An unmanned mine truck path planning method for a loading and unloading area of an open pit mine

Abstract

The present application relates to a kind of unmanned mine truck path planning method for open-pit mine loading and unloading area, belong to mine intelligent driving technical field, solve the problem that existing technology cannot effectively select the best parking point for open-pit mine loading area, and carry out segmented path planning.The method comprises: obtaining the first starting point pose of the mine truck driving into the loading and unloading area, real-time sensing obstacle position, uploading the first starting point pose of the mine truck and the obstacle position to cloud control platform combined with reference map, to build the first local map;The cloud control platform delimits the parking area range according to the attribute of the loading and unloading area, determines a plurality of parking endpoints;The cloud control platform generates the optimal collision-free path according to the plurality of parking endpoints, the first starting point pose, the first local map and the kinematic parameters of the mine truck. Realize the selection of the best parking endpoint of open-pit mine loading area, and carry out segmented path planning.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of mine intelligent driving, and in particular to an unmanned mine truck path planning method for a loading and unloading area of an open-pit mine. BACKGROUND

[0002] With the gradual maturity of unmanned driving technology, the demand for improving engineering efficiency and reducing labor costs is increasingly urgent. The mine area has become one of the most suitable scenarios for landing of unmanned driving technology due to its relatively closed and unmanned road, relatively fixed transportation route, no need to operate the material during loading and unloading, and low vehicle driving speed. In the unmanned transportation system of the open-pit mine, the local path generation of the unmanned mine truck in the material loading and unloading area is the most difficult to overcome because of the multiple steering operations, large angle changes, and multiple parking points. At the same time, the sand and stone road conditions of the mine area are different from ordinary roads, and the volume and tonnage of the unmanned mine truck are much larger than ordinary commercial vehicles, which have a great impact on the tire wear of the unmanned mine truck. Therefore, for improving the efficiency of the mine unmanned transportation system and reducing its use cost, the path planning method for the loading and unloading area of the open-pit mine is crucial, which needs to have universality, efficiency, and durability.

[0003] Most of the path planning methods for the loading and unloading area of the open-pit mine in the industry use the traditional way of tracing and reproducing the driving trajectory of the skilled mine driver to solve the problem of path planning of the unmanned mine truck in the loading and unloading area of the open-pit mine. This scheme does not have universality, and because the path is fixed, once an emergency stop occurs due to sand or other obstacles, the system will be paralyzed and need external force to clean up the obstacles, which does not have efficiency.

[0004] Currently, most of the path planning algorithms for vehicles are based on the hybrid A-star algorithm, but most of them are applied to the unmanned driving scenarios of commercial unmanned vehicles and household passenger vehicles, and there are few algorithms for mine large vehicles, and the mine large vehicle usage scenarios are not deeply adapted, and the situations such as too many steering operations and steering wheel reverse dead in place when parking in the parking scenario of the unmanned mine truck on the road with great friction are not considered, which will greatly damage the durability of the tires and power steering system of the unmanned mine truck, thus many problems and safety hazards will easily occur in actual application.

[0005] Therefore, there is a lack of an unmanned mine truck path planning method for the loading and unloading area of the open-pit mine in the prior art, which can further improve the working efficiency and durability of the mine truck by evaluating and selecting multiple parking points in the loading and unloading area of the open-pit mine and reasonably allocating the best parking point for the mine truck from different starting points. SUMMARY

[0006] In view of the above analysis, the embodiments of the present application aim to provide an unmanned mine truck path planning method for a loading and unloading area of an open-pit mine, so as to solve the problem that the existing method cannot effectively select the optimal parking point for the parking area range of the loading area of the open-pit mine and perform segmented path planning according to the working characteristics of the mine truck in the open-pit mine.

[0007] In one aspect, the embodiments of the present application provide an unmanned mine truck path planning method for a loading and unloading area of an open-pit mine, comprising:

[0008] obtaining a first starting point pose of a mine truck entering the loading and unloading area, sensing a position of an obstacle in real time, uploading the first starting point pose of the mine truck and the position of the obstacle to a cloud control platform in combination with a reference map to construct a first local map;

[0009] The cloud control platform delimits a parking area range according to the attributes of the loading and unloading area and determines a plurality of parking end points;

[0010] The cloud control platform generates an optimal collision-free path according to the plurality of parking end points, the first starting point pose, the first local map and the kinematic parameters of the mine truck.

[0011] Further, the generation of the optimal collision-free path comprises:

[0012] determining a backward-reserved point pose and a path turning radius for each of the parking end points according to the kinematic parameters of the mine truck; the backward-reserved point is a position set at a distance from the parking end point in a forward straight line;

[0013] determining an intersection point of a first initial path and a second initial path corresponding to each of the backward-reserved point poses according to the backward-reserved point pose, the first starting point pose and the path turning radius in combination with a first heuristic function and a second heuristic function;

[0014] In each of the intersection points, the intersection point with the minimum sum of the total cost of the first heuristic function and the second heuristic function is selected as a turning point, and the corresponding parking end point is an optimal parking end point;

[0015] the path points from the first starting point pose to the turning point form a first optimal path, and the mine truck travels forward along the first optimal path;

[0016] the path points from the turning point to the backward-reserved point form a second optimal path, and the mine truck travels backward along the second optimal path;

[0017] the path points from the backward-reserved point to the optimal parking end point form a third optimal path, and the mine truck travels backward along the third optimal path.

[0018] Further, the first heuristic function is calculated starting from the first start point pose; and the second heuristic function is calculated starting from the reverse point pose.

[0019] Further, the first and second heuristic functions are expressed as:

[0020]

[0021]

[0022]

[0023]

[0024]

[0025] wherein m=1 is a first heuristic function corresponding parameter, and m=2 is a second heuristic function corresponding parameter; f mi is a total cost of the mth initial path of the ith stop end, is a cost of the jth point of the mth initial path of the ith stop end, and J is a node number of the mth initial path of the ith stop end, is a search cost estimation function of the jth point of the mth initial path of the ith stop end, is a pre-generated path sub-node cost function of the jth point of the mth initial path of the ith stop end, is an estimation path cost function of the jth point of the mth initial path of the ith stop end only considering obstacle avoidance, is an estimation path cost function of the jth point of the mth initial path of the ith stop end only considering vehicle kinematics, is an existing cost of the jth point of the mth initial path of the ith stop end, is an additional penalty of the jth point of the mth initial path of the ith stop end to the child node expansion, a mi is a reverse cost weight coefficient of the mth initial path of the ith stop end, is a reverse cost of the jth point of the mth initial path of the ith stop end, b mi is a turning cost weight coefficient of the mth initial path of the ith stop end, is a turning cost of the jth point of the mth initial path of the ith stop end, c mi is a turning radius size weight coefficient of the mth initial path of the ith stop end, is a turning radius size of the jth point of the mth initial path of the ith stop end, d mi is a front wheel turning angle weight coefficient of the mth initial path of the ith stop end, The front wheel turning angle of the i-th stop end point of the m-th initial path of the j-th point.

[0026] Further, the kinematic parameters of the mining truck include: the length, the width and the minimum turning radius of the mining truck.

[0027] The turning radius of the path is within the range of the minimum turning radius of the mining truck and a preset turning radius, so that the path curve curvature is smooth, and the front wheel turning angle of the mining truck at the path end point is within the range of the first turning angle threshold.

[0028] Further, the cloud control platform delimits the stop area range according to the loading and unloading area attribute, and the delimitation includes:

[0029] If it is a loading area, the excavator located in the loading area delimits the loading area stop area range through the excavator environment perception module and the second navigation positioning module, and uploads the delimitation to the cloud control platform;

[0030] If it is an unloading area, the cloud control platform determines the safe stop area range of the unloading area based on the first local map, sets a map edge and an obstacle edge in the first local map, and determines the safe stop area range of the unloading area.

[0031] Further, the cloud control platform monitors the use frequency of each stop end point, and suspends the use of the stop end point when the use frequency exceeds a use frequency threshold.

[0032] Further, before the unmanned mining truck enters the loading and unloading area, the area laser radar perceives the road condition information of the loading and unloading area, generates a SLAM grid map, and uploads the SLAM grid map to the cloud control platform to generate the reference map.

[0033] Further, the obstacle position is determined by the vehicle-mounted laser radar and / or millimeter wave radar, and the first navigation positioning module.

[0034] Further, it further includes:

[0035] When the vehicle-mounted environment perception module detects obstacle information not recorded in the first local map during the stop process, the unmanned mining truck immediately brakes, uploads the obstacle position to the cloud control platform, and the cloud control platform updates the first local map information to the second local map;

[0036] The pose of the unmanned mining truck after braking is a second starting point pose, and the cloud control platform re-plans the path according to the second starting point pose and the second local map.

[0037] Compared with the prior art, the present application can at least realize one of the following beneficial effects:

[0038] 1. The open-pit mine loading and unloading area unmanned mine truck path planning method provided by the application, which is adapted to the open-pit mine loading and unloading area scene through an intelligent path planning algorithm.

[0039] 2. The selection of the parking endpoint of the path, the segmented planning, the curvature optimization and the path smoothing ensure the safety and durability of the mine truck during the loading and unloading operation.

[0040] 3. The path generation method can ensure the universality and durability of the application of the unmanned mine truck in the loading and unloading area parking and driving out of the scene.

[0041] 4. Through real-time map updating, obstacle avoidance during travel is realized, thereby replacing the existing non-intelligent sampling scheme, avoiding system shutdown caused by insufficient adaptability of the sampling scheme, and realizing the efficiency of the path planning scheme.

[0042] In the application, the above technical solutions can be combined with each other to realize more preferred combination solutions. Other features and advantages of the application will be described in the subsequent specification, and some advantages will become apparent from the specification or be understood through the implementation of the application. The purpose and other advantages of the application can be achieved and obtained through the contents specifically indicated in the specification and the drawings. BRIEF DESCRIPTION OF DRAWINGS

[0043] The accompanying drawings are included to provide a further understanding of the application and are incorporated herein and constitute a part of the application. The same reference numerals in the several figures indicate the same or similar components.

[0044] Figure 1 The flowchart of the open-pit mine loading and unloading area unmanned mine truck path planning method shown in one embodiment of the application;

[0045] Figure 2 The collision-free path generation schematic diagram shown in one embodiment of the application. DETAILED DESCRIPTION

[0046] The preferred embodiments of the application will be specifically described below with reference to the accompanying drawings, wherein the drawings constitute a part of the application and are used to illustrate the principles of the embodiments of the application, but are not used to limit the scope of the application.

[0047] As shown in the drawings, Figure 1 one specific embodiment of the application discloses an open-pit mine loading and unloading area unmanned mine truck path planning method, which comprises:

[0048] S10, the first starting point pose of the mine truck driving into the loading and unloading area is obtained, the obstacle position is sensed in real time, the first starting point pose of the mine truck and the obstacle position are uploaded to the cloud control platform combined with the reference map, and the first local map is constructed.

[0049] wherein the mine card is an unmanned mine card, the first starting point pose refers to information such as a parking position, a parking direction and a front wheel turning angle of a starting point of the mine card, which is obtained by a vehicle-mounted laser radar and other environment perception modules and a navigation positioning module.

[0050] Specifically, before the unmanned mine card enters the loading and unloading area, the area laser radar perceives the road condition information of the loading and unloading area, generates a SLAM grid map, and uploads the SLAM grid map to the cloud control platform to generate the reference map.

[0051] The obstacle position is determined by the vehicle-mounted laser radar and / or millimeter wave radar of the mine card and the first navigation positioning module. More specifically, the cloud control platform combines the first starting point pose of the mine card, the obstacle position and the reference map to construct a first local map with obstacle position information.

[0052] S20, the cloud control platform delimits a parking area range and determines a plurality of parking endpoints according to the attribute of the loading and unloading area;

[0053] Specifically, the attribute of the loading and unloading area includes a loading area and an unloading area; the cloud control platform delimits a parking area range according to the attribute of the loading and unloading area, including:

[0054] If it is a loading area, the excavator located in the loading area delimits the loading area parking area range through the excavator environment perception module and the second navigation positioning module, and uploads it to the cloud control platform;

[0055] If it is an unloading area, the cloud control platform sets a map edge and an obstacle edge in the first local map based on the first local map to determine a safe unloading area parking area range. More specifically, the cloud control platform directly selects the unloading area parking area range for the parking endpoint selection according to the grid map edge safety distance and the obstacle edge safety distance information; alternatively, the grid map edge safety distance and the obstacle edge safety distance information can be set artificially.

[0056] Specifically, within the safe parking area range, a certain number of parking points are selected according to the kinematic parameters of the mine card, the parking endpoint spacing and the parking pose are set by the cloud control platform based on the pose difference [dx, dy, dtheta] between the starting point start_point and the parking endpoint end_point, and the weighted evaluation function score(start_point, end_point) = f(dx, dy, dtheta) is evaluated. By assigning different weights to each term of the function, the screening of specific requirements can be realized, such as selecting the parking point that can minimize the change of the start and end pose angle in this embodiment, although the parking point that is easier to reach can reduce the difficulty of path implementation.

[0057] Specifically, the cloud control platform monitors the usage count of each docking point. When the usage count exceeds a threshold, the use of that docking point is suspended. More specifically, due to the weight of mining trucks, repeated use of the same docking point can cause deformation of the docking area, leading to danger. Therefore, the cloud control platform needs to consider the usage count (times) of each docking area to achieve balanced use of each area. Optionally, times can be set as a variable and a weighted evaluation function can be added to achieve a more comprehensive selection of multiple docking points. That is, after a task is completed, based on the status of the docking points after the task, the platform can choose to clear the usage count or deduct unusable docking points.

[0058] S30. The cloud control platform generates an optimal collision-free path based on the multiple stopping endpoints, the known pose of the first starting point, the first local map, and the kinematic parameters of the mining truck.

[0059] Specifically, based on the length and width dimensions of the mining truck, the parking point position, and the minimum turning radius, the path segmentation is determined. More specifically, under normal circumstances, a straight reverse drive is used to end the reverse, and the front wheel angle theta_front is fixed at 0 during the reverse process. Therefore, a reversing pre-set point is set to avoid turning the steering wheel all the way in the opposite direction while stationary. Therefore, the front wheel angle theta_front at the turning point between forward and reverse driving needs to be as small as possible. Thus, the parking path is divided into three segments during the path division process, as shown in the figure. Figure 2 As shown: ① is the first segment, driving forward from the starting point to the turning point; ② is the second segment, driving backward from the turning point to the reversing reference point; ③ is the third segment, driving backward from the reversing reference point to the stopping point.

[0060] Specifically, generating the optimal collision-free path includes:

[0061] S301. Based on the kinematic parameters of each stopping point and the mining truck, determine the position and turning radius of each reversing reserved point; the reversing reserved point is a position set at a distance in a straight line forward from the stopping point; optionally, the set distance is 1.5 times the length of the mining truck.

[0062] Specifically, the kinematic parameters of the mining truck include: the truck's length, width, and minimum turning radius;

[0063] Specifically, the turning radius of the path is within the minimum turning radius and turning radius threshold range of the mining truck, ensuring that the curvature of the path curve is smooth, and the front wheel turning angle range of the mining truck at the end of the path is within the first turning angle threshold range.

[0064] More specifically, a turning radius range larger than the minimum turning radius is provided according to a vehicle kinematic model, and in the curvature optimization, the turning radius can be selected between the minimum turning radius turn ing_radius_min and a preset turning radius turn ing_radius_1 larger than the minimum turning radius, so as to achieve the smoothness of the path curvature and the curvature rate while minimizing the front wheel turning angle theta_front at the turning point. For example, when the minimum turning radius of the mining truck in the kinematic model is 10 meters and the front wheel turning angle range is [-35°, 35°], the turning_radius_1 is set to 15 meters to ensure the smoothness of the curve curvature, and the front wheel turning angle range [-5°, 5°] at the end of the setting is to ensure the smoothness of the connection of different path segments.

[0065] S302, according to the pose of each said reverse reservation point, the pose of the first starting point and the path turning radius, combining the first heuristic function and the second heuristic function, determining the intersection point of the first segment initial path and the second segment initial path corresponding to each said reverse reservation point pose;

[0066] The first heuristic function is calculated from the first starting point; and the second heuristic function is calculated from the reverse reservation point.

[0067] The first or second heuristic function is expressed as:

[0068]

[0069]

[0070]

[0071]

[0072]

[0073] wherein m=1 is the first heuristic function corresponding parameter, and m=2 is the second heuristic function corresponding parameter; f mi is the total cost of the mth segment initial path corresponding to the mth heuristic function of the ith stop end, is the cost of the jth point of the mth segment initial path of the ith stop end, and J is the number of nodes of the mth segment initial path of the ith stop end, is the search cost estimation function of the jth point of the mth segment initial path of the ith stop end, is the pre-generated path sub-node cost function of the jth point of the mth segment initial path of the ith stop end, is the estimated path cost function of the jth point of the mth segment initial path of the ith stop end only considering obstacle avoidance, an estimated path cost function for the mth segment of the initial path for the ith stop end point at the jth point only considering vehicle kinematics, an existing cost for the mth segment of the initial path for the ith stop end point at the jth point, an additional penalty for the mth segment of the initial path for the ith stop end point at the jth point to expand to a child node, a mi a reverse cost weight coefficient for the mth segment of the initial path for the ith stop end point, a reverse cost for the mth segment of the initial path for the ith stop end point at the jth point, b mi a turning cost weight coefficient for the mth segment of the initial path for the ith stop end point, a turning cost for the mth segment of the initial path for the ith stop end point at the jth point, c mi a turning radius size weight coefficient for the mth segment of the initial path for the ith stop end point, a turning radius size for the mth segment of the initial path for the ith stop end point at the jth point, d mi a front wheel steering angle weight coefficient for the mth segment of the initial path for the ith stop end point, a front wheel steering angle for the mth segment of the initial path for the ith stop end point at the jth point.

[0074] More specifically, the first heuristic function mainly considers obstacle avoidance, and determines the turning point and the first path in combination with the second heuristic function, thus, giving greater weights to the reverse cost and the turning cost, and in the estimated path cost function of the first heuristic function only considering vehicle kinematics is prior to the estimated path cost function only considering vehicle kinematics so that the function body has a greater weight coefficient, the reverse cost weight a 1i and the turning cost weight b 1i are greater, the turning radius size weight c 1i is smaller, and the front wheel steering angle weight d 1i at the end is maximum.

[0075] More specifically, the second heuristic function mainly considers pose change, and in the estimated path cost function of the second heuristic function only considering vehicle kinematics is prior to the estimated path cost function only considering obstacle avoidance the estimated path cost function only considering obstacle avoidance the function body has a greater coefficient, the reverse cost weight a 2i and the turning cost weight b 2i are smaller, the turning radius size weight c 2i is greater, and the front wheel steering angle weight d 2i at the end is maximum.

[0076] The intersection point of the first initial path and the second initial path corresponding to each of the reverse reserving point poses is determined by combining the first heuristic function and the second heuristic function, including: the first heuristic function takes the starting point pose as the starting point, and plans the path forward according to the set conditions of the first heuristic function; the second heuristic function takes a reverse reserving point pose as the starting point, and plans the path forward according to the set conditions of the first heuristic function; until the paths planned by the first heuristic function and the second heuristic function intersect, the first heuristic function and the second heuristic function stop calculating, and the intersection point corresponds to the reverse reserving point, that is, the stop end point.

[0077] In each of the intersection points, the intersection point with the minimum sum of total costs of the first heuristic function and the second heuristic function is selected as the turning point, and the corresponding stop end point is the optimal stop end point.

[0078] The path from the first starting point pose to the turning point is the first optimal path, and the mine truck drives forward along the first optimal path.

[0079] The path from the turning point to the reverse reserving point is the second optimal path, and the mine truck drives backward along the second optimal path.

[0080] The path from the reverse reserving point to the optimal stop end point is the third optimal path, and the mine truck drives backward along the third optimal path.

[0081] Specifically, the cloud control platform sends the generated optimal path to the unmanned mine truck, and the unmanned mine truck starts the stopping process along the path; during the stopping process, the vehicle-mounted laser radar and / or multi-millimeter wave radar, and other environment perception modules, are used to monitor the obstacles not recorded in the local map in real time; if the vehicle-mounted environment perception module detects obstacle information not recorded in the first local map during the stopping process, the unmanned mine truck immediately brakes, and uploads the obstacle position to the cloud control platform, and the cloud control platform updates the first local map information to the second local map.

[0082] The pose of the unmanned mine truck after braking is the second starting point pose, and the cloud control platform re-plans the path based on the second starting point pose and the second local map. More specifically, the cloud control platform re-divides the stopping area based on the second starting point pose after braking, allocates the stop end point, and re-plans the optimal path using the above steps S301-S306 based on the second local map and the start and end positions of the mine truck.

[0083] Specifically, after the unmanned mine truck completes the stopping and waits for the loading and unloading task to be completed, it is driven out by the cloud control platform.

[0084] In the embodiment, for the unmanned mine truck path planning problem of the loading and unloading area of the open-pit mine, a segmented multi-path route is formed through an intelligent path planning algorithm, which can be deeply adapted to the loading and unloading area scene of the open-pit mine. The loading and unloading area actually has a slope, so it is easy to cause rollover or even rollover due to errors. Through the selection of the parking end point of the path, segmented planning, curvature optimization and path smoothing, the safety and durability of the mine truck during the loading and unloading operation process are ensured. It can effectively prevent the unmanned mine truck from turning too much and the steering wheel from being reversed in place when parking on a road surface with high friction, which can cause great damage to the durability of the tire and the power-assisted steering system of the unmanned mine truck, thereby causing safety hazards. The path generation method can ensure the universality and durability of the unmanned mine truck in the loading and unloading area. Through the evaluation and selection of multiple parking points, the loading and unloading area of the open-pit mine has multiple parking points, and the best parking point is reasonably allocated to the mine truck from different starting points, which can further improve the efficiency and durability. Through real-time map updating, obstacle avoidance during travel is realized, thereby replacing the existing non-intelligent sampling scheme, avoiding system downtime caused by insufficient adaptability of the sampling scheme, and realizing the efficiency of the path planning scheme. In the event of an emergency stop caused by sand or other obstacles, the system will be paralyzed and will need external force to clean the obstacles, which does not have the problem of high efficiency.

[0085] Those skilled in the art can understand that all or part of the processes of the above-mentioned embodiments can be completed by a computer program instructing related hardware, and the program can be stored in a computer readable storage medium. The computer readable storage medium includes a magnetic disk, an optical disk, a read-only memory, a random access memory, etc.

[0086] The above is only a preferred specific embodiment of the present application, but the protection scope of the present application is not limited thereto. Any skilled person in the art can easily think of changes or replacements within the technical scope disclosed by the present application, which should be covered within the protection scope of the present application.

Claims

1. A method for unmanned mining truck path planning in open-pit mine loading and unloading areas, characterized in that, include: The first starting point pose of the mining truck entering the loading and unloading area is obtained, the position of obstacles is perceived in real time, and the first starting point pose of the mining truck and the position of obstacles are uploaded to the cloud control platform and combined with the reference map to construct the first local map. The cloud control platform defines the parking area and determines multiple parking destinations based on the attributes of the loading and unloading area. The cloud control platform generates an optimal collision-free path based on the multiple stopping endpoints, the pose of the first starting point, the first local map, and the kinematic parameters of the mining truck. The generation of the optimal collision-free path includes: Based on the kinematic parameters of each stopping point and the mining truck, determine the position and turning radius of each reversing reserved point; The reversing reference point is a position set at a straight line distance ahead of the parking endpoint; Based on the pose of each reversing reserved point, the pose of the first starting point, and the turning radius of the path, and combined with the first heuristic function and the second heuristic function, the intersection point of the first initial path and the second initial path corresponding to each reversing reserved point pose is determined. In each of the aforementioned intersection points, the intersection point with the minimum sum of the total costs of the first and second heuristic functions is selected as the turning point, and the corresponding stopping point is the optimal stopping point. The path points from the first starting point pose to the turning point form the first optimal path, and the mining truck travels forward along the first optimal path. The path points from the turning point to the reversing reserved point form the second optimal path, and the mining truck reverses along the second optimal path. The reversing reserved point to the optimal stopping point is the third optimal path, and the mining truck reverses along the third optimal path.

2. The unmanned mining truck path planning method for open-pit mine loading and unloading areas according to claim 1, characterized in that, The first heuristic function is calculated starting from the first starting point pose; the second heuristic function is calculated starting from the reversing reserved point pose.

3. The unmanned mining truck path planning method for open-pit mine loading and unloading areas according to claim 1, characterized in that, The first and second heuristic functions are expressed as follows: in, The time is the parameter corresponding to the first heuristic function. The time is the parameter corresponding to the second heuristic function; For the first The first stop at the destination The heuristic function corresponding to the first The total cost of the initial path segment. For the first The first stop at the destination Initial path segment The cost of points For the first The first stop at the destination The number of nodes in the initial path of the segment. For the first The first stop at the destination Initial path segment The search cost prediction function for points. For the first The first stop at the destination Initial path segment The cost function for pre-generated path child nodes of a point. For the first The first stop at the destination Initial path segment The point only considers the predicted path cost function for obstacle avoidance. For the first The first stop at the destination Initial path segment The predicted path cost function only considers vehicle kinematics. For the first The first stop at the destination Initial path segment There is already a price to pay. For the first The first stop at the destination Initial path segment Additional penalty for extending a node to its child nodes. For the first The first stop at the destination The reversing cost weighting coefficient of the initial path segment. No. The first stop at the destination Initial path segment The cost of reversing For the first The first stop at the destination The turning cost weighting coefficient of the initial path segment. No. The first stop at the destination Initial path segment The cost of turning at a point For the first The first stop at the destination The weighting coefficient of the turning radius of the initial path segment. No. The first stop at the destination Initial path segment The turning radius of the point, For the first The first stop at the destination The front wheel steering angle weighting coefficient of the initial path segment. No. The first stop at the destination Initial path segment The front wheel steering angle at the point.

4. The unmanned mining truck path planning method for open-pit mine loading and unloading areas according to claim 1, characterized in that, The kinematic parameters of the mining truck include: the truck length, width, and minimum turning radius; The turning radius of the path is within the range of the minimum turning radius and the preset turning radius of the mining truck, ensuring that the curvature of the path curve is smooth, and that the front wheel turning angle of the mining truck is within the range of the first turning angle threshold at the end of the path.

5. The unmanned mining truck path planning method for open-pit mine loading and unloading areas according to claim 1, characterized in that, The cloud control platform defines the parking area range based on the loading / unloading area attributes, including: If it is a loading area, the excavator located in the loading area uses the excavator environment perception module and the second navigation and positioning module to delineate the parking area of ​​the loading area and upload it to the cloud control platform; If it is an unloading area, the cloud control platform determines the safe parking area range of the unloading area by setting map edges and obstacle edges in the first local map.

6. The unmanned mining truck path planning method for open-pit mine loading and unloading areas according to claim 5, characterized in that, The cloud control platform monitors the number of times each docking point is used, and suspends the use of this docking point when the number of uses exceeds the usage threshold.

7. The unmanned mining truck path planning method for open-pit mine loading and unloading areas according to claim 1, characterized in that, Before the unmanned mining truck enters the loading and unloading area, the regional lidar senses the road conditions in the loading and unloading area, generates a SLAM grid map, and uploads the SLAM grid map to the cloud control platform to generate the base map.

8. The unmanned mining truck path planning method for open-pit mine loading and unloading areas according to claim 1, characterized in that, The location of the obstacle is determined by vehicle-mounted lidar and / or millimeter-wave radar, as well as the first navigation and positioning module.

9. The unmanned mining truck path planning method for open-pit mine loading and unloading areas according to claim 1, characterized in that, Also includes: When the vehicle-mounted environmental perception module detects obstacle information that has not been recorded in the first local map during the parking process, the unmanned mining truck immediately brakes and uploads the obstacle location to the cloud control platform. The cloud control platform then updates the first local map information to the second local map. The pose of the unmanned mining truck after braking is the second starting point pose, and the cloud control platform re-plans the path based on the second starting point pose and the second local map.

Images