention, as shown below. Figure 3 As shown, the road is flanked by messy grass, with shaded areas where the grass extends into the road. The grass is characterized by being short and soft. In this case, the grass extending into the road can be identified as an obstacle.

[0103] Optionally, if the information of all roads within the region of interest of the vehicle is analyzed and it is found that there are unstructured roads in the region, the region of interest can be temporarily marked to facilitate subsequent secondary discrimination of the road boundaries in the region, further determining whether there are unstructured roads in the region. If so, the mark can be retained, and the area within the road boundary of the unstructured road can be determined as the unstructured sub-region of the vehicle. In this embodiment of the invention, by taking into account the possibility of unstructured roads existing within the region of interest and performing secondary discrimination, the technical effect of improving the accuracy of unstructured road identification is achieved.

[0104] For example, Figure 4 This is a schematic diagram of a region of interest containing unstructured roads according to an embodiment of the present invention, such as... Figure 4 As shown, there are structured and unstructured roads in the region of interest. The two roads are flanked by messy grass. The shaded areas are weeds, shrubs, and branches that extend into the roads. The drivable area in this case can be marked to facilitate secondary identification of the region of interest later.

[0105] In this embodiment of the invention, all drivable areas can be divided into multiple grids according to resolution. The set of points with the most points in each grid can be determined by clustering methods. A portion of the point cloud can be randomly extracted from the point cloud. Based on the portion of the point cloud, candidate ground points can be determined. The candidate ground points can be fitted to obtain the local ground equation for each grid. Based on the local ground equation, the point cloud can be filtered to obtain ground points. Based on the ground points and the distance between the ground points and the local ground equation, the local ground height and ground undulation can be determined.

[0106] For example, the purpose of the above steps is to identify the degree of road undulation, and to output the local ground height and undulation. Unlike smooth structured ground, unstructured ground may include road conditions such as potholes in cement roads and gravel roads. Figure 5(a) is a schematic diagram of the scanning point distribution of a mechanical rotating lidar according to an embodiment of the present invention. As shown in Figure 5(a), when the point cloud generated by the mechanical rotating lidar is evenly distributed on a scanning circle of equal radius, that is, on an equidistant circle at a distance from the vehicle, all adjacent point clouds have equal horizontal projection distance and equal arc length distance. Figure 5(b) is a schematic diagram of the scanning point distribution of a solid-state lidar according to an embodiment of the present invention. As shown in Figure 5(b), the point cloud scanned by the solid-state lidar is unevenly distributed. On the same line of sight, the distribution of the point cloud does not show a linear relationship from near to far. On unstructured roads, relying on solid-state lidar to perceive obstacles can lead to difficulty in distinguishing small obstacles and slight undulations on the road surface.

[0107] In an embodiment of the present invention, Figure 6 This is a flowchart of a method for determining road types based on local ground height and ground undulation according to an embodiment of the present invention, such as... Figure 6 As shown, the method may include the following steps:

[0108] Step S602: Divide the region of interest into grids.

[0109] In the technical solution provided by step S602 of the present invention, the entire drivable area of ​​the vehicle can be divided into m*n grids according to the resolution. Since the drivable area is a three-dimensional range, the divided grids are three-dimensional columns.

[0110] Step S604: Fit a local plane within the grid to determine the grid ground point, local ground height, and ground undulation.

[0111] In the technical solution provided by step S604 of the present invention, a local plane can be fitted within the grid to further determine the ground points of the grid, and the local ground height and ground undulation can be determined for the ground points.

[0112] Optionally, each grid cell can be clustered, for example, by density clustering, to find the set of points with the most points in the point cloud. A portion of the point cloud in the set can be randomly extracted as a seed point cloud. Multiple ground candidate points can be obtained by using a random sampling consensus algorithm on the seed point cloud. The ground candidate points can be fitted to determine the local ground plane equation within each grid cell.

[0113] Optionally, by using the local ground plane equation based on each grid, the distance from each point cloud in the grid to the local ground plane equation can be calculated. The result can be used to determine the relationship between the distance from all point clouds to the local ground screen equation and the distance threshold. If the distance between the two is less than the distance threshold, the point cloud can be identified as a ground point; if the distance between the two is greater than or equal to the distance threshold, the distance between the two can be used as an indicator to measure the ground undulation.

[0114] Optionally, the distance between all point clouds in each grid and the corresponding local ground equation is determined. The calculation results of all distances less than a distance threshold in a grid are summed and the average value is taken. This average value can be determined as the local ground height of the grid. The distance between all point clouds in each grid and the corresponding local ground equation is determined. The maximum value among the calculation results of all distances greater than or equal to the distance threshold in a grid can be determined as the ground relief of the grid.

[0115] Step S606: Determine whether the road is a structured road or an unstructured road.

[0116] In the technical solution provided by step S606 of the present invention, the relationship between the ground undulation degree and the ground undulation degree threshold of the structured road and the ground undulation degree threshold of the unstructured road can be determined. If the ground undulation degree of the grid is less than or equal to the ground undulation degree threshold of the structured road, the grid can be identified as a structured sub-region, that is, the road is a structured road. If the ground undulation degree of the grid is greater than the ground undulation degree threshold of the structured road and less than the ground undulation degree threshold of the unstructured road, the grid can be identified as an unstructured sub-region, that is, the road is an unstructured road.

[0117] In this embodiment of the invention, obstacle perception can be performed on all obstacles within the drivable area. For example, an artificial intelligence obstacle perception model can be used to perceive obstacles and obtain a three-dimensional bounding box for all obstacles. The center point, minimum height, and maximum height of the three-dimensional bounding box can be determined. By comparing the height information of the perceived result with the local ground height using local ground elevation and undulation, all obstacles within the drivable area can be filtered. For example, the minimum height of the perceived result can be compared with the local ground height. If the minimum height is less than the local ground height, the obstacle can be removed. If the minimum height is greater than or equal to the local ground height, the obstacle can be retained and identified as a target obstacle.

[0118] Optionally, the relationship between the lowest height and the local ground height in the height information can be determined. If the lowest height is greater than the local ground height, the obstacle can be identified as a target obstacle. All target obstacles at this point can be grouped into a first obstacle set. The point cloud in the drivable area, excluding ground points, can be clustered and segmented to obtain an initial obstacle set. It can be determined whether the center point of the initial obstacle set is within an unstructured sub-region. Alternatively, the relationship between the highest height in the height information of the initial obstacle set and the ground undulation thresholds of the structured and unstructured sub-regions can be determined. If the center point of an obstacle in the initial obstacle set is within an unstructured sub-region and the highest height is within a structured sub-region... The obstacle can be filtered between the ground undulation of the region and the ground undulation of the unstructured sub-region. The remaining obstacles can form a second obstacle set. The drivable area can be expanded outward at equal intervals. The obstacle information of the obstacles within the target range after the equal-distance expansion can be determined. The obstacles within the target range can be formed into a third obstacle set. The intersection of the second obstacle set and the third obstacle set can be determined. The obstacles in the intersection are identified as interference obstacles. All interference obstacles can be formed into an interference obstacle set. The second obstacle set and the third obstacle set can be merged. After merging, the interference obstacle set can be removed to obtain a fourth obstacle set. The fourth obstacle set and the first obstacle set can be merged to obtain the target obstacle set.

[0119] Optionally, the second obstacle set and the third obstacle set can be merged. Since there is overlap between the obstacles in the second and third obstacle sets, if the overlapping part meets a certain threshold requirement and the occupied grid includes both the grid within and outside the drivable area, then the obstacles in the overlapping part can be obstacles such as vegetation and rocks extending from outside the road boundary into the road boundary, i.e., interference obstacles. The interference obstacle set in the merged set can be removed to obtain the fourth obstacle set. Figure 7 This is a schematic diagram of different road areas within a region of interest according to an embodiment of the present invention, such as... Figure 7 As shown, the overlapping portions of the second obstacle set S1 and the third obstacle set S2, but not included in the first obstacle set S1, can be identified as the obstacle set to be removed, i.e., the interfering obstacle set. The fourth obstacle set can be fused with the first obstacle set to obtain the target obstacle set, thereby achieving the technical effect of improving the accuracy of obstacle recognition on unstructured roads.

[0120] For example, Figure 8 This is a schematic diagram of fallen leaves extending from a road boundary onto the road according to an embodiment of the present invention, such as... Figure 8As shown, there is a road in the region of interest, with messy grass on both sides of the road. There are shaded areas containing obstacles such as leaves that extend into the road. Based on this situation, the obstacles can be removed.

[0121] This invention, through its embodiments, determines the drivable area of ​​a vehicle from a map. This area can be divided into multiple grids, and the local ground height and undulation of each grid are determined. By processing the local ground height and undulation, the target drivable area of ​​the vehicle can be determined from the drivable area. Because the drivable area is divided into multiple grids for analysis, the problem of inaccurate determination of the target drivable area is avoided, thereby achieving the goal of more accurately planning the vehicle's driving route. This solves the technical problem of poor rationality in vehicle route planning and achieves the technical effect of improving the rationality of vehicle route planning.

[0122] Example 3

[0123] According to an embodiment of the present invention, a device for determining a driving route is also provided. It should be noted that this device for determining a driving route can be used to execute the method for determining a driving route in Embodiment 1.

[0124] Figure 9 This is a schematic diagram of a route determination device according to an embodiment of the present invention. Figure 9 As shown, the route determination device 900 may include: an acquisition unit 902, a first determination unit 904, a second determination unit 906, and a planning unit 908.

[0125] Acquisition unit 902 is used to acquire the drivable area of ​​the vehicle on the map.

[0126] The first determining unit 904 is used to determine the ground points of the grid that makes up the drivable area, and to determine the local ground height and ground undulation of the grid based on the ground points.

[0127] The second determining unit 906 is used to determine the target drivable area in the drivable area based on the local ground height and ground undulation.

[0128] Planning unit 908 is used to plan the driving route of the vehicle in the target drivable area.

[0129] Optionally, the first determining unit 904 may include: a partitioning module for partitioning the drivable area into grids; a first processing module for performing clustering processing on each grid and determining the local ground equation for each grid based on the obtained clustering processing results; a calculation module for calculating the point cloud in each grid based on the local ground equations and obtaining the calculation results; and a first determining module for determining the point cloud as a ground point in response to the calculation result being less than a distance threshold.

[0130] Optionally, the first determining unit 904 may include: a second determining module, used to determine the calculation result of the ground point in each grid in the local ground equation corresponding to the grid; and a third determining module, used to determine the average value of the sum of all calculation results as the local ground height of the grid.

[0131] Optionally, the first determining unit 904 may include: a fourth determining module, used to determine the calculation results of the point cloud in each grid other than the ground point in the local ground equation corresponding to the grid; and a fifth determining module, used to determine the maximum value in the calculation results as the ground relief of the grid.

[0132] Optionally, the device further includes a third determining unit, used to determine grids whose ground undulation is greater than the structured road ground undulation threshold and less than the unstructured road ground undulation threshold, and to determine the grids as unstructured sub-regions.

[0133] Optionally, the second determining unit 906 may include: a sensing module for sensing obstacles in the drivable area and obtaining sensing results; a sixth determining module for determining the center point position and height information of the obstacle based on the sensing results; a seventh determining module for determining the target obstacle in the obstacles based on the center point position and height information, local ground height and ground undulation; and an eighth determining module for determining the target drivable area based on the target obstacle.

[0134] Optionally, the seventh determining module may include: a first determining submodule, configured to determine an obstacle as a target obstacle in response to a first height in the height information being greater than the local ground height, thereby obtaining a first obstacle set containing at least one obstacle; a clustering and separating submodule, configured to cluster and separate the point cloud in the drivable area excluding ground points, thereby obtaining an initial obstacle set; and a filtering submodule, configured to filter out obstacles in the initial obstacle set whose center point is located in an unstructured sub-region and whose second height in the height information is between the ground relief threshold of the structured sub-region and the ground relief threshold of the unstructured sub-region, thereby obtaining a second obstacle set, wherein the first height of the same obstacle... The height is less than the second height; the second determining submodule is used to expand the drivable area outward at equal intervals, determine the obstacle information within the target range, and obtain the third obstacle set; the third determining submodule is used to determine the obstacles intersecting the second obstacle set and the third obstacle set as interfering obstacles, and obtain interfering obstacles, wherein the interfering obstacles are obstacles that cannot obstruct the vehicle's driving path; the processing submodule is used to merge the second obstacle set and the third obstacle set, and remove the interfering obstacle set from the obstacle set obtained after merging, and obtain the fourth obstacle set; the fusion submodule is used to merge the fourth obstacle set and the first obstacle set to obtain the target obstacle set with at least one target obstacle.

[0135] In this embodiment of the invention, the acquisition unit acquires the drivable area of ​​the vehicle in the map, the first determination unit determines the ground points of the grid that makes up the drivable area, and determines the local ground height and ground undulation of the grid based on the ground points, the second determination unit determines the target drivable area in the drivable area based on the local ground height and ground undulation, and the planning unit plans the driving route of the vehicle to be driven in the target drivable area, thereby solving the technical problem of poor rationality of vehicle route planning and achieving the technical effect of improving the rationality of vehicle route planning.

[0136] Example 4

[0137] According to an embodiment of the present invention, a computer-readable storage medium is also provided, the storage medium including a stored program, wherein the program executes the method for determining the driving route described in Embodiment 1.

[0138] Example 5

[0139] According to an embodiment of the present invention, a processor is also provided for running a program, wherein the program executes the method for determining the driving route described in Embodiment 1.

[0140] Example 6

[0141] According to an embodiment of the present invention, a vehicle is also provided, which is used to perform the method for determining the driving route according to the embodiments of the present invention.

[0142] The sequence numbers of the above embodiments of the present invention are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0143] In the above embodiments of the present invention, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0144] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. The device embodiments described above are merely illustrative; for example, the division of units can be a logical functional division, and in actual implementation, there may be other division methods. For instance, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the displayed or discussed mutual coupling, direct coupling, or communication connection may be through some interfaces; the indirect coupling or communication connection between units or modules may be electrical or other forms.

[0145] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0146] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0147] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, read-only memory (ROM), random access memory (RAM), portable hard drives, magnetic disks, or optical disks.

[0148] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method of determining a travel route, characterized by, The method comprises: acquiring a drivable area of a vehicle in a map; determining a ground point of a grid constituting the drivable area, and determining a local ground height and a ground roughness of the grid based on the ground point; determining a target drivable area in the drivable area based on the local ground height and the ground roughness; planning a driving route to be driven by the vehicle in the target drivable area; determining a target drivable area in the drivable area based on the local ground height and the ground roughness comprises: determining a target obstacle in the obstacles based on a center point position of the obstacle, height information of the obstacle, the local ground height and the ground roughness, wherein the center point position and the height information are obtained by obstacle perception on the obstacles in the drivable area; and determining the target drivable area based on the target obstacle; determining a target obstacle in the obstacles based on a center point position of the obstacle, height information of the obstacle, the local ground height and the ground roughness comprises: determining the obstacle as a target obstacle in response to a first height in the height information being greater than the local ground height, to obtain a first obstacle set containing at least one obstacle; clustering and separating point clouds in the drivable area except the ground point, to obtain an initial obstacle set; filtering out obstacles in the initial obstacle set whose center point position is located in an unstructured sub-region and whose second height in the height information is located between a ground roughness threshold of the structured sub-region and a ground roughness threshold of the unstructured sub-region, to obtain a second obstacle set, wherein the first height of the same obstacle is less than the second height; performing equidistant outward expansion on the drivable area to determine obstacle information within a target range, to obtain a third obstacle set; determining obstacles in the second obstacle set and the third obstacle set as interference obstacles, to obtain an interference obstacle set, wherein the interference obstacle is an obstacle that cannot block the driving route of the vehicle; fusing the second obstacle set and the third obstacle set, and removing the interference obstacle set from the obstacle set obtained after fusion, to obtain a fourth obstacle set; and fusing the fourth obstacle set and the first obstacle set, to obtain a target obstacle set containing at least one target obstacle.

2. The method of claim 1, wherein, The method further comprises: dividing the drivable area to obtain a plurality of grids; performing clustering processing on each grid, and determining a local ground equation of each grid based on the clustering processing result obtained; calculating point clouds in each grid based on the local ground equation, to obtain a calculation result; determining the point clouds as the ground point in response to the calculation result being less than a distance threshold.

3. The method of claim 2, wherein, The method further comprises: determining a calculation result of the ground point in the local ground equation corresponding to the grid in each grid; and An average of a sum of all the calculation results is determined as the local ground height of the grid.

4. The method of claim 2, wherein, The ground relief of the grid is determined based on the ground points, including: determining a calculation result of the point cloud in each grid except the ground points in the local ground equation corresponding to the grid; a maximum value in the calculation results is determined as the ground relief of the grid.

5. The method according to any one of claims 1 to 4, characterized in that, The method further includes: determining a grid with the ground relief greater than a structured road ground relief threshold and less than an unstructured road ground relief threshold as an unstructured sub-region.

6. The method of claim 1, wherein, The method further includes: obstacle perception is performed on obstacles in the drivable area to obtain a perception result; center point positions and height information of the obstacles are determined based on the perception result.

7. A travel route determination device characterized by comprising: including: an acquisition unit configured to acquire a drivable area of a vehicle in a map; a first determination unit configured to determine ground points of grids constituting the drivable area, and determine local ground heights and ground reliefs of the grids based on the ground points; a second determination unit configured to determine a target drivable area in the drivable area based on the local ground heights and the ground reliefs; a planning unit configured to plan a driving route to be driven by the vehicle in the target drivable area; the second determination unit is further configured to determine a target obstacle in the obstacles based on center point positions of the obstacles, height information of the obstacles, the local ground heights and the ground reliefs, wherein the center point positions and the height information are obtained by performing obstacle perception on the obstacles in the drivable area; and determine the target drivable area based on the target obstacle; the second determination unit is further configured to, in response to a first height in the height information being greater than the local ground height, determine the obstacle as the target obstacle to obtain a first obstacle set containing at least one obstacle; perform clustering and separation on a point cloud in the drivable area except the ground points to obtain an initial obstacle set; filter out obstacles in the initial obstacle set with the center point positions located in an unstructured sub-region and the second height in the height information located between a ground relief threshold of the structured sub-region and a ground relief threshold of the unstructured sub-region to obtain a second obstacle set, wherein the first height of the same obstacle is less than the second height; perform equidistant outward expansion on the drivable area to determine obstacle information in a target range to obtain a third obstacle set; determine obstacles intersected between the second obstacle set and the third obstacle set as interference obstacles to obtain an interference obstacle set, wherein the interference obstacles are obstacles that cannot block the driving route of the vehicle; fuse the second obstacle set and the third obstacle set, and remove the interference obstacle set from an obstacle set obtained after the fusion to obtain a fourth obstacle set; and fuse the fourth obstacle set and the first obstacle set to obtain a target obstacle set containing at least one target obstacle.

8. A computer-readable storage medium, characterized in that, The computer readable storage medium includes a stored program, wherein the program, when executed, controls a device in which the computer readable storage medium is located to perform the method of any one of claims 1 to 6.

9. A vehicle characterized by comprising: A computer program product for performing the method of any one of claims 1 to 6. A computer program product for performing the method of any one of claims 1 to 6.

Images