Loading area parking position calculation method and device, equipment and storage medium

Abstract

The invention discloses a loading area parking position calculation method and device, equipment and a storage medium, and the method comprises the steps: taking a loading point course value of a loaded vehicle as a direction, taking each preset loading point as a base point, carrying out the region division, obtaining three rectangular regions, and obtaining each non-ground point cloud clustering target on a backward trajectory; determining the category of each non-ground point cloud clustering target based on the position of each non-ground point cloud clustering target and the comparison result of each non-ground point cloud clustering target and the preset height threshold in the three rectangular areas; obtaining a vehicle model point cloud and a mountain voxel grid, farthest from the preset loading point position, of a backward trajectory of a loaded vehicle; and starting from the preset loading point position, detecting the position relationship between each point of the vehicle model point cloud corresponding to the backward trajectory point and the mountain voxel grid one by one along the backward trajectory so as to determine an optimal stop point. According to the invention, the material mountain and the common obstacle can be effectively distinguished, and the loading point position parking can be accurately carried out.

Description

Technical Field

[0001] This invention relates to the technical field of intelligent mining truck transportation, and in particular to a method, apparatus, device, and storage medium for calculating the parking position in the loading area. Background Technology

[0002] With the intelligent development of open-pit mining operations, unmanned vehicles are increasingly widely used in mining transportation. The mining truck working face mainly consists of three areas: loading area, unloading area, and structured road. The equipment area is where the material stripping surface is located. After the mining truck reaches the appropriate loading position, it uses an electric shovel to load the material. After that, it moves to the unloading area to unload the material. Before entering the loading area, the loading position is determined by the electric shovel operator, and a driving route from the entry point to the loading point is generated. However, since the loading position is determined manually by the electric shovel operator, the set point may be located on the bottom material (coal / rock). During the process of the mining truck reversing to the preset loading point, it may mistake the material for an obstacle, causing the vehicle to stop prematurely and preventing the electric shovel from loading. Alternatively, it may not recognize the material as an obstacle, and after reaching the loading point, the vehicle tires may run over it, resulting in rocks getting stuck and causing tire wear. Therefore, given the probability that the preset loading location may be inaccurate, it is essential to accurately calculate the loading and stopping location of the unmanned mining truck during its reversal in the loading area, so that it can stop loading at a more reasonable location. Summary of the Invention

[0003] The purpose of this invention is to provide at least one method, apparatus, device, and storage medium for calculating the loading area parking position, which can at least solve the technical problem of inaccurate calculation of loading position points in the prior art.

[0004] To address the aforementioned technical problems, at least one embodiment of this application provides a method for calculating the docking location of a loading area, comprising: Obtain each preset loading point and the heading value and backward trajectory line of the loading vehicle based on each preset loading point; Using the heading value of the loading point of the loaded vehicle as the direction, and taking each of the preset loading points as the base point, the area is divided into three rectangular areas. The length direction of the three rectangular areas is along the heading value of the loading point, and the width direction of the three rectangular areas is along the direction perpendicular to the heading value of the loading point. The length and width of the three rectangular areas are different. For each of the preset loading points, obtain the non-ground point cloud cluster targets on the backward trajectory line of the loading vehicle; The category of each non-ground point cloud cluster target is determined based on the location of each non-ground point cloud cluster target and the comparison results between each non-ground point cloud cluster target and the preset height threshold in the three rectangular regions. The categories of the non-ground point cloud cluster targets include mountains and obstacles. For each of the preset loading points, obtain the vehicle model point cloud and the mountain voxel grid that is furthest from the preset loading point on the backward trajectory line of the loaded vehicle. Starting from the preset loading point, the positional relationship between each point of the vehicle model point cloud corresponding to each point of the backward trajectory and the mountain voxel grid is detected, thereby determining the optimal stopping point corresponding to each preset loading point.

[0005] At least one embodiment of this application also provides a loading area docking position calculation device, comprising: The first acquisition module is used to acquire each preset loading point and the loading point heading value and backward trajectory line determined by the loading vehicle based on each preset loading point; The region division module is used to divide the region into three rectangular regions, with the heading value of the loading point of the loaded vehicle as the direction and each preset loading point as the base point. The length direction of the three rectangular regions is along the heading value of the loading point, and the width direction of the three rectangular regions is along the direction perpendicular to the heading value of the loading point. The length and width of the three rectangular regions are different. The clustering target determination module is used to obtain the non-ground point cloud clustering targets on the backward trajectory line of the loading vehicle for each of the preset loading points. The target category determination module is used to determine the category of each non-ground point cloud cluster target based on the location of each non-ground point cloud cluster target and the comparison result between each non-ground point cloud cluster target and a preset height threshold in the three rectangular regions. The categories of the non-ground point cloud cluster targets include mountains and obstacles. The second acquisition module is used to acquire, for each of the preset loading points, the point cloud of the vehicle model and the voxel grid of the mountain where the backward trajectory line of the loaded vehicle is furthest from the preset loading point. The stopping point determination module is used to detect the positional relationship between each point of the vehicle model point cloud and the mountain voxel grid, starting from the preset loading point and moving along the backward trajectory line, thereby determining the optimal stopping point corresponding to each preset loading point.

[0006] At least one embodiment of this application also provides an electronic device, including: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform the above-described loading area docking position calculation method.

[0007] At least one embodiment of this application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described method for calculating the loading area docking position.

[0008] The loading area parking position calculation method, device, electronic device, and computer-readable storage medium provided in the embodiments of this application can effectively distinguish between material hills and ordinary obstacles. While ensuring improved accuracy of hill collision analysis, it reduces the probability of missing obstacles and avoids vehicles getting too close to obstacles. Furthermore, based on the planned trajectory, it can perform dynamic analysis on the material hills scanned in real time during vehicle operation and combine it with the vehicle's shape for collision analysis. This allows for more accurate parking at loading points, preventing vehicles from stopping too early and failing to load materials or from running over materials and causing tire wear due to failure to stop in time.

[0009] In some optional embodiments, the three rectangular regions are respectively a first rectangular region, a second rectangular region, and a third rectangular region. The process of dividing the region into three rectangular regions using the heading value of the loading point of the loaded vehicle as the direction and each preset loading point as a base point, wherein the length direction of the three rectangular regions is along the heading value of the loading point, and the width direction of the three rectangular regions is along the direction perpendicular to the heading value of the loading point, and the length and width of the three rectangular regions are different, includes: The width of the first rectangular region is greater than the width of the second rectangular region, the width of the second rectangular region is greater than the width of the third rectangular region, and the third rectangular region is composed of three sub-regions with equal widths. The three sub-regions are arranged side by side along the direction of the backward trajectory line. The three sub-regions are the first sub-region, the second sub-region, and the third sub-region. The two sides of the second sub-region along the length direction overlap with the one side of the first sub-region along the length direction and the one side of the third sub-region along the length direction, respectively. The length of the first sub-region is equal to the length of the first rectangular region, and the sum of the lengths of the second and third sub-regions is equal to the length of the second rectangular region.

[0010] In some optional embodiments, the step of dividing the area into three rectangular regions based on the heading value of the loading point of the loaded vehicle and each preset loading point as a base point, wherein the length direction of the three rectangular regions is along the heading value of the loading point, and the width direction of the three rectangular regions is along the direction perpendicular to the heading value of the loading point, and the length and width of the three rectangular regions are different, further includes: The width of the first rectangular area is the sum of the width of the vehicle body and a first preset distance; the width of the second rectangular area is the sum of the width of the vehicle body and a second preset distance; and the width of the third rectangular area is the sum of the width of the vehicle body and a third preset distance. The first preset distance is greater than the second preset distance, and the second preset distance is greater than the third preset distance. The length of the first rectangular area is the sum of a first distance and a second distance. The first distance is a preset distance along the backward trajectory line with the preset loading point as the base point, and the second distance is a preset distance in the opposite direction of the backward trajectory line with the preset loading point as the base point. The length of the second rectangular area is preset to be the third distance, which is a preset distance along the backward trajectory line with the preset loading point as the base point; The length of the first sub-region is preset to be equal to the length of the first rectangular region, the length of the second sub-region is a fourth distance, the fourth distance is a preset distance along the retreat trajectory line, the length of the third sub-region is a fifth distance, the fifth distance is a preset distance along the retreat trajectory line, and the sum of the fourth distance and the fifth distance is equal to the length of the second rectangular region.

[0011] In some optional embodiments, the step of determining the category of each non-ground point cloud cluster target based on the location of each non-ground point cloud cluster target and the comparison result of each non-ground point cloud cluster target with a preset height threshold in the three rectangular regions, wherein the category of the non-ground point cloud cluster target includes mountains and obstacles, includes: Detect whether each point of the non-ground point cloud clustering target is within the second rectangular region; When each point of the non-ground point cloud cluster target is within the second rectangular area, the elevation value of the lowest point of the non-ground point cloud cluster target within the second rectangular area is determined to be the elevation value of the target's lowest point. When none of the points of the non-ground point cloud cluster target are within the second rectangular area, the elevation value of the lowest point of the non-ground point cloud cluster target is determined to be the elevation value of the lowest point of the target. Based on the sub-region where each of the non-ground point cloud cluster targets is located, the difference between the highest point elevation value and the lowest point elevation value of the target in the corresponding sub-region is calculated to obtain the relative height; The category of each non-ground point cloud cluster target is determined based on the comparison between the location of each non-ground point cloud cluster target and the relative height with the preset height threshold in each sub-region.

[0012] In some optional embodiments, the step of determining the category of each non-ground point cloud cluster target based on the comparison result of the location of each non-ground point cloud cluster target and the relative height with a preset height threshold in each sub-region includes: Detect whether at least one point of the non-ground point cloud clustering target is located within a first sub-region, and whether the relative height is less than a first preset height threshold; When at least one point of the non-ground point cloud cluster target is located within a first sub-region and the relative height is less than a first preset height threshold, the category of the non-ground point cloud cluster target located in the first sub-region is determined to be an obstacle. Detect whether at least one point of the non-ground point cloud clustering target is located within the second sub-region, and whether the relative height is less than a second preset height threshold; When at least one point of the non-ground point cloud cluster target is located within the second sub-region, and the relative height is less than a second preset height threshold, the category of the non-ground point cloud cluster target located in the second sub-region is determined to be an obstacle; Detect whether at least one point of the non-ground point cloud clustering target is located within a third sub-region, and whether the relative height is less than a third preset height threshold; When at least one point of the non-ground point cloud cluster target is located within the third sub-region, and the relative height is less than a third preset height threshold, the category of the non-ground point cloud cluster target located in the third sub-region is determined to be an obstacle. When at least one point of the non-ground point cloud cluster target is located in a first sub-region and the relative height is greater than or equal to a first preset height threshold, or when at least one point of the non-ground point cloud cluster target is located in a second sub-region and the relative height is greater than or equal to a second preset height threshold, or when at least one point of the non-ground point cloud cluster target is located in a third sub-region and the relative height is greater than or equal to a third preset height threshold, the category of the non-ground point cloud cluster target is determined to be a mountain. When at least one point of the non-ground point cloud cluster target is located within the first rectangular local area, the category of the non-ground point cloud cluster target is determined to be a mountain.

[0013] In some optional embodiments, the step of detecting the positional relationship between each point of the vehicle model point cloud corresponding to each point of the backward trajectory line and the mountain voxel grid, starting from a preset loading point, and then determining the optimal stopping point corresponding to each preset loading point includes: Starting from the preset loading point, along the direction of the backward trajectory line, detect whether each point of the vehicle model point cloud corresponding to each backward trajectory line point falls within the mountain voxel grid; When each point of the vehicle model point cloud corresponding to the retreat trajectory line point falls into the mountain voxel grid, it is determined that the loaded vehicle will come into contact with the mountain, and it is detected whether each point of the vehicle model point cloud corresponding to the next trajectory line point along the retreat trajectory line direction falls into the mountain voxel grid. When the points of the vehicle model point cloud corresponding to the retreat trajectory line do not fall within the mountain voxel grid, it is determined that the loaded vehicle will not come into contact with the mountain, and the retreat trajectory line is determined to be the optimal stopping point of the preset loading point.

[0014] In some optional embodiments, the steps of acquiring the vehicle model point cloud and constructing the mountain voxel grid include: A point cloud model of the vehicle's rear is constructed based on the vehicle's vehicle model parameters. The vehicle's rear includes the wheels, rear axle, and cargo box. A voxel grid is established based on the ground height of the hillside furthest from the preset loading point along the vehicle's reversing trajectory. The height of the voxel grid is set such that the lowest point is the difference between the ground height and a preset height, and the highest point is the sum of the ground height, the height from the ground to the top of the cargo box, and a preset error range. The length of the voxel grid is the sum of the length from the vehicle's rear axle center to the bottom of the cargo box, a first preset length, and a second preset length. The first preset length is a preset distance along the reverse direction of the reversing trajectory, with the preset loading point as the base point. The sum of the length from the vehicle's rear axle center to the bottom of the cargo box and the second preset length is a preset distance along the reversing trajectory, with the preset loading point as the base point. The width of the voxel grid is the sum of the vehicle body width and a third preset length, with the preset loading point located at the midpoint of the voxel grid's width. Attached Figure Description

[0015] One or more embodiments are illustrated by way of example with reference to the accompanying drawings, and these illustrative descriptions do not constitute a limitation on the embodiments.

[0016] Figure 1 A flowchart illustrating a method for calculating the parking position in a loading area according to an embodiment of this application; Figure 2A schematic diagram illustrating the determination of the backward and forward routes based on preset loading (unloading) points; Figure 3 A schematic diagram illustrating the division of a first rectangular region provided in one embodiment of this application; Figure 4 A schematic diagram illustrating the division of a second rectangular region provided in one embodiment of this application; Figure 5 A schematic diagram illustrating the division of a third rectangular region is provided for one embodiment of this application; Figure 6 A schematic diagram of a method for calculating the loading area docking position according to another embodiment of this application; Figure 7 A schematic diagram of a vehicle model point cloud provided for another embodiment of this application; Figure 8 A schematic diagram showing the lowest and highest points and length of a mountain voxel grid in the height direction, provided for another embodiment of this application; Figure 9 A schematic diagram of the width of a mountain voxel grid provided for another embodiment of this application; Figure 10 A schematic diagram of a vehicle model point cloud and a mountain voxel grid provided for another embodiment of this application; Figure 11 A schematic diagram of a loading area docking position calculation device provided for another embodiment of this application; Figure 12 A schematic diagram of the structure of an electronic device provided for another embodiment of this application.

[0017] In the accompanying drawings, the same parts are referred to by the same reference numerals, and the drawings are not drawn to scale. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the various embodiments of this application will be described in detail below with reference to the accompanying drawings. However, those skilled in the art will understand that many technical details have been provided in the various embodiments of this application to help readers better understand this application. However, the technical solutions claimed in this application can be implemented even without these technical details and various changes and modifications based on the following embodiments. The division of the various embodiments below is for the convenience of description and should not constitute any limitation on the specific implementation of this application. The various embodiments can be combined with and referenced by each other without contradiction.

[0019] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this disclosure 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 this disclosure 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 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.

[0020] 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, and 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.

[0021] Example 1: Figure 1 This is a flowchart illustrating a method for calculating the parking position in a loading area, as provided in an embodiment of this disclosure. Figure 1 As shown, a method for calculating the parking position in a loading area includes: Step 110: Obtain each preset loading point and the loading point heading value and backward trajectory line determined by the loading vehicle based on each preset loading point.

[0022] Specifically, such as Figure 2 As shown, after the mining truck (i.e., the loading vehicle) arrives at the loading area, the cloud generates a series of loading points based on the boundaries of the loading area. The mining truck automatically selects a valid loading point and plans the corresponding driving route. The planned driving route is divided into two segments: a forward segment (i.e., the forward trajectory line) and a reverse segment (i.e., the reverse trajectory line). When the mining truck travels along the forward route to the shift point, it switches to reverse gear to reverse and park at a suitable location for loading. Furthermore, the system obtains the preset loading point selected by the loading vehicle, as well as the reverse trajectory line determined based on that preset loading point and the loading point heading value of the loading vehicle.

[0023] Step 120: Using the heading value of the loading point of the loaded vehicle as the direction, and using each of the preset loading points as the base points, divide the area to obtain three rectangular areas. The length direction of the three rectangular areas is along the heading value of the loading point, and the width direction of the three rectangular areas is along the direction perpendicular to the heading value of the loading point. The length and width of the three rectangular areas are different.

[0024] Specifically, taking the preset loading point location as the base point and the loading point heading of the loading vehicle as the direction, three rectangular areas of different sizes are divided. The length of the rectangular area is along the loading point heading, and the width of the rectangular area is along the direction perpendicular to the loading point heading. The length and width of each rectangular area are different.

[0025] Step 130: For each of the preset loading points, obtain the non-ground point cloud cluster targets on the backward trajectory line of the loading vehicle.

[0026] Specifically, extract the non-ground point cloud cluster targets during the reversing process of the loading vehicle, that is, obtain the non-ground point cloud cluster targets on the reversing trajectory line of the loading vehicle.

[0027] Step 140: Determine the category of each non-ground point cloud cluster target based on the location of each non-ground point cloud cluster target and the comparison results between each non-ground point cloud cluster target and the preset height threshold in the three rectangular regions. The categories of the non-ground point cloud cluster targets include mountains and obstacles.

[0028] Specifically, the categories of each non-ground point cloud cluster target may be obstacles or mountain structures. It is necessary to identify the categories of each ground point cloud cluster target to avoid situations where loading vehicles misidentify the categories of non-ground point cloud cluster targets during reversal, resulting in inaccurate stopping point determination. Further, the category of each non-ground point cloud cluster target is determined by comparing its location with a preset height threshold in each rectangular region; that is, the category of non-ground point cloud cluster targets is distinguished based on their position relative to each rectangular region and by using different preset height thresholds for each rectangular region. In this embodiment, the region is segmented based on the preset loading point location. Different height thresholds are selected in different regions to label the point cloud as either a mountain or an obstacle, avoiding directly classifying other vehicles being loaded as mountains, which could lead to vehicles failing to properly stop and ultimately stopping too close together.

[0029] Step 150: For each preset loading point, obtain the vehicle model point cloud and the mountain voxel grid where the backward trajectory of the loaded vehicle is furthest from the preset loading point.

[0030] Specifically, after identifying the categories of each non-ground point cloud cluster target on the retreat trajectory line, the vehicle model point cloud and the mountain voxel grid of the retreat trajectory of the loaded vehicle are obtained, which are far away from the preset loading point. Then, the optimal stopping point on the retreat trajectory line of the subsequent loaded vehicle is determined based on the vehicle model point cloud and the mountain voxel grid.

[0031] Step 160: Starting from the preset loading point, detect the positional relationship between each point of the vehicle model point cloud corresponding to each point of the backward trajectory line and the mountain voxel grid, thereby determining the optimal stopping point corresponding to each preset loading point.

[0032] Specifically, starting from the preset loading point, the optimal stopping point is determined along the reverse trajectory line based on the positional relationship between each point in the vehicle model point cloud corresponding to each reverse trajectory line point and the mountain voxel grid. Further, based on any reverse trajectory line point, the corresponding vehicle model point cloud is traversed, and the position of the vehicle model point cloud and the mountain voxel grid is determined to ascertain whether the vehicle will come into contact with the mountain. Based on the determination results at each point, the optimal stopping point for the loaded vehicle based on the preset loading point is determined. In this way, a vehicle point cloud model is generated using vehicle model parameters, the mountain is detected during operation, and a mountain point cloud voxel grid is constructed. The collision situation between the vehicle and the mountain is analyzed in real time, and the best position point is selected as the final stopping position on the planned reverse trajectory route.

[0033] In one example, the non-ground point cloud cluster target identified as a mountain and the vehicle model point cloud are used to determine whether the loading vehicle will collide with the mountain during the reversing process. Then, the reversing trajectory point when there is no collision with the mountain is determined as the optimal stopping point for the loading vehicle.

[0034] The loading area parking position calculation method provided in this embodiment can effectively distinguish between material hills and ordinary obstacles. While ensuring improved accuracy of hill collision analysis, it reduces the probability of missing obstacles and avoids vehicles getting too close to obstacles. Furthermore, based on the planned trajectory, it can perform dynamic analysis of the material hills scanned in real time during vehicle operation and combine it with the vehicle's shape for collision analysis. This allows for more accurate parking at loading points, preventing vehicles from stopping too early and failing to load materials or from running over materials and causing tire wear due to failure to stop in time.

[0035] In some embodiments, the three rectangular regions are respectively a first rectangular region, a second rectangular region, and a third rectangular region. The step of dividing the region into three rectangular regions using the heading value of the loading point of the loaded vehicle as the direction and each preset loading point as a base point, wherein the length direction of the three rectangular regions is along the heading value of the loading point, and the width direction of the three rectangular regions is along the direction perpendicular to the heading value of the loading point, and the length and width of the three rectangular regions are different, includes: The width of the first rectangular region is greater than the width of the second rectangular region, the width of the second rectangular region is greater than the width of the third rectangular region, and the third rectangular region is composed of three sub-regions with equal widths. The three sub-regions are arranged side by side along the direction of the backward trajectory line. The three sub-regions are the first sub-region, the second sub-region, and the third sub-region. The two sides of the second sub-region along the length direction overlap with the one side of the first sub-region along the length direction and the one side of the third sub-region along the length direction, respectively. The length of the first sub-region is equal to the length of the first rectangular region, and the sum of the lengths of the second and third sub-regions is equal to the length of the second rectangular region.

[0036] In some embodiments, the step of dividing the area into three rectangular regions using the heading value of the loading point of the loaded vehicle as the direction and each preset loading point as the base point, wherein the length direction of the three rectangular regions is along the heading value of the loading point and the width direction of the three rectangular regions is along the direction perpendicular to the heading value of the loading point, and the length and width of the three rectangular regions are different, further includes: The width of the first rectangular area is the sum of the width of the vehicle body and a first preset distance; the width of the second rectangular area is the sum of the width of the vehicle body and a second preset distance; and the width of the third rectangular area is the sum of the width of the vehicle body and a third preset distance. The first preset distance is greater than the second preset distance, and the second preset distance is greater than the third preset distance. The length of the first rectangular area is the sum of a first distance and a second distance. The first distance is a preset distance along the backward trajectory line with the preset loading point as the base point, and the second distance is a preset distance in the opposite direction of the backward trajectory line with the preset loading point as the base point. The length of the second rectangular area is preset to be the third distance, which is a preset distance along the backward trajectory line with the preset loading point as the base point; The length of the first sub-region is preset to be equal to the length of the first rectangular region, the length of the second sub-region is a fourth distance, the fourth distance is a preset distance along the retreat trajectory line, the length of the third sub-region is a fifth distance, the fifth distance is a preset distance along the retreat trajectory line, and the sum of the fourth distance and the fifth distance is equal to the length of the second rectangular region.

[0037] Specifically, such as Figure 3 , Figure 4 as well as Figure 5 As shown, the specific process of dividing the area into three rectangular regions using the heading value of the loading point of the loaded vehicle as the direction and each preset loading point as the base point is as follows: The large rectangular region, i.e., the first rectangular region, is divided as follows: Figure 3 As shown, the length of the first rectangular area is along the heading direction of the loading point, and the width of the first rectangular area is along the direction perpendicular to the heading direction of the loading point. The width of the first rectangular area is the sum of the width of the vehicle body and the first preset distance. The preset loading point is defined as A. The distance AB is the preset distance along the backward trajectory line with the preset loading point as the base point, and the distance AC is the preset distance in the opposite direction of the backward trajectory line with the preset loading point as the base point. The length of the first rectangular area is the sum of the AB and AC segments.

[0038] In one example, the first preset distance is 6m, the distance between segments AB is 1m, and the distance between segments AC is 10m.

[0039] The division of the middle rectangular region, i.e., the second rectangular region, is as follows: Figure 4 As shown, the length of the second rectangular area is along the heading direction of the loading point, and the width of the second rectangular area is along the direction perpendicular to the heading direction of the loading point. The width of the second rectangular area is the sum of the width of the vehicle body and the second preset distance. The preset loading point is defined as A, and the distance AB is the preset distance along the backward trajectory line with the preset loading point as the base point. The length of the second rectangular area is the distance of the AB segment.

[0040] In one example, the second preset distance is 4m, and the distance between segments AB is 8m.

[0041] The division of the small rectangular region, i.e., the third rectangular region, is as follows: Figure 5As shown, the length of the third rectangular region is along the heading direction of the loading point, and the width of the third rectangular region is perpendicular to the heading direction of the loading point. The third rectangular region consists of three sub-regions, the width of which is the sum of the width of the loaded vehicle body and a third preset distance. The first preset distance is greater than the second preset distance, and the second preset distance is greater than the third preset distance. The three sub-regions are arranged side-by-side and are designated as the first sub-region S3, the second sub-region S2, and the third sub-region S1. The two sides of the second sub-region S2 in the length direction overlap with one side of the first sub-region S3 in the length direction and one side of the third sub-region S1 in the length direction, respectively. The following is a definition of a rectangular region: The first sub-region S3 has a length BE equal to the length of the first rectangular region. Specifically, the BE segment distance is the sum of the AB segment distance and the AE segment distance. The AB segment distance is a preset distance along the backward trajectory line with the preset loading point as the base point, and the AE segment distance is a preset distance along the backward trajectory line with the preset loading point as the base point in the opposite direction. The second sub-region S2 has a length BC that is a fourth distance, which is a preset distance along the backward trajectory line. The third sub-region S1 has a length CD that is a fifth distance, which is a preset distance along the backward trajectory line. The sum of the fourth and fifth distances is equal to the length of the second rectangular region.

[0042] In one example, the third preset distance is set to 2m, the AB segment distance is 1m, the AE segment distance is 10m, and the sum of the BC segment distance and the CD segment distance is 8m.

[0043] In some embodiments, the step of determining the category of each non-ground point cloud cluster target based on the location of each non-ground point cloud cluster target and the comparison result between each non-ground point cloud cluster target and a preset height threshold in the three rectangular regions, wherein the category of the non-ground point cloud cluster target includes mountains and obstacles, includes: Detect whether each point of the non-ground point cloud clustering target is within the second rectangular region; When each point of the non-ground point cloud cluster target is within the second rectangular area, the elevation value of the lowest point of the non-ground point cloud cluster target within the second rectangular area is determined to be the elevation value of the target's lowest point. When none of the points of the non-ground point cloud cluster target are within the second rectangular area, the elevation value of the lowest point of the non-ground point cloud cluster target is determined to be the elevation value of the lowest point of the target. Based on the sub-region where each of the non-ground point cloud cluster targets is located, the difference between the highest point elevation value and the lowest point elevation value of the target in the corresponding sub-region is calculated to obtain the relative height; The category of each non-ground point cloud cluster target is determined based on the comparison between the location of each non-ground point cloud cluster target and the relative height with the preset height threshold in each sub-region.

[0044] In some embodiments, the step of determining the category of each non-ground point cloud cluster target based on the comparison result of the location of each non-ground point cloud cluster target and the relative height with a preset height threshold in each sub-region includes: Detect whether at least one point of the non-ground point cloud clustering target is located within a first sub-region, and whether the relative height is less than a first preset height threshold; When at least one point of the non-ground point cloud cluster target is located within a first sub-region and the relative height is less than a first preset height threshold, the category of the non-ground point cloud cluster target located in the first sub-region is determined to be an obstacle. Detect whether at least one point of the non-ground point cloud clustering target is located within the second sub-region, and whether the relative height is less than a second preset height threshold; When at least one point of the non-ground point cloud cluster target is located within the second sub-region, and the relative height is less than a second preset height threshold, the category of the non-ground point cloud cluster target located in the second sub-region is determined to be an obstacle; Detect whether at least one point of the non-ground point cloud clustering target is located within a third sub-region, and whether the relative height is less than a third preset height threshold; When at least one point of the non-ground point cloud cluster target is located within the third sub-region, and the relative height is less than a third preset height threshold, the category of the non-ground point cloud cluster target located in the third sub-region is determined to be an obstacle. When at least one point of the non-ground point cloud cluster target is located in a first sub-region and the relative height is greater than or equal to a first preset height threshold, or when at least one point of the non-ground point cloud cluster target is located in a second sub-region and the relative height is greater than or equal to a second preset height threshold, or when at least one point of the non-ground point cloud cluster target is located in a third sub-region and the relative height is greater than or equal to a third preset height threshold, the category of the non-ground point cloud cluster target is determined to be a mountain. When at least one point of the non-ground point cloud cluster target is located within the first rectangular local area, the category of the non-ground point cloud cluster target is determined to be a mountain.

[0045] Specifically, the steps for determining the category of each non-ground point cloud cluster target based on the comparison between the location of each non-ground point cloud cluster target and the relative height with a preset height threshold in each sub-region are as follows: A. For each non-ground point cloud cluster target extracted during the backtracking process, find the lowest point elevation value low_z of the points within the second rectangular area. If none of the cluster block points are within this area, low_z is taken as its own lowest point elevation value. B. Traverse each point in the cluster target, convert it to a global coordinate system, and obtain the relative height h through the point elevation value z - low_z. If the point is located in region S1 and h > 1 meter, the target is directly determined as a non-mountain and classified as an ordinary obstacle; if the point is located in region S2 and h > 2 meters, the target is directly determined as a non-mountain; if the point is located in region S3 and h > 8 meters, the target is directly determined as a non-mountain. C. If at least one point is located in region S1, S2 or S3, but none of the three steps in step B identified it as a normal obstacle that is not a mountain, then it is classified as a mountain. D. If there are no cluster targets within the S1, S2, or S3 regions, traverse each point. If a point is located within the large rectangular region (i.e., the first rectangular region), then it is determined to be a mountain.

[0046] In some embodiments, the steps of acquiring the vehicle model point cloud and constructing the mountain voxel grid include: A point cloud model of the vehicle's rear is constructed based on the vehicle's vehicle model parameters. The vehicle's rear includes the wheels, rear axle, and cargo box. A voxel grid is established based on the ground height of the hillside furthest from the preset loading point along the vehicle's reversing trajectory. The height of the voxel grid is set such that the lowest point is the difference between the ground height and a preset height, and the highest point is the sum of the ground height, the height from the ground to the top of the cargo box, and a preset error range. The length of the voxel grid is the sum of the length from the vehicle's rear axle center to the bottom of the cargo box, a first preset length, and a second preset length. The first preset length is a preset distance along the reverse direction of the reversing trajectory, with the preset loading point as the base point. The sum of the length from the vehicle's rear axle center to the bottom of the cargo box and the second preset length is a preset distance along the reversing trajectory, with the preset loading point as the base point. The width of the voxel grid is the sum of the vehicle body width and a third preset length, with the preset loading point located at the midpoint of the voxel grid's width.

[0047] Specifically, the steps for obtaining the vehicle model point cloud and constructing the mountain voxel raster include the following: A. Construct a point cloud model of the rear of the vehicle based on the vehicle model parameters, including the wheels, rear axle, and cargo box; B. During the backward process, the first point cloud of the mountain is extracted, and the ground height at this time is recorded as record_bottom_height; C. Construct a voxel grid. The lowest point in the z-axis height is the difference between `record_bottom_height` and 2m, and the highest point is the sum of `record_bottom_height`, the height from the ground to the top of the cargo plane, and 0.3m. The length is the sum of 5m, `car_back_length` (the length from the center of the rear axle to the bottom of the cargo plane), and 2m. The width is the sum of half the vehicle width and 3m on both sides, with the preset loading point as the midpoint. The grid dimensions are 0.3m. Construct the grid using these parameters.

[0048] In some embodiments, the step of detecting the positional relationship between each point of the vehicle model point cloud corresponding to each point of the backward trajectory line and the mountain voxel grid, starting from a preset loading point, and then determining the optimal stopping point corresponding to each preset loading point includes: Starting from the preset loading point, along the direction of the backward trajectory line, detect whether each point of the vehicle model point cloud corresponding to each backward trajectory line point falls within the mountain voxel grid; When each point of the vehicle model point cloud corresponding to the retreat trajectory line point falls into the mountain voxel grid, it is determined that the loaded vehicle will come into contact with the mountain, and it is detected whether each point of the vehicle model point cloud corresponding to the next trajectory line point along the retreat trajectory line direction falls into the mountain voxel grid. When the points of the vehicle model point cloud corresponding to the retreat trajectory line do not fall within the mountain voxel grid, it is determined that the loaded vehicle will not come into contact with the mountain, and the retreat trajectory line is determined to be the optimal stopping point of the preset loading point.

[0049] Specifically, starting from the preset loading point, the steps of detecting the positional relationship between each point in the vehicle model point cloud corresponding to each point on the backward trajectory line and the mountain voxel grid, and then determining the optimal stopping point corresponding to each preset loading point, include the following: A. During the loading and rollback process, for each detected mountain, the mountain points are traversed, the position of the voxel grid is calculated, and each grid is set to occupied. B. Starting from the last point of the backward trajectory line (i.e. the preset loading point), obtain the coordinates and heading angle information of the position point, and at the same time obtain the roll pitch information of the current parking space. Based on this information, a rotation and translation matrix can be constructed to transfer the vehicle point cloud to the corresponding global coordinate system. C. Traverse the point cloud of the vehicle model, determine whether the voxel containing the point occupies a grid cell, and if it is occupied. If it is occupied, it means that the vehicle will come into contact with the mountain. Then continue to the previous trajectory point to check until a point position index that will not collide with the vehicle is found. D. Obtain the final parking location coordinates and send them to the decision-making and planning module for parking.

[0050] The loading area parking position calculation method provided in this embodiment can effectively distinguish between material hills and ordinary obstacles. While ensuring improved accuracy of hill collision analysis, it reduces the probability of missing obstacles and avoids vehicles getting too close to obstacles. Furthermore, based on the planned trajectory, it can perform dynamic analysis of the material hills scanned in real time during vehicle operation and combine it with the vehicle's shape for collision analysis. This allows for more accurate parking at loading points, preventing vehicles from stopping too early and failing to load materials or from running over materials and causing tire wear due to failure to stop in time.

[0051] Example 2: Another embodiment of this application relates to a method for calculating the accurate loading and stopping position of an unmanned mining truck in a loading area, such as... Figure 6 As shown, the method includes: Once the mining truck arrives at the loading area, the cloud will generate a series of loading points based on the boundaries of the loading area. The mining truck will then automatically select a valid loading point and plan the corresponding driving route. The planned driving route is divided into two segments: a forward segment and a reverse segment. When the mining truck reaches the shift point along the forward route, it will switch to reverse gear to park at a suitable location for loading.

[0052] 1. Area division based on preset loading points Using the preset loading point as the base point, and its heading value as the direction, divide the area into large, medium, and small rectangular regions. The large rectangular region is shown below. Figure 3 As shown, AC is 10m, AB is 1m, and the width is the vehicle width + 6m.

[0053] medium rectangular area Figure 4 As shown, AB is 8m and the width is the vehicle width + 4m.

[0054] The small rectangular area is divided into three regions: S1, S2, and S3, as follows: Figure 5 As shown, the width is vehicle width + 2m. Here, A is the preset loading point, S1 is a rectangle with length CD, S2 is a rectangle with length BC, and S3 is a rectangle with length BE.

[0055] 2. Different height thresholds are used to distinguish mountains from obstacles in different areas. A. For each non-ground point cloud cluster target extracted during the backtracking process, find the lowest point elevation value low_z of the points within the rectangular area. If none of the cluster block points are within this area, low_z is taken as its own lowest point elevation value. B. Traverse each point in the cluster target, convert it to a global coordinate system, and obtain the relative height h through the point elevation value z - low_z. If the point is located in region S1 and h > 1 meter, the target is directly determined as a non-mountain and classified as an ordinary obstacle; if the point is located in region S2 and h > 2 meters, the target is directly determined as a non-mountain; if the point is located in region S3 and h > 8 meters, the target is directly determined as a non-mountain. C. If at least one point is located in region S1, S2 or S3, but it was not identified as a normal obstacle other than a mountain in any of the three steps in step B, then it is classified as a mountain. D. If there are no cluster targets within the S1, S2, or S3 regions, traverse each point. If a point is located within a large rectangular region, it is determined to be a mountain.

[0056] 3. Generate vehicle point clouds and construct a mountain voxel occupancy raster. A. Construct a point cloud model of the rear of the vehicle based on vehicle model parameters, including wheels, rear axle, and cargo box, etc. Figure 7 As shown; B. During the backward process, the first point cloud of the mountain is extracted, and the ground height at this time is recorded as record_bottom_height; C. Construct a voxel grid. The lowest point in the z-axis is `record_bottom_height - 2m`, and the highest point is `record_bottom_height + height from ground to top of cargo plane + 0.3m`. The length is 5m + `car_back_length` (length from rear axle center to bottom of cargo plane) + 2m. The width is half the width of the vehicle on each side + 3m. For example... Figure 8 and Figure 9 As shown. The grid dimensions are 0.3m (length, width, height). It is constructed using these parameters.

[0057] 4. Analyze vehicle-mountain collisions and select the optimal stopping point. A. During the loading and rollback process, according to step 2, for each mountain detected, the mountain points are traversed, the position of the voxel grid is calculated, and each grid is set as occupied; B. Starting from the last point of the backward trajectory line (i.e. the preset loading point), obtain the coordinates and heading angle information of the position point, and at the same time obtain the roll pitch information of the current parking space. Based on this information, a rotation and translation matrix can be constructed to transfer the vehicle point cloud to the corresponding global coordinate system. C. Traverse the vehicle model point cloud, determine if the voxel containing the point occupies a grid cell, and if so, if it is occupied. If occupied, it means the vehicle will come into contact with the mountain. Continue checking the previous trajectory point until a point index that does not collide with the vehicle is found. Figure 10 As shown; D. Obtain the final parking location coordinates and send them to the decision-making and planning module for parking.

[0058] The loading area parking position calculation method provided in this embodiment generates a vehicle point cloud model through vehicle model parameters, detects mountains during operation, constructs a mountain point cloud voxel grid, analyzes vehicle-mountain collisions in real time, and selects the optimal location as the final parking position on the planned reverse trajectory. Based on the preset loading point location, the area is segmented, and different height thresholds are selected in different segments to mark the point cloud as either a mountain or an obstacle. This avoids directly classifying other vehicles being loaded as mountains, leading to improper parking and vehicles parking too close together. This method can dynamically analyze the material mountains scanned in real time during vehicle operation, combined with collision analysis based on the vehicle's morphology, enabling more accurate parking at loading points. It avoids vehicles stopping too early, preventing loading failure, or running over materials and causing tire wear due to failure to stop in time. Furthermore, it effectively distinguishes between material mountains and ordinary obstacles, ensuring improved accuracy in mountain collision analysis while reducing the probability of missed obstacles and preventing vehicles from getting too close to obstacles.

[0059] Example 3: Another embodiment of this application relates to a loading area parking position calculation device. The implementation details of this loading area parking position calculation device are described below. The following details are provided for ease of understanding and are not essential for implementing this solution. A schematic diagram of the loading area parking position calculation device in this embodiment can be seen as follows: Figure 11 As shown, it includes a first acquisition module 1101, a region division module 1102, a clustering target determination module 1103, a target category determination module 1104, a second acquisition module 1105, and a stop point determination module 1106.

[0060] The first acquisition module 1101 is used to acquire each preset loading point and the loading point heading value and backward trajectory line determined by the loading vehicle based on each preset loading point; The region division module 1102 is used to divide the region into three rectangular regions with the heading value of the loading point of the loaded vehicle as the direction and each preset loading point as the base point. The length direction of the three rectangular regions is along the heading value of the loading point, and the width direction of the three rectangular regions is along the direction perpendicular to the heading value of the loading point. The length and width of the three rectangular regions are different. Clustering target determination module 1103 is used to obtain the clustering targets of each non-ground point cloud on the backward trajectory line of the loading vehicle for each preset loading point. The target category determination module 1104 is used to determine the category of each non-ground point cloud cluster target based on the location of each non-ground point cloud cluster target and the comparison result between each non-ground point cloud cluster target and a preset height threshold in the three rectangular regions. The categories of the non-ground point cloud cluster targets include mountains and obstacles. The second acquisition module 1105 is used to acquire, for each of the preset loading points, the point cloud of the vehicle model and the voxel grid of the mountain where the backward trajectory line of the loaded vehicle is furthest from the preset loading point. The stopping point determination module 1106 is used to detect the positional relationship between each point of the vehicle model point cloud corresponding to each point of the backward trajectory line and the mountain voxel grid, starting from the preset loading point, and then determine the optimal stopping point corresponding to each preset loading point.

[0061] In some optional embodiments, the region division module is further configured to preset that the width of the first rectangular region is greater than the width of the second rectangular region, the width of the second rectangular region is greater than the width of the third rectangular region, the third rectangular region is composed of three sub-regions, the widths of the three sub-regions are equal, the three sub-regions are arranged side by side along the direction of the backward trajectory line, the three sub-regions are respectively the first sub-region, the second sub-region and the third sub-region, the two sides of the second sub-region in the length direction overlap with the one side of the first sub-region in the length direction and the one side of the third sub-region in the length direction, respectively; the length of the first sub-region is equal to the length of the first rectangular region, and the sum of the lengths of the second sub-region and the third sub-region is equal to the length of the second rectangular region.

[0062] In some optional embodiments, the region division module is further configured to preset the width of the first rectangular region to be the sum of the width of the vehicle body and a first preset distance, the width of the second rectangular region to be the sum of the width of the vehicle body and a second preset distance, and the width of the third rectangular region to be the sum of the width of the vehicle body and a third preset distance, wherein the first preset distance is greater than the second preset distance, and the second preset distance is greater than the third preset distance; The length of the first rectangular area is the sum of a first distance and a second distance. The first distance is a preset distance along the backward trajectory line with the preset loading point as the base point, and the second distance is a preset distance in the opposite direction of the backward trajectory line with the preset loading point as the base point. The length of the second rectangular area is preset to be the third distance, which is a preset distance along the backward trajectory line with the preset loading point as the base point; The length of the first sub-region is preset to be equal to the length of the first rectangular region, the length of the second sub-region is a fourth distance, the fourth distance is a preset distance along the retreat trajectory line, the length of the third sub-region is a fifth distance, the fifth distance is a preset distance along the retreat trajectory line, and the sum of the fourth distance and the fifth distance is equal to the length of the second rectangular region.

[0063] In some optional embodiments, the target category determination module is further configured to detect whether each point of the non-ground point cloud clustering target is within the second rectangular region; When each point of the non-ground point cloud cluster target is within the second rectangular area, the elevation value of the lowest point of the non-ground point cloud cluster target within the second rectangular area is determined to be the elevation value of the target's lowest point. When none of the points of the non-ground point cloud cluster target are within the second rectangular area, the elevation value of the lowest point of the non-ground point cloud cluster target is determined to be the elevation value of the lowest point of the target. Based on the sub-region where each of the non-ground point cloud cluster targets is located, the difference between the highest point elevation value and the lowest point elevation value of the target in the corresponding sub-region is calculated to obtain the relative height; The category of each non-ground point cloud cluster target is determined based on the comparison between the location of each non-ground point cloud cluster target and the relative height with the preset height threshold in each sub-region.

[0064] In some optional embodiments, the target category determination module is further configured to detect whether at least one point of the non-ground point cloud clustering target is located within a first sub-region and whether the relative height is less than a first preset height threshold. When at least one point of the non-ground point cloud cluster target is located within a first sub-region and the relative height is less than a first preset height threshold, the category of the non-ground point cloud cluster target located in the first sub-region is determined to be an obstacle. Detect whether at least one point of the non-ground point cloud clustering target is located within the second sub-region, and whether the relative height is less than a second preset height threshold; When at least one point of the non-ground point cloud cluster target is located within the second sub-region, and the relative height is less than a second preset height threshold, the category of the non-ground point cloud cluster target located in the second sub-region is determined to be an obstacle; Detect whether at least one point of the non-ground point cloud clustering target is located within a third sub-region, and whether the relative height is less than a third preset height threshold; When at least one point of the non-ground point cloud cluster target is located within the third sub-region, and the relative height is less than a third preset height threshold, the category of the non-ground point cloud cluster target located in the third sub-region is determined to be an obstacle. When at least one point of the non-ground point cloud cluster target is located in a first sub-region and the relative height is greater than or equal to a first preset height threshold, or when at least one point of the non-ground point cloud cluster target is located in a second sub-region and the relative height is greater than or equal to a second preset height threshold, or when at least one point of the non-ground point cloud cluster target is located in a third sub-region and the relative height is greater than or equal to a third preset height threshold, the category of the non-ground point cloud cluster target is determined to be a mountain. When at least one point of the non-ground point cloud cluster target is located within the first rectangular local area, the category of the non-ground point cloud cluster target is determined to be a mountain.

[0065] In some optional embodiments, the docking point determination module is further configured to detect, starting from the preset loading point, whether each point of the vehicle model point cloud corresponding to each retreating trajectory point falls within the mountain voxel grid along the retreating trajectory line direction; When each point of the vehicle model point cloud corresponding to the retreat trajectory line point falls into the mountain voxel grid, it is determined that the loaded vehicle will come into contact with the mountain, and it is detected whether each point of the vehicle model point cloud corresponding to the next trajectory line point along the retreat trajectory line direction falls into the mountain voxel grid. When the points of the vehicle model point cloud corresponding to the retreat trajectory line do not fall within the mountain voxel grid, it is determined that the loaded vehicle will not come into contact with the mountain, and the retreat trajectory line is determined to be the optimal stopping point of the preset loading point.

[0066] In some optional embodiments, the second acquisition module is further configured to construct a model point cloud of the vehicle rear based on the vehicle model parameters, wherein the vehicle rear includes wheels, a rear axle, and a cargo box. A voxel grid is established based on the ground height of the hillside furthest from the preset loading point along the vehicle's reversing trajectory. The height of the voxel grid is set such that the lowest point is the difference between the ground height and a preset height, and the highest point is the sum of the ground height, the height from the ground to the top of the cargo box, and a preset error range. The length of the voxel grid is the sum of the length from the vehicle's rear axle center to the bottom of the cargo box, a first preset length, and a second preset length. The first preset length is a preset distance along the reverse direction of the reversing trajectory, with the preset loading point as the base point. The sum of the length from the vehicle's rear axle center to the bottom of the cargo box and the second preset length is a preset distance along the reversing trajectory, with the preset loading point as the base point. The width of the voxel grid is the sum of the vehicle body width and a third preset length, with the preset loading point located at the midpoint of the voxel grid's width.

[0067] It is worth mentioning that all modules involved in this embodiment are logical modules. In practical applications, a logical unit can be a physical unit, a part of a physical unit, or a combination of multiple physical units. Furthermore, to highlight the innovative aspects of this application, this embodiment does not introduce units that are not closely related to solving the technical problems proposed in this application; however, this does not mean that other units are absent in this embodiment.

[0068] It should be noted that, in this disclosure, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element limited by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0069] Example 4: Another embodiment of this application relates to an electronic device, such as... Figure 12 As shown, it includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform the steps of the above-described loading area docking position calculation method.

[0070] The memory and processor are connected via a bus, which can include any number of interconnecting buses and bridges, connecting various circuits of one or more processors and memories. The bus can also connect various other circuits, such as peripheral devices, voltage regulators, and power management circuits, which are well known in the art and will not be described further herein. The bus interface provides an interface between the bus and the transceiver. The transceiver can be a single element or multiple elements, such as multiple receivers and transmitters, providing a unit for communicating with various other devices over a transmission medium. Data processed by the processor is transmitted over the wireless medium via an antenna, which further receives data and transmits it to the processor.

[0071] The processor manages the bus and general processing, and also provides various functions, including timing, peripheral interfaces, voltage regulation, power management, and other control functions. Memory is used to store data used by the processor during operation.

[0072] The processor may include, but is not limited to, one or more processors or microprocessors. Each processor may be implemented as an Application Specific Integrated Circuit (ASIC), Digital Signal Processor (DSP), Digital Signal Processing Device (DSPD), Programmable Logic Device (PLD), Field Programmable Gate Array (FPGA), controller, microcontroller, microprocessor, or other electronic component, for executing the methods in the above embodiments.

[0073] Example 5: Another embodiment of this application relates to a computer-readable storage medium storing a computer program. When executed by a processor, the computer program implements the steps of the above-described loading area docking position calculation method.

[0074] That is, those skilled in the art will understand that all or part of the steps in the methods of the above embodiments can be implemented by a program instructing related hardware. This program is stored in a storage medium and includes several instructions to cause a device (which may be a microcontroller, chip, etc.) or processor to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0075] Computer-readable storage media may also store at least one computer-executable program / instruction, such as computer-readable instructions. Computer-readable storage media include, but are not limited to, volatile memory and / or non-volatile memory. Volatile memory may include, for example, random access memory (RAM) and / or cache memory. Computer-readable storage media may include, for example, read-only memory (ROM), hard disk, flash memory, etc. For example, a non-transitory computer-readable storage medium may be connected to a computing device such as a computer, and then, when the computing device executes the computer-readable instructions stored on the computer-readable storage medium, the various methods described above can be performed.

[0076] In addition, the computer device may include (but is not limited to) a data bus, an input / output (I / O) bus, a display, and input / output devices (e.g., keyboard, mouse, speakers, etc.).

[0077] The processor can communicate with external devices via the I / O bus through wired or wireless networks.

[0078] Those skilled in the art will understand that the above embodiments are specific embodiments for implementing this application, and in practical applications, various changes can be made to them in form and detail without departing from the spirit and scope of this application.

Claims

1. A method of calculating a loading zone stopping position, characterized by, The method comprises the following steps: acquiring each preset loading point and a loading point heading value and a retreat trajectory line determined by a loading vehicle based on each preset loading point; dividing the area with the loading point heading value of the loading vehicle as the direction and each preset loading point as the base point to obtain three rectangular areas, the length direction of the three rectangular areas is along the loading point heading value direction, the width direction of the three rectangular areas is perpendicular to the loading point heading value direction, and the length and width of the three rectangular areas are not equal; for each preset loading point, acquiring each non-ground point cloud clustering target on the retreat trajectory line of the loading vehicle; determining the category of each non-ground point cloud clustering target based on the position of each non-ground point cloud clustering target and the comparison result of each non-ground point cloud clustering target and the preset height threshold in the three rectangular areas, the category of the non-ground point cloud clustering target includes mountains and obstacles; for each preset loading point, acquiring a vehicle model point cloud and a mountain voxel grid farthest from the preset loading point on the retreat trajectory line of the loading vehicle; starting from the preset loading point, detecting the position relationship between each point of the vehicle model point cloud corresponding to each retreat trajectory line point on the retreat trajectory line and the mountain voxel grid, and then determining the optimal parking point corresponding to each preset loading point.

2. The method of claim 1, wherein, The three rectangular areas are a first rectangular area, a second rectangular area, and a third rectangular area, the area division with the loading point heading value of the loading vehicle as the direction and each preset loading point as the base point to obtain three rectangular areas, the length direction of the three rectangular areas is along the loading point heading value direction, the width direction of the three rectangular areas is perpendicular to the loading point heading value direction, and the length and width of the three rectangular areas are not equal, which comprises the following steps: The width of the first rectangular area is greater than the width of the second rectangular area, the width of the second rectangular area is greater than the width of the third rectangular area, the third rectangular area is composed of three sub-areas, the widths of the three sub-areas are equal, the three sub-areas are arranged side by side along the direction of the retreat trajectory line, the three sub-areas are a first sub-area, a second sub-area, and a third sub-area, the two side edges of the length direction of the second sub-area are respectively overlapped with one side edge of the first sub-area along the length direction and one side edge of the third sub-area along the length direction; the length of the first sub-area is equal to the length of the first rectangular area, and the sum of the length of the second sub-area and the length of the third sub-area is equal to the length of the second rectangular area.

3. The method of claim 2, wherein, The area division with the loading point heading value of the loading vehicle as the direction and each preset loading point as the base point to obtain three rectangular areas, the length direction of the three rectangular areas is along the loading point heading value direction, the width direction of the three rectangular areas is perpendicular to the loading point heading value direction, and the length and width of the three rectangular areas are not equal, which further comprises the following steps: The width of the first rectangular region is preset as the sum of the vehicle body width of the loading vehicle and a first preset distance, the width of the second rectangular region is preset as the sum of the vehicle body width of the loading vehicle and a second preset distance, and the width of the third rectangular region is preset as the sum of the vehicle body width of the loading vehicle and a third preset distance, wherein the first preset distance is greater than the second preset distance, and the second preset distance is greater than the third preset distance. The length of the first rectangular region is preset as the sum of a first distance and a second distance, the first distance is a preset distance in the direction of the retreat trajectory line with the preset loading point as a base point, and the second distance is a preset distance in the opposite direction of the retreat trajectory line with the preset loading point as a base point. The length of the second rectangular region is preset as a third distance, which is a preset distance in the direction of the retreat trajectory line with the preset loading point as a base point. The length of the first sub-region is equal to the length of the first rectangular region, the length of the second sub-region is a fourth distance, which is a preset distance in the direction of the retreat trajectory line, and the length of the third sub-region is a fifth distance, which is a preset distance in the direction of the retreat trajectory line, and the sum of the fourth distance and the fifth distance is equal to the length of the second rectangular region.

4. The method of claim 2, wherein, The step of determining the category of each non-ground point cloud clustering target based on the position of each non-ground point cloud clustering target and the comparison result of the relative height and the preset height threshold in each sub-region includes: detecting whether each point of the non-ground point cloud clustering target is in the second rectangular region; when each point of the non-ground point cloud clustering target is in the second rectangular region, determining the lowest point elevation value of the non-ground point cloud clustering target in the second rectangular region as a target lowest point elevation value; when none of the points of the non-ground point cloud clustering target is in the second rectangular region, determining the lowest point elevation value of the non-ground point cloud clustering target as the target lowest point elevation value; calculating the difference between the highest point elevation value of the non-ground point cloud clustering target in the corresponding sub-region and the target lowest point elevation value based on the sub-region where each non-ground point cloud clustering target is located, to obtain a relative height; determining the category of each non-ground point cloud clustering target based on the position of each non-ground point cloud clustering target and the comparison result of the relative height and the preset height threshold in each sub-region.

5. The method of claim 4, wherein, The step of determining the category of each non-ground point cloud clustering target based on the position of each non-ground point cloud clustering target and the comparison result of the relative height and the preset height threshold in each sub-region includes: detecting whether at least one point of the non-ground point cloud clustering target is in the first sub-region and whether the relative height is less than a first preset height threshold; determining that the non-ground point cloud clustering target is a mountain when at least one point of the non-ground point cloud clustering target is in the first sub-region and the relative height is greater than or equal to the first preset height threshold, or at least one point of the non-ground point cloud clustering target is in the second sub-region and the relative height is greater than or equal to the second preset height threshold, or at least one point of the non-ground point cloud clustering target is in the third sub-region and the relative height is greater than or equal to the third preset height threshold; determining that the non-ground point cloud clustering target is a mountain when at least one point of the non-ground point cloud clustering target is in the first sub-region and the relative height is greater than or equal to the first preset height threshold, or at least one point of the non-ground point cloud clustering target is in the second sub-region and the relative height is greater than or equal to the second preset height threshold, or at least one point of the non-ground point cloud clustering target is in the third sub-region and the relative height is greater than or equal to the third preset height threshold; determining that the non-ground point cloud clustering target is a mountain when at least one point of the non-ground point cloud clustering target is in the first sub-region and the relative height is greater than or equal to the first preset height threshold, or at least one point of the non-ground point cloud clustering target is in the second sub-region and the relative height is greater than or equal to the second preset height threshold, or at least one point of the non-ground point cloud clustering target is in the third sub-region and the relative height is greater than or equal to the third preset height threshold; determining that the non-ground point cloud clustering target is a mountain when at least one point of the non-ground point cloud clustering target is in the first sub-region and the relative height is greater than or equal to the first preset height threshold, or at least one point of the non-ground point cloud clustering target is in the second sub-region and the relative height is greater than or equal to the second preset height threshold, or at least one point of the non-ground point cloud clustering target is in the third sub-region and the relative height is greater than or equal to the third preset height threshold; determining that the non-ground point cloud clustering target is a mountain when at least one point of the non-ground point cloud clustering target is in the first sub-region and the relative height is greater than or equal to the first preset height threshold, or at least one point of the non-ground point cloud clustering target is in the second sub-region and the relative height is greater than or equal to the second preset height threshold, or at least one point of the non-ground point cloud clustering target is in the third sub-region and the relative height is greater than or equal to the third preset height threshold; determining that the non-ground point cloud clustering target is a mountain when at least one point of the non-ground point cloud clustering target is in the first sub-region and the relative height is greater than or equal to the first preset height threshold, or at least one point of the non-ground point cloud clustering target is in the second sub-region and the relative height is greater than or equal to the second preset height threshold, or at least one point of the non-ground point cloud clustering target is in the third sub-region and the relative height is greater than or equal to the third preset height threshold; the step of detecting the position relationship between each point of the vehicle model point cloud corresponding to each retreat trajectory line point and the mountain voxel grid along the retreat trajectory line from the preset loading point, and further determining the optimal parking point corresponding to each preset loading point comprises:

6. The method of claim 1, wherein, detecting whether each point of the vehicle model point cloud corresponding to each retreat trajectory line point falls in the mountain voxel grid along the direction of the retreat trajectory line from the preset loading point; when each point of the vehicle model point cloud corresponding to the retreat trajectory line point falls in the mountain voxel grid, determining that the loaded vehicle will contact the mountain, and detecting whether each point of the vehicle model point cloud corresponding to the next trajectory line point in the direction of the retreat trajectory line falls in the mountain voxel grid; when each point of the vehicle model point cloud corresponding to the retreat trajectory line point does not fall in the mountain voxel grid, determining that the loaded vehicle will not contact the mountain, and determining the retreat trajectory line point as the optimal parking point of the preset loading point. the steps of obtaining the vehicle model point cloud and constructing the mountain voxel grid comprise:

7. The method of claim 1, wherein, constructing a model point cloud of the vehicle tail according to the vehicle type parameters of the loaded vehicle, the vehicle tail including wheels, a rear axle, and a cargo box; ​ According to the ground height of the mountain farthest from the preset loading point of the loading vehicle, a voxel grid is established, wherein the height of the voxel grid is set as the difference between the ground height and a preset height, the lowest point in the height direction is the difference between the ground height and the preset height, and the highest point is the sum of the ground height, the height from the ground to the top of the container, and a preset error range; the length of the voxel grid is set as the sum of the length from the center of the rear axle of the vehicle to the bottom of the container, a first preset length, and a second preset length, the first preset length is a preset distance in the reverse direction of the retreat trajectory line based on the preset loading point, and the sum of the length from the center of the rear axle of the vehicle to the bottom of the container and the second preset length is a preset distance in the direction of the retreat trajectory line based on the preset loading point; the width of the voxel grid is set as the sum of the width of the vehicle body and a third preset length, and the preset loading point is located at the midpoint of the width distance of the voxel grid.

8. A loading area stop position calculating device characterized by comprising: Comprise: The first acquisition module is used for acquiring each preset loading point and the loading point heading value and the retreat trajectory line determined by the loading vehicle based on each preset loading point; The region division module is used for respectively performing region division with the loading point heading value of the loading vehicle as the direction and each preset loading point as the base point to obtain three rectangular regions, the length direction of the three rectangular regions is along the loading point heading value direction, the width direction of the three rectangular regions is perpendicular to the loading point heading value direction, and the three rectangular regions are not equal in length and width; The clustering target determination module is used for acquiring, for each preset loading point, each non-ground point cloud clustering target on the retreat trajectory line of the loading vehicle; The target category determination module is used for determining the category of each non-ground point cloud clustering target based on the position where each non-ground point cloud clustering target is located and the comparison result of each non-ground point cloud clustering target and the preset height threshold in the three rectangular regions, and the category of the non-ground point cloud clustering target comprises a mountain and an obstacle; The second acquisition module is used for acquiring, for each preset loading point, the vehicle model point cloud and the mountain voxel grid farthest from the preset loading point of the retreat trajectory line of the loading vehicle; The parking point determination module is used for detecting the position relationship between each point of the vehicle model point cloud corresponding to each retreat trajectory line point and the mountain voxel grid from the preset loading point along the retreat trajectory line, and further determining the optimal parking point corresponding to each preset loading point.

9. An electronic device, comprising: Comprise: At least one processor; And The memory is in communication connection with the at least one processor; wherein The memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the loading area parking position calculation method in any one of claims 1 to 7.

10. A computer readable storage medium storing a computer program, characterized in that, The computer program is executed by the processor to implement the loading area parking position calculation method in any one of claims 1 to 7.

Images