Mining Truck Obstacle Avoidance Path Planning Method and System

By narrowing the search range, integrating bounding boxes and boundaries, segmented planning, and real-time perception in the obstacle avoidance path planning of mining trucks, the problem of excessively long calculation time for obstacle avoidance paths of mining trucks has been solved, achieving efficient and safe path generation and meeting the real-time requirements of high-frequency operations in mining areas.

CN123111289APending Publication Date: 2026-11-17SANY INTELLIGENT MINING TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611448024.2
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-09-17
Publication Date
2026-11-17

AI Technical Summary

Technical Problem

In existing technologies, the calculation of obstacle avoidance path planning for mining trucks takes too long, making it difficult to meet the real-time requirements of high-frequency, continuous operations in mining areas.

Method used

By narrowing the search range of the path search algorithm, utilizing prior knowledge of the loading area operation scenario to tailor the search space, combining bounding box modeling of the mining truck with the fusion of the drivable area boundary, obstacle avoidance paths are generated, and sensing devices are introduced for real-time updates. Segmented path planning and safety detection are adopted to reduce the algorithm complexity.

Benefits of technology

It significantly improves the efficiency and safety of path planning, and can generate reliable obstacle avoidance paths in milliseconds, meeting the real-time requirements of high-frequency and continuous operations in mining areas, reducing computational overhead, and improving adaptability.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123111289A_ABST
    Figure CN123111289A_ABST
Patent Text Reader

Abstract

This application relates to the field of autonomous driving technology, specifically to a method and system for obstacle avoidance path planning for mining trucks. The method includes: determining the search space of a path search algorithm based on the operational scenario of the loading area; identifying obstacle mining trucks based on their status and performing bounding box modeling on these obstacle trucks; obtaining the boundary of a base map, which includes the spatial distribution of unexcavated and excavated areas within the loading area, and determining the boundary of the drivable area based on the base map boundary and the bounding boxes of the obstacle mining trucks; performing bounding box modeling on the current mining truck and geometrically fusing the bounding box of the current mining truck with the boundary of the drivable area to determine the boundary of the passable area; and generating an obstacle avoidance path within the search space, constrained by the boundary of the passable area, based on the path search algorithm. The obstacle avoidance path planning method for mining trucks provided in this application can meet the real-time requirements of high-frequency, continuous operations in mining areas.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This application relates to the field of autonomous driving technology, and more specifically, to a method and system for obstacle avoidance path planning for mining trucks. Background Technology

[0002] In open-pit mining operations, the collaborative operation of mining trucks (hereinafter referred to as "mining trucks") and excavators is a core production link. In the loading area, the excavator is usually located in the unexploded original ore pile area, while the mining truck drives into the loading point in reverse, with its rear end close to the excavator's working range, to complete the loading of materials.

[0003] To improve overall mining efficiency, the loading area often adopts an intensive operation mode of "dual loading points and dual waiting points." In this mode, when a mining truck finishes loading and needs to leave the loading area, there is usually another mining truck waiting to be loaded about 10 meters ahead, forming a dynamic obstacle. Because the roads in the mining area are unstructured and the space is extremely narrow, if the waiting mining truck is tilted or misaligned while parking, it will directly block the path of the exiting mining truck, forcing the exiting truck to perform obstacle avoidance operations within a limited area.

[0004] Currently, existing technologies for obstacle avoidance in narrow passages mostly employ the traditional Hybrid A* search algorithm for path planning. However, this method has significant inherent drawbacks in the context of mining loading areas. First, when expanding nodes in a narrow, multi-obstacle configuration space, the Hybrid A* algorithm generates a large number of search branches, resulting in excessively long computation times for obstacle avoidance paths, making it difficult to meet the real-time requirements of high-frequency, continuous operations in mining areas. Summary of the Invention

[0005] In view of this, this application aims to provide a method and system for planning obstacle bypass paths for mining trucks, which can solve the problem that the calculation time for obstacle bypass paths in related technologies is too long and it is difficult to meet the real-time requirements of high-frequency and continuous operations in mining areas.

[0006] To achieve the above objectives, the first aspect of this application provides a method for planning obstacle avoidance paths for mining trucks.

[0007] The second aspect of this application provides a mining truck obstacle avoidance path planning system.

[0008] According to a first aspect of this application, a method for planning obstacle avoidance paths for mining trucks is provided, comprising: determining the search space of a path search algorithm based on the operational scenario of the loading area; identifying obstacle mining trucks based on the status of the mining trucks and performing bounding box modeling on the obstacle mining trucks; obtaining the boundary of a base map, which includes the spatial distribution of unexcavated and excavated areas within the loading area, and determining the boundary of a drivable area based on the boundary of the base map and the bounding box of the obstacle mining trucks; performing bounding box modeling on the current mining truck and geometrically merging the bounding box of the current mining truck and the boundary of the drivable area to determine the boundary of the drivable area; and generating an obstacle avoidance path within the search space, constrained by the boundary of the drivable area, based on the path search algorithm.

[0009] In the above technical solution, the search range of the path search algorithm is pre-narrowed by utilizing the operational scenario of the loading area, allowing the algorithm to run in areas where effective paths are likely to be generated. This reduces the expansion of invalid nodes and improves the convergence speed. Simultaneously, the bounding box of the current mining truck and the boundary of the actual drivable area are merged into a unified drivable area constraint, making the vehicle body and the boundary spatially integrated. This naturally aligns with the input requirement of the hybrid A* algorithm for obstacle spatial rasterization, eliminating the need for additional complex obstacle abstraction transformations and thus reducing the deployment difficulty of the algorithm. In this way, the real-time requirements of high-frequency, continuous operations in mining areas can be met.

[0010] In some technical solutions, the length of the search space can be optionally set as the longitudinal distance from the current position of the current mining card to the boundary of the loading area outlet, and the width of the search space can be determined according to the working width of the loading area.

[0011] This configuration allows the search space of the path search algorithm to be dynamically adjusted according to the current location of the mining card and the geometric features of the work area. It ensures that critical path areas are not overlooked and reduces invalid calculations, thereby further improving the efficiency of path planning and better meeting the real-time requirements of high-frequency, continuous operations in the mining area.

[0012] In some technical solutions, the obstacle avoidance path planning method for mining trucks may optionally include: using the sensing devices carried by the mining trucks during entry and exit processes to collect real-time data on the obstacle boundaries around the loading area in order to update the boundaries of the drivable area.

[0013] In the above technical solution, the sensing devices carried by the current mining trucks during entry and exit processes are used to collect the obstacle boundaries around the loading area in real time. This allows for the updating of the boundaries of the drivable area, thereby further improving the safety and reliability of obstacle avoidance path planning and enhancing the adaptive capability of path planning.

[0014] In some technical solutions, optionally, the geometry of the enclosure of the obstacle mining card is greater than or equal to the safety verification threshold.

[0015] In the above technical solution, by setting a safety verification threshold and making the geometric dimensions of the bounding box of the obstacle mine card greater than or equal to the safety verification threshold, high-risk areas near obstacles can be avoided during path planning. This allows the generated obstacle bypass path to have sufficient safety buffer, eliminating the need for additional safety verification or path correction after planning is completed. This further shortens the overall time of path planning, better meeting the real-time requirements of high-frequency, continuous operations in the mining area.

[0016] In some technical solutions, optionally, based on a path search algorithm, within the search space and constrained by the boundary of the passable area, an obstacle bypass path is generated. Specifically, this includes: starting from the search starting position in front of the current mining truck, performing discrete point sampling within the search space to obtain candidate points; performing safety checks on the candidate points based on the boundary of the passable area, the bounding box of the obstacle mining truck, and the bounding box of the current mining truck, and selecting candidate points that pass the safety check as intermediate points; generating a first path using the current position of the current mining truck as the starting point and the intermediate point as the ending point using the path search algorithm; generating a second path using the intermediate point as the starting point and the exit of the loading area as the ending point; and concatenating the first path and the second path to obtain the obstacle bypass path.

[0017] In the above technical solution, the path planning is divided into two segments: from the current position of the mining truck to the midpoint, and from the midpoint to the loading area exit. This decomposes the complex single-segment obstacle avoidance path planning into two relatively simple forward path search subproblems, which can significantly reduce the node expansion scale of the path search algorithm and lower the complexity of the path search. Furthermore, since each segment is a forward search, it can improve the reversing and reversing requirements caused by branch explosion in single-segment search in narrow spaces, ensuring that the entire path consists only of forward trajectories.

[0018] Meanwhile, this application introduces bounding box collision detection of mining trucks corresponding to path points as a safety criterion during the planning process, which can ensure that each pose point on the path meets the spatial safety margin requirements, thus providing a reliable foundation for subsequent high-precision trajectory tracking control.

[0019] In some technical solutions, optionally, discrete point sampling search is performed within the search space, specifically including: discretizing along the width direction of the search space within the search space, and sampling search at each discrete position along the direction perpendicular to the width.

[0020] This setup allows for the traversal of potential paths within the search space, reducing the risk of overlooking key areas. It also provides an ordered queue of candidate points for subsequent path search methods, facilitating the algorithm to attempt expansions sequentially and further improving planning efficiency.

[0021] In some technical solutions, optionally, the safety check includes: whether the candidate point is located inside the boundary of the drivable area, whether the bounding box of the current mining card corresponding to the candidate point collides with the boundary of the drivable area, and whether the bounding box of the current mining card corresponding to the candidate point overlaps with the bounding box of the obstacle mining card.

[0022] This setup ensures that each pose point on the path meets the spatial safety margin requirements, thus providing a reliable foundation for subsequent high-precision trajectory tracking control.

[0023] In some technical solutions, optionally, a second path is generated with the intermediate point as the starting point and the outlet of the loading area as the ending point. Specifically, this includes using a hybrid A* algorithm or a Reeds-Shepp curve algorithm to search and generate a second path connecting the intermediate point and the outlet of the loading area.

[0024] In the above technical solution, by introducing the Reeds-Shepp curve algorithm as an alternative to the hybrid A* algorithm, the path generation efficiency can be significantly improved and the computational overhead can be reduced in barrier-free or low-barrier environments, while ensuring the executability of the path.

[0025] According to a second aspect of this application, a mining truck obstacle avoidance path planning system is provided, comprising: a mining truck; a roadside base station, and a communication connection with the mining truck; the roadside base station is used to store operational scenario information of the loading area and can implement the steps of the mining truck obstacle avoidance path planning method provided by any of the above technical solutions to generate an obstacle avoidance path. Therefore, the mining truck obstacle avoidance path planning system includes all the beneficial effects of any of the above technical solutions, which will not be elaborated further here.

[0026] In some technical solutions, the mining truck obstacle avoidance path planning system may optionally include a cloud platform; the cloud platform is communicatively connected to the roadside base station and is used to receive and store the operational data uploaded by the roadside base station.

[0027] In the above technical solution, by setting up a cloud platform, managers can view the obstacle avoidance status of each mine card in the mining area in real time through the cloud platform, and make manual intervention or global scheduling when necessary.

[0028] Additional aspects and advantages of the technical solutions of this application will become apparent in the following description or may be learned by practice of this application. Attached Figure Description

[0029] Figure 1 This is one of the flowcharts illustrating the obstacle avoidance path planning method for mining trucks in some embodiments of this application;

[0030] Figure 2 This is a second flowchart illustrating the obstacle avoidance path planning method for mining trucks in some embodiments of this application;

[0031] Figure 3 This is a schematic diagram of the loading of the mining truck in some embodiments of this application;

[0032] Figure 4 This is a schematic diagram of the movement of the mining card in some embodiments of this application;

[0033] Figure 5 This is a schematic diagram of the framework of a mining truck obstacle avoidance path planning system in some embodiments of this application.

[0034] Figure label:

[0035] 100 mining cards; 200 roadside base stations; 300 cloud platforms. Detailed Implementation

[0036] The technical solutions of the embodiments of this application will be clearly described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application are within the scope of protection of this application.

[0037] The terms "first," "second," etc., used in the specification and claims of this application are used to distinguish similar objects and not to describe a specific order or sequence. It should be understood that such use of data can be interchanged where appropriate so that embodiments of this application can be implemented in orders other than those illustrated or described herein, and the objects distinguished by "first," "second," etc., are generally of the same class and the number of objects is not limited; for example, a first object can be one or more. Furthermore, in the specification and claims, "and / or" indicates at least one of the connected objects, and the character " / " generally indicates that the preceding and following objects are in an "or" relationship.

[0038] The following is combined with Figures 1 to 5 The method and system for planning obstacle bypass paths for mining trucks provided in this application will be described in detail through specific embodiments and application scenarios.

[0039] Reference Figure 1 , Figure 3 and Figure 4 In some embodiments, this application provides a method for planning obstacle avoidance paths for mining trucks, including the following steps:

[0040] S102, Determine the range of the search space for the path search algorithm based on the operating scenario of the loading area;

[0041] S104, Based on the status of the mining card, identify the obstacle mining card and perform bounding box modeling on the obstacle mining card;

[0042] S106, Obtain the basic map boundary. The basic map boundary includes the spatial distribution of the unexcavated area and the excavated area within the loading area. Determine the boundary of the drivable area based on the basic map boundary and the bounding box of the obstacle mine card.

[0043] S108, perform bounding box modeling on the current mining card, and geometrically merge the bounding box of the current mining card with the boundary of the drivable area to determine the boundary of the drivable area;

[0044] S110, based on a path search algorithm, generates obstacle avoidance paths within the search space, constrained by the boundaries of passable areas.

[0045] Specifically, prior knowledge of the work scenario, such as the geographical coordinates of the loading area, the working range of the excavator, and common driving routes, is used to trim the search space. This narrows the search space, allowing the path search algorithm to run in areas where effective paths are likely to be generated, thereby improving the search convergence speed, reducing invalid calculations, and thus reducing the computation time of path planning and the complexity of the algorithm to meet the real-time requirements of high-frequency, continuous operations in the mining area.

[0046] The path search algorithm can be a hybrid A* algorithm, which is an improved algorithm for vehicle path planning. Its characteristic is that it combines discrete grid search with a continuous vehicle kinematics model to generate a smooth path that conforms to the vehicle's actual steering capabilities.

[0047] It should be understood that path search algorithms can also be sampling algorithms, such as RRT (Rapidly-exploring Random Tree).

[0048] Bounding box modeling is a geometric simplification method that uses a simple, regular geometric shape (such as a cuboid or sphere) to approximate the shape of a complex object, in order to facilitate rapid collision detection, space occupancy assessment, or path planning.

[0049] Understandably, the irregular shape of the mining truck makes it easy to quickly determine whether it interferes with the obstacle-avoiding mining truck by using bounding box modeling, which can abstract the complex shape of the mining truck into a regular geometric shape. This reduces computational costs and makes it easier to quickly determine whether there is interference with the obstacle-avoiding mining truck.

[0050] Subsequently, the boundaries of the drivable area are determined based on the bounding boxes of the base map and obstacle mines, thus providing an initial boundary constraint for path search.

[0051] It should be understood that the mining truck itself has physical dimensions and cannot be considered a point mass. Therefore, after determining the boundary of the drivable area, a bounding box model is performed on the current mining truck, and the bounding box of the current mining truck is geometrically merged with the boundary of the drivable area. This converts the physical dimensions of the current mining truck itself into a safety margin to determine the boundary of the drivable area. In this way, the boundary of the drivable area has actually deducted the space occupied by the current mining truck itself. During subsequent path search, the path search algorithm only needs to treat the current mining truck as a point mass to ensure that the vehicle body does not collide with obstacles or terrain boundaries during actual driving. This makes the vehicle body and the boundary integrated in spatial representation, which meets the input requirement of the hybrid A* algorithm for obstacle spatial rasterization. No additional complex obstacle abstraction transformation is required, thereby reducing the deployment difficulty of the algorithm.

[0052] Among them, geometric fusion can refer to equidistantly shrinking the current mining card's bounding box along the inner side of the boundary of the drivable area.

[0053] Finally, within the search space, and constrained by the boundaries of the passable area, a path search algorithm is used to generate obstacle avoidance paths.

[0054] In the above embodiments, the search range of the path search algorithm is pre-narrowed by utilizing the operational scenario of the loading area, allowing the algorithm to run in areas where effective paths are likely to be generated. This reduces the expansion of invalid nodes and improves the convergence speed. Simultaneously, the bounding box of the current mining truck and the boundary of the actual drivable area are merged into a unified drivable area constraint, making the vehicle body and the boundary spatially integrated. This naturally aligns with the input requirement of the hybrid A* algorithm for obstacle spatial rasterization, eliminating the need for additional complex obstacle abstraction transformations and thus reducing the deployment difficulty of the algorithm. In this way, the real-time requirements of high-frequency, continuous operations in mining areas can be met.

[0055] In some embodiments, the length of the search space is set as the longitudinal distance from the current position of the current mining card to the outlet boundary of the loading area, and the width of the search space is determined according to the working width of the loading area.

[0056] Specifically, the length of the search space is set as the longitudinal distance from the current position of the mining card to the boundary of the loading area exit. With this setting, path planning only needs to cover the entire interval from the current position of the mining card to leaving the loading area, thereby reducing invalid searches to the rear or beyond the exit direction. This further compresses the search depth, reduces unnecessary node expansion, and improves the convergence speed.

[0057] Meanwhile, the width of the search space is determined based on the working width of the loading area. This ensures that the obstacle avoidance path has sufficient lateral adjustment margin, while preventing the introduction of a large number of invalid grids due to an excessively wide search range.

[0058] It should be understood that the width of the search space can be set based on the excavator's turning radius, the width of adjacent lanes, or the actual working width measured on site.

[0059] In the above embodiments, the search space is determined by the above method, so that the search space of the path search method can be dynamically adjusted according to the current location of the mining card and the geometric features of the work area. This ensures that critical path areas are not missed and that invalid calculations are reduced, thereby further improving the efficiency of path planning and better meeting the real-time requirements of high-frequency and continuous operations in the mining area.

[0060] In some embodiments, the obstacle avoidance path planning method for mining trucks further includes:

[0061] By utilizing the sensing devices currently installed on mining trucks during entry and exit, the boundary of obstacles around the loading area is collected in real time to update the boundary of the drivable area.

[0062] It is understandable that the working environment in the mining area is highly dynamic. As the excavators continue to work, the boundary between the unexcavated area and the excavated area will continue to change; other mining trucks may also change their position or status at any time.

[0063] Therefore, in the above embodiments, by using the sensing devices carried by the current mining trucks during the entry and exit processes to collect the obstacle boundaries around the loading area in real time, the boundaries of the drivable area can be updated, thereby further improving the safety and reliability of obstacle avoidance path planning and enhancing the adaptive capability of path planning.

[0064] It should be understood that sensing devices include, but are not limited to, one or more combinations of LiDAR, millimeter-wave radar, ultrasonic sensors, and visual cameras. During the process of a mining truck entering or leaving the loading area, the sensing devices can continuously scan the surrounding environment, detect changes in the boundaries of unexcavated areas, the expansion of excavated areas, the appearance of temporarily stockpiled materials, and the occupancy status of other mobile devices in real time, and fuse and compare this dynamic obstacle information with the pre-loaded base map boundaries.

[0065] In some embodiments, the geometry of the bounding box of the obstacle mining card is greater than or equal to the security verification threshold.

[0066] Understandably, if the bounding box of the obstacle mining truck only takes its physical outline size, the planned path may theoretically be collision-free, but in actual execution, it may approach or even touch the edge of the obstacle due to positioning errors, control deviations or sensor noise, causing subsequent modules to determine that it is unsafe and refuse to execute.

[0067] The safety verification threshold is a pre-agreed safety distance standard. When selecting a path, it is necessary to assess whether the clearance distance between the path and obstacles meets the safety margin requirements; when performing trajectory tracking, it is also necessary to verify whether the deviation between the actual vehicle posture and the obstacle is within the allowable range.

[0068] Therefore, in the above embodiments, by setting a safety verification threshold and making the geometric dimensions of the bounding box of the obstacle mine card greater than or equal to the safety verification threshold, high-risk areas near obstacles can be avoided during path planning, so that the generated obstacle bypass path has sufficient safety buffer. Thus, no additional safety verification or path correction needs to be performed after the planning is completed, thereby further shortening the overall time of path planning and better meeting the real-time requirements of high-frequency and continuous operations in the mining area.

[0069] Reference Figures 2 to 4 In some embodiments, based on a path search algorithm, within the search space and constrained by the boundaries of the passable area, a path around the obstacle is generated, specifically including:

[0070] S202, Starting from the current search start position in front of the mining card, perform discrete point sampling search within the search space to obtain candidate points;

[0071] S204. Based on the boundary of the passable area, the bounding box of the obstacle mining card, and the bounding box of the current mining card, a safety check is performed on the candidate points, and the candidate points that pass the safety check are selected as intermediate points.

[0072] S206: Starting from the current position of the mining card and using the intermediate point as the end point, the first path is generated using a path search algorithm.

[0073] S208, starting from the midpoint and ending at the outlet of the loading area, generates a second path;

[0074] S210, the first path and the second path are joined together to obtain the obstacle bypass path.

[0075] Specifically, starting from the current search starting position in front of the mining card, discrete point sampling is performed within the search space to obtain possible candidate points.

[0076] The search starting position is pre-set, determined based on the current location, heading, and a certain distance ahead of the mining truck. Specifically, it is set at a fixed offset position directly in front of the mining truck along the direction of travel, ensuring the search begins in front of the truck and avoiding searches from behind or to the side, thus saving computational resources. A buffer distance is also included between the starting point and the current truck's location to prevent overlap between the starting point and the actual truck body due to positioning errors.

[0077] After obtaining candidate points, collision safety checks are performed on the candidate points by combining the boundary of the passable area, the bounding box of the obstacle mining card, and the bounding box of the current mining card. Candidate points that pass the safety check are used as intermediate points.

[0078] At this point, taking the current position of the mining card as the starting point and the intermediate point as the ending point, a local obstacle avoidance path, namely the first path, is planned using a path search algorithm.

[0079] After the first path is successfully planned, the subsequent path, namely the second path, is planned with the midpoint as the starting point and the loading area exit as the ending point.

[0080] Finally, connecting the first and second paths will form a complete obstacle avoidance path.

[0081] In the above embodiment, the path planning is divided into two segments: from the current position of the current mining truck to the midpoint, and from the midpoint to the loading area exit. This decomposes the complex single-segment obstacle avoidance path planning into two relatively simple forward path search subproblems, significantly reducing the node expansion scale of the path search algorithm and lowering the complexity of the path search. Simultaneously, since each segment involves a forward search, it mitigates the reversing and reversing requirements caused by branch explosion in single-segment searches in narrow spaces, ensuring the entire path consists only of forward trajectories. This effectively avoids reversing behavior, thereby improving exit safety while significantly enhancing the continuity and efficiency of loading operations in the mining area.

[0082] Experimental results show that this application can generate effective obstacle avoidance paths in milliseconds (ms), and the spacing between adjacent path points is controlled in centimeters, which significantly reduces the real-time calculation of path planning and can fully meet the requirements of high-frequency continuous operation in mining areas for planning response speed.

[0083] Meanwhile, this application introduces bounding box collision detection of mining trucks corresponding to path points as a safety criterion during the planning process, which can ensure that each pose point on the path meets the spatial safety margin requirements, thus providing a reliable foundation for subsequent high-precision trajectory tracking control.

[0084] Furthermore, the aforementioned intermediate point-guided segmented planning strategy can be seamlessly integrated and collaboratively applied with existing mainstream path planning algorithms (such as distance sampling-based path search algorithms and B-spline curve smoothing algorithms), facilitating rapid integration and engineering implementation in existing autonomous driving systems. This makes it applicable to various models and sizes of mining dump trucks in mining scenarios.

[0085] In some embodiments, performing discrete point sampling search within the search space specifically includes: discretizing within the search space along the width direction of the search space, and performing sampling search at each discrete location along a direction perpendicular to the width.

[0086] Specifically, the search space is divided into several strips with equal or variable spacing along the width direction, and each strip corresponds to a fixed horizontal coordinate.

[0087] Subsequently, at each discrete lateral position, sampling is performed along a direction perpendicular to the width direction to obtain a series of candidate points.

[0088] This setup allows for the traversal of potential paths within the search space, reducing the risk of overlooking key areas. It also provides an ordered queue of candidate points for subsequent path search algorithms, facilitating sequential attempts to expand the search and further improving planning efficiency.

[0089] For example, when performing discrete sampling within the search space, a vertical sampling search is conducted along the width direction approximately 15 meters in front of the current mine card. If all candidate points along this vertical line fail the safety check, the search range is extended to approximately 30 meters in front of the current mine card, and vertical sampling continues until all possible positions within the entire search space have been traversed. This progressive search strategy prioritizes shorter obstacle avoidance paths while ensuring search completeness, thus reducing travel distance during obstacle avoidance.

[0090] In some embodiments, the security check specifically includes: whether the candidate point is located inside the boundary of the drivable area, whether the bounding box of the current mining card corresponding to the candidate point collides with the boundary of the drivable area, and whether the bounding box of the current mining card corresponding to the candidate point overlaps with the bounding box of the obstacle mining card.

[0091] Specifically, candidate points can only be used as intermediate points if they pass a security check, which includes three conditions:

[0092] (1) The candidate point itself is located inside the boundary of the passable area, so that the path does not cross the boundary.

[0093] (2) The current mining card's bounding box cannot collide with the boundary of the passable area so that the current mining card does not cross the line or exceed the boundary.

[0094] (3) The bounding box of the current mining card does not overlap with the bounding box of the obstacle mining card, that is, the two do not contact each other.

[0095] Only when all three conditions are met can a candidate point pass the safety check and become an intermediate point in the path planning.

[0096] This setup ensures that each pose point on the path meets the spatial safety margin requirements, thus providing a reliable foundation for subsequent high-precision trajectory tracking control.

[0097] In some embodiments, generating a second path with the midpoint as the starting point and the outlet of the loading area as the ending point specifically includes: using a hybrid A* algorithm or a Reeds-Shepp curve algorithm to search and generate a second path connecting the midpoint and the outlet of the loading area.

[0098] The Reeds-Shepp curve algorithm is based on the vehicle's minimum turning radius and forward / backward model. It is constructed from arcs, straight lines, and direction-switching segments and is suitable for fast and smooth connections between the intermediate point and the exit when there are no complex obstacles.

[0099] In the above embodiments, by introducing the Reeds-Shepp curve algorithm as an alternative to the hybrid A* algorithm, the path generation efficiency can be significantly improved and the computational overhead can be reduced in barrier-free or low-obstacle environments, while ensuring the executability of the path.

[0100] Reference Figures 1 to 5 In some embodiments, this application also provides a mining truck obstacle avoidance path planning system, including: a mining truck 100; a roadside base station 200, which is communicatively connected to the mining truck 100; the roadside base station 200 is used to store the operation scene information of the loading area and can implement the steps of the mining truck obstacle avoidance path planning method provided in any of the above embodiments to generate an obstacle avoidance path. Therefore, the mining truck obstacle avoidance path planning system has all the beneficial effects of any of the above embodiments, which will not be elaborated further here.

[0101] Among them, the Roadside Base Station (RSU) 200 refers to fixed communication and computing equipment deployed on both sides of the road or mining operation area. Through the V2X (Vehicle-to-Everything) wireless communication protocol, it exchanges real-time data with passing mining trucks 100, transmitting road conditions, obstacle information, dispatch instructions, etc.

[0102] In some embodiments, the mining truck obstacle avoidance path planning system further includes a cloud platform 300. The cloud platform 300 is communicatively connected to the roadside base station 200 and is used to receive and store the operational data uploaded by the roadside base station 200.

[0103] The cloud platform 300 refers to the software and hardware system deployed in a remote data center or cloud server, possessing large-scale data storage, high-performance computing, and distributed processing capabilities. In the unmanned driving scenario in the mining area, the cloud platform 300 is responsible for aggregating data uploaded from multiple roadside base stations 200 and mining trucks 100, and providing advanced functions such as global scheduling, historical data backtracking, and machine learning model training.

[0104] In the above embodiments, by setting up a cloud platform 300, managers can view the obstacle avoidance status of each mining card in the mining area in real time through the cloud platform 300, and make manual intervention or global scheduling when necessary.

[0105] It should be noted that, in this document, 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 limitations, an element defined 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. Furthermore, it should be noted that the scope of the methods and apparatuses in the embodiments of this application is not limited to performing functions in the order shown or discussed, but may also include performing functions substantially simultaneously or in the reverse order, depending on the functions involved. For example, the described methods may be performed in a different order than described, and various steps may be added, omitted, or combined. Additionally, features described with reference to certain examples may be combined in other examples.

[0106] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a computer software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions to cause a terminal (which may be a mobile phone, computer, server, or network device, etc.) to execute the methods of the various embodiments of this application.

[0107] The embodiments of this application have been described above with reference to the accompanying drawings. However, this application is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of this application without departing from the spirit and scope of the claims, and all of these forms are within the protection scope of this application.

Claims

1. A method for planning obstacle avoidance paths for mining trucks, characterized in that, include: Based on the operational scenario in the loading area, determine the range of the search space for the path search algorithm; Based on the status of the mining card, identify the obstacle mining card and perform bounding box modeling on the obstacle mining card; Obtain the basic map boundary, which includes the spatial distribution of unexcavated and excavated areas within the loading area, and determine the boundary of the drivable area based on the basic map boundary and the bounding box of the obstacle mine card. A bounding box model is performed on the current mining truck, and the bounding box of the current mining truck and the boundary of the drivable area are geometrically merged to determine the boundary of the drivable area. Based on the path search algorithm, within the search space, an obstacle avoidance path is generated with the boundary of the passable area as a constraint.

2. The obstacle avoidance path planning method for mining trucks according to claim 1, characterized in that, The length of the search space is set as the longitudinal distance from the current position of the current mining card to the boundary of the loading area exit, and the width of the search space is determined according to the working width of the loading area.

3. The obstacle avoidance path planning method for mining trucks according to claim 1, characterized in that, Also includes: Using the sensing devices mounted on the current mining truck during entry and exit, the boundary of obstacles around the loading area is collected in real time to update the boundary of the drivable area.

4. The obstacle avoidance path planning method for mining trucks according to claim 1, characterized in that, The geometric dimensions of the enclosure of the obstacle mining card are greater than or equal to the security verification threshold.

5. The mining truck obstacle avoidance path planning method according to any one of claims 1 to 4, characterized in that, Based on the path search algorithm, within the search space, and constrained by the boundaries of the passable area, an obstacle avoidance path is generated, specifically including: Starting from the search start position in front of the current mining card, a discrete point sampling search is performed within the search space to obtain candidate points; The candidate points are subjected to security checks based on the boundary of the passable area, the bounding box of the obstacle mine card, and the bounding box of the current mine card, and the candidate points that pass the security checks are selected as intermediate points. Starting from the current position of the current mining card and ending at the intermediate point, a first path is generated using the path search algorithm. A second path is generated, starting from the midpoint and ending at the outlet of the loading area. The obstacle avoidance path is obtained by concatenating the first path and the second path.

6. The obstacle avoidance path planning method for mining trucks according to claim 5, characterized in that, Discrete point sampling search is performed within the search space, specifically including: Within the search space, discretization is performed along the width direction of the search space, and a sampling search is performed at each discrete location along the direction perpendicular to the width.

7. The obstacle avoidance path planning method for mining trucks according to claim 5, characterized in that, The security check specifically includes: Whether the candidate point is located inside the boundary of the drivable area, whether the bounding box of the current mining card corresponding to the candidate point collides with the boundary of the drivable area, and whether the bounding box of the current mining card corresponding to the candidate point overlaps with the bounding box of the obstacle mining card.

8. The obstacle avoidance path planning method for mining trucks according to claim 5, characterized in that, Starting from the midpoint and ending at the outlet of the loading area, a second path is generated, specifically including: A second path connecting the intermediate point and the outlet of the loading area is generated by using a hybrid A* algorithm or the Reeds-Shepp curve algorithm.

9. A mining truck obstacle avoidance path planning system, characterized in that, include: Mining Card; A roadside base station is connected to the mining truck for communication; the roadside base station is used to store the operation scene information of the loading area and can implement the steps of the mining truck obstacle avoidance path planning method as described in any one of claims 1 to 8 to generate an obstacle avoidance path.

10. The mining truck obstacle avoidance path planning system according to claim 9, characterized in that, It also includes cloud platforms; The cloud platform and the roadside base station are connected for communication, and the cloud platform is used to receive and store the operation data uploaded by the roadside base station.