Automatic driving path planning method for mining truck

Abstract

The invention relates to the technical field of automatic driving, and discloses a mining truck automatic driving path planning method comprising the following steps: obtaining mine environment data and vehicle operation data; performing data fusion on the mine environment data and the vehicle operation data to obtain environment sensing data; constructing a mine environment three-dimensional digital map according to the environment perception data; a virtual potential field environment is constructed on a three-dimensional digital map of a mine environment by using an artificial potential field method, and a driving path is dynamically planned based on a virtual force mechanism in combination with vehicle operation data. The obstacle avoidance capability and the path planning efficiency can be improved.

Description

Technical Field

[0001] The present invention relates to the technical field of autonomous driving, and particularly to a method for path planning of autonomous driving of mining trucks. Background Art

[0002] With the development of mining automation and intelligence, the autonomous driving technology of mining trucks has become a key way to improve mining operation efficiency, reduce labor costs, and enhance operation safety. Facing the challenges of complex and changeable mining environments, such as rough terrains, variable weather conditions, dense transportation lines, and dynamically changing operation areas, traditional path planning and decision-making algorithms often seem inadequate, suffering from problems such as low planning efficiency, unreasonable paths, and slow decision-making responses. Summary of the Invention

[0003] In view of the above deficiencies in the prior art, the present invention provides a method for path planning of autonomous driving of mining trucks.

[0004] To achieve the above invention objective, the technical solution adopted by the present invention is as follows: A method for path planning of autonomous driving of mining trucks, comprising the following steps: Obtain mining environment data and vehicle operation data; Perform data fusion on the mining environment data and the vehicle operation data to obtain environment perception data; Construct a three-dimensional digital map of the mining environment based on the environment perception data; Construct a virtual potential field environment on the three-dimensional digital map of the mining environment by using the artificial potential field method, and in combination with the vehicle operation data, dynamically plan the driving path based on the virtual force mechanism.

[0005] Preferably, obtaining the mining environment data and the vehicle operation data includes: Use a lidar to obtain point cloud data of the mining environment; Use a millimeter-wave radar to detect the speed and position of distant targets; Use a camera to obtain visual data of the mining environment; Use an inertial navigation system to obtain the attitude, speed, and position information of the vehicle.

[0006] Preferably, the three-dimensional digital map of the mining environment includes terrain, obstacles, road network, and operation area information.

[0007] Preferably, constructing a virtual potential field environment on the three-dimensional digital map of the mining environment by using the artificial potential field method includes: Set the stopping point of the mining truck as the gravitational source, and calculate the gravitational value of each position point according to the distance from the center of the gravitational source; Set the obstacles and restricted areas as repulsive force sources, and calculate the repulsive force value at each position point according to the distance from the center of the repulsive force source and the obstacle attributes; According to the resultant force of gravitational and repulsive forces acting on the mining truck at the current position, select the direction of the resultant force as the moving direction, and gradually update the position of the mining truck until it reaches the target point or the preset number of iterations is reached.

[0008] Preferably, when calculating the repulsive force value at each position point according to the distance from the center of the repulsive force source and the obstacle attributes, set a repulsive force attenuation factor to dynamically adjust the magnitude and direction of the repulsive force.

[0009] Preferably, setting the repulsive force attenuation factor to dynamically adjust the magnitude and direction of the repulsive force specifically means: , where, is the repulsive force, is the repulsive force gain coefficient, is the repulsive force attenuation factor, d is the distance between the mining truck and the obstacle, is the obstacle influence radius, is the obstacle position, is the speed of the mining truck, is the control parameter for adjusting the influence degree of the speed difference on the repulsive force direction, is the obstacle speed.

[0010] Preferably, when dynamically planning the driving path based on the virtual force mechanism, compare the resultant force of gravitational and repulsive forces acting on the mining truck at the current position with the preset resultant force threshold value, and generate a virtual thrust pointing to the target point according to the comparison result.

[0011] Preferably, generating a virtual thrust pointing to the target point according to the comparison result specifically means: , where, is the virtual thrust, is the virtual thrust gain coefficient, is the target point position, is the current position of the mining truck, is the resultant force of gravitational and repulsive forces acting on the mining truck at the current position, is the preset resultant force threshold, is the preset distance threshold, is the Euclidean norm.

[0012] The present invention has the following beneficial effects: (1) By optimizing the repulsive force field function, the mining truck can more accurately perceive the position and properties (such as size, shape, mobility, etc.) of surrounding obstacles, thereby dynamically adjusting the magnitude and direction of the repulsive force, effectively avoiding collisions with obstacles. This dynamic adjustment not only improves the flexibility of obstacle avoidance but also enhances the adaptability of the mining truck in complex environments.

[0013] (2) The improved APF algorithm adopted by the present invention introduces a repulsive force attenuation factor, making the repulsive force gradually weaken when far from the obstacle, avoiding unnecessary calculations in the area far from the obstacle, thereby improving the efficiency of path planning. At the same time, the algorithm can comprehensively consider multiple factors (such as distance, road conditions, traffic flow, etc.) and quickly select the optimal path to ensure that the mining truck can complete the transportation task efficiently and safely.

[0014] (3) The present invention integrates functions such as high-precision map positioning, automatic path planning, and decision-making, and can make intelligent decisions based on real-time perception data and preset algorithm logics. During the obstacle avoidance process, the algorithm can evaluate the feasibility and safety of different obstacle avoidance schemes in real time and select the optimal obstacle avoidance path to ensure the safe driving of the mining truck.

[0015] (4) By introducing the improved APF algorithm, the mining truck can perceive and avoid potential dangerous areas in advance during the automatic driving process, greatly reducing the risk of safety accidents caused by human operation errors or environmental mutations. At the same time, the real-time performance and accuracy of the algorithm also improve the ability of the mining truck to cope with emergencies, ensuring the safety of the operation process.

[0016] (5) The driverless mining truck of the present invention has higher working efficiency and lower labor costs compared with the manned mining truck. Through the optimization of the intelligent path planning and decision-making algorithm, the mining truck can independently complete the transportation task, reducing the dependence on drivers and lowering the labor cost. At the same time, the optimization of the algorithm also improves the transportation efficiency of the mining truck, further reducing the operation cost.

[0017] In summary, the intelligent path planning and decision-making method for the automatic driving of the mining truck adopting the improved APF technology of the present invention has achieved remarkable technical effects in aspects such as improving the obstacle avoidance ability, path planning efficiency, decision-making intelligence, operation safety, and reducing the operation cost. Brief Description of the Drawings

[0018] Figure 1 It is a schematic flow chart of a path planning method for the automatic driving of a mining truck. Detailed Embodiments

[0019] The following describes the specific embodiments of the present invention to facilitate those skilled in the art of this technology to understand the present invention. However, it should be clear that the present invention is not limited to the scope of the specific embodiments. For those of ordinary skill in the art of this technology, as long as various changes are within the spirit and scope of the present invention defined and determined by the appended claims, these changes are obvious, and all inventions and creations using the concept of the present invention are within the scope of protection.

[0020] As Figure 1 shown, a method for path planning of an autonomous driving mining truck provided by an embodiment of the present invention includes the following steps S1 to S4: S1. Obtain mine environment data and vehicle operation data; In an optional embodiment of the present invention, step S1 of obtaining mine environment data and vehicle operation data includes: Use lidar to obtain mine environment point cloud data; Use millimeter-wave radar to detect the speed and position of distant targets; Use a camera to obtain visual data of the mine environment; Use an inertial navigation system to obtain the attitude, speed, and position information of the vehicle.

[0021] In this embodiment, the original data of the mine environment is collected in real time through multi-source sensors, including terrain information, obstacle positions, speed, direction, and the attitude, speed, and position information of the vehicle itself. Specifically, through the high-precision point cloud data of lidar, the fine three-dimensional structure of the mine environment can be obtained; millimeter-wave radar is good at detecting the speed and position of distant targets under adverse weather conditions; the camera provides rich visual information, including color, texture, etc., which helps to identify specific obstacles or traffic signs; while the inertial navigation system provides the attitude, speed, and position information of the vehicle itself.

[0022] S2. Perform data fusion on the mine environment data and vehicle operation data to obtain environment perception data; In an optional embodiment of the present invention, in step S2, in the autonomous driving system of the mining truck, the environment perception ability is improved through multi-source sensor data fusion.

[0023] In this embodiment, preprocessing operations such as denoising and filtering are first performed on the multi-source sensor data to improve the data quality. Then, it is processed through data fusion algorithms, such as Kalman filtering, particle filtering, or deep learning methods, to eliminate redundant information, reduce noise interference, and fuse into a consistent and accurate environment perception result, significantly improving the robustness and safety of the autonomous driving mining truck in a complex mine environment.

[0024] S3. Construct a three-dimensional digital map of the mine environment according to the environment perception data; In an alternative embodiment of the present invention, the three-dimensional digital map of the mine environment constructed in step S3 includes terrain, obstacles, road networks, and operation area information.

[0025] Based on the data after multi-source sensor fusion, this embodiment constructs a three-dimensional digital map of the mine environment. This map not only contains high-precision terrain information such as slope and elevation, but also details various obstacles (such as rocks, trees, other vehicles), road networks (including width, curvature, material), operation area division, etc. Through advanced SLAM (Simultaneous Localization and Mapping) technology and GIS (Geographic Information System) technology, this embodiment can update the map information in real time to ensure the accuracy and timeliness of path planning. The high-precision environmental modeling provides a solid foundation for the path planning of autonomous mining trucks, enabling them to make accurate and efficient decisions in the complex and changing mine environment.

[0026] S4. Use the artificial potential field method to construct a virtual potential field environment on the three-dimensional digital map of the mine environment, and combine vehicle operation data to dynamically plan the driving path based on the virtual force mechanism.

[0027] In an alternative embodiment of the present invention, step S4 using the artificial potential field method to construct a virtual potential field environment on the three-dimensional digital map of the mine environment includes: Set the docking point of the mining truck as the gravitational source, and calculate the gravitational value of each position point according to the distance from the center of the gravitational source; Set obstacles and restricted areas as repulsive sources, and calculate the repulsive value of each position point according to the distance from the center of the repulsive source and the obstacle attributes; According to the resultant force of gravity and repulsion received by the mining truck at the current position, select the direction of the resultant force as the moving direction, and gradually update the position of the mining truck until it reaches the target point or reaches the preset number of iterations.

[0028] This embodiment adopts the optimized APF (Artificial Potential Field) algorithm as the core path search and optimization strategy for the changing traffic conditions and operation requirements in the mine environment. The APF algorithm constructs a virtual potential field environment, sets the target point as the center of gravity to attract the autonomous mining truck to move towards it; at the same time, regards obstacles, restricted areas, and other factors that may affect driving as repulsive sources to repel the mining truck to avoid collisions and bad road conditions. The optimized APF algorithm provides an efficient, flexible and safe path planning solution for autonomous mining trucks in mines through strategies such as dynamic repulsive field construction, integration of gravitational field and task priority, path smoothing and safety verification, and vehicle state and environmental adaptability adjustment. This dynamic path optimization strategy enables the mining truck to flexibly handle various challenges in the mine environment and ensure the smooth completion of transportation tasks.

[0029] In this embodiment, to more efficiently implement path planning and obstacle avoidance for mining trucks, the present invention adopts an improved artificial potential field (APF) algorithm. In this algorithm, the target point (such as the ore loading point or unloading point) is set as the gravitational center, and the gravitational value of each position point is calculated according to the distance, and the gravitational value decreases as the distance increases. Obstacles, restricted areas, etc. are regarded as repulsive force sources, and the repulsive force value of each position point is calculated according to the distance and obstacle attributes (such as size, shape, mobility), and the repulsive force value increases as the distance decreases to ensure that the mining truck can avoid these areas. In the constructed potential field, the autonomous driving mining truck selects the direction of the resultant force as the moving direction according to the resultant force of the gravitational and repulsive forces received at the current position. Through iterative calculation, the position of the mining truck is gradually updated until it reaches the target point or reaches the preset number of iterations.

[0030] To avoid the repulsive force and gravitational force canceling each other out in some cases, resulting in the resultant force of the mining truck being zero and the truck stopping moving, the present invention optimizes the repulsive force field function. Specifically, according to the relative position, speed of the obstacle and the mining truck, and the properties of the obstacle (such as size, shape, mobility, etc.), the magnitude and direction of the repulsive force are dynamically adjusted. At the same time, a repulsive force attenuation factor is set to ensure that the repulsive force gradually weakens when moving away from the obstacle, avoiding unnecessary influence on the path planning far from the obstacle. Specifically: , where, is the repulsive force, is the repulsive force gain coefficient, is the repulsive force attenuation factor, d is the distance between the mining truck and the obstacle, is the obstacle influence radius, is the obstacle position, is the speed of the mining truck, is the control parameter for adjusting the influence degree of the speed difference on the repulsive force direction. When it is equal to 0, the speed at time 0 is only related to the position difference. When it is greater than 0, it is considered that the repulsive force is affected by the relative speed between the obstacle and the mining truck, so that the mining truck can avoid the approaching obstacle; is the obstacle speed.

[0031] The traditional APF algorithm is prone to falling into local minima, that is, the resultant force on the mining truck at a certain position is zero, resulting in the inability to continue moving towards the target point. To solve this problem, the present invention introduces a virtual force mechanism. When it is detected that the mining truck is in an area where it may fall into local minima (such as the resultant force is less than the preset threshold and lasts for a period of time), the system generates a virtual thrust pointing to the target point to help the mining truck get out of the local trap and continue to move forward towards the target. Specifically: , where, is the virtual thrust, is the virtual thrust gain coefficient, is the target point position, is the current position of the mining truck, is the resultant force of gravitational and repulsive forces acting on the mining truck at the current position, is the preset resultant force threshold for detecting local minima; is the preset distance threshold for preventing the mining truck from being still affected by virtual forces when approaching the target point; is the Euclidean norm.

[0032] Through the above improvements, the present invention not only retains the high efficiency and intuitiveness of the APF algorithm in path planning, but also effectively solves the problems of local minima and the stagnation problem caused by the mutual cancellation of repulsive and attractive forces, enabling the mining truck to achieve more stable and reliable autonomous driving in complex mine environments.

[0033] To further improve the intelligent level of autonomous driving mining trucks, we integrate machine learning or deep learning models as the core of the intelligent decision-making module. This module uses big data analysis technology to learn and train a large amount of historical data to master the laws and patterns of various events occurring in the mine environment. During the path planning process, the intelligent decision-making module can intelligently evaluate and adjust the planning results, such as predicting potential risk areas and evaluating the feasibility of different paths. At the same time, when encountering sudden situations (such as the sudden appearance of obstacles, road congestion, etc.), the intelligent decision-making module can respond quickly, make safe and reasonable decisions after comprehensively analyzing various factors, and ensure the safe driving of the autonomous driving mining truck. This intelligent decision-making ability not only improves the flexibility and adaptability of the autonomous driving system, but also greatly reduces the necessity of human intervention.

[0034] 1. Data collection and processing • Data sources: The intelligent decision-making module relies on a variety of data sources, including but not limited to vehicle sensor data (such as radar, camera, lidar, etc.), mine environment data (such as topographic maps, meteorological data), historical driving records, and human intervention records, etc.

[0035] • Data preprocessing: Clean, denoise, normalize and other processing on the collected raw data to improve data quality and provide reliable input for subsequent learning and training.

[0036] 2. Machine learning / deep learning model construction • Model selection: According to the specific requirements of mine autonomous driving, select appropriate machine learning or deep learning models, such as convolutional neural network (CNN), recurrent neural network (RNN), long short-term memory network (LSTM), reinforcement learning models, etc.

[0037] • Feature Engineering: Extract useful features for intelligent decision-making from the preprocessed data, such as obstacle distance, speed change, road type, weather conditions, etc.

[0038] • Model Training: Utilize big data analysis techniques to learn and train a large amount of historical data, enabling the model to master the laws and patterns of various events occurring in the mine environment.

[0039] 3. Intelligent Evaluation and Adjustment • Path Planning Evaluation: During the path planning process, the intelligent decision-making module uses the trained model to intelligently evaluate the planning results, predict potential risk areas, and evaluate the feasibility, safety, and efficiency of different paths.

[0040] • Dynamic Adjustment: Based on the evaluation results, the intelligent decision-making module can dynamically adjust the path planning strategy, such as selecting a safer and more efficient path, or taking obstacle avoidance measures when necessary.

[0041] 4. Emergency Handling • Real-time Monitoring: The intelligent decision-making module can real-time monitor the vehicle status and changes in the mine environment, including sudden obstacles, road blockages, weather mutations, etc.

[0042] • Comprehensive Analysis: When encountering an emergency, the intelligent decision-making module can quickly respond, comprehensively analyze the vehicle status, environmental data, and historical experience, and make safe and reasonable decisions.

[0043] • Decision Execution: Based on the decision results, the intelligent decision-making module can control the vehicle to perform corresponding operations, such as emergency braking, bypassing obstacles, adjusting the driving speed, etc., to ensure the safe driving of the autonomous mining truck.

[0044] 5. Continuous Optimization and Iteration • Feedback Mechanism: The intelligent decision-making module has the ability of self-learning and optimization, and can continuously adjust the model parameters according to the feedback data during the actual driving process (such as human intervention records, driving effect evaluations, etc.) to improve the accuracy and reliability of the decision-making.

[0045] • Iterative Update: With the continuous development of the mine environment and autonomous driving technology, the intelligent decision-making module needs to be updated and iterated regularly to adapt to the new environment and requirements.

[0046] To ensure the safety and controllability of the autonomous mining truck system, we have designed a user-friendly human-machine interface and a complete monitoring system. The human-machine interface provides an intuitive operation interface and rich information display functions, allowing remote operators to monitor the running status of the mining truck, the results of environmental perception, and the path planning situation in real time. At the same time, the interface also supports remote control and parameter adjustment functions for manual intervention or system performance optimization when necessary. The monitoring system comprehensively monitors the running status of the mining truck and the surrounding environment through real-time monitoring and data analysis technologies of multi-source sensors. Once an abnormal situation or potential risk is detected, the system will immediately issue an alarm and activate the emergency response mechanism. This design of human-machine interaction and monitoring not only improves the safety and reliability of the system but also enhances the operator's trust and control over the system.

[0047] This invention is described with reference to the flowcharts and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It should be understood that each flow and / or block in the flowchart and / or block diagram, and the combinations of flows and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to produce a machine such that the instructions executed by the processor of the computer or other programmable data processing devices generate means for implementing the functions specified in Figure 1 one or more of the flows Figure 1 or blocks or combinations of blocks.

[0048] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, such that the instructions stored in the computer-readable memory produce a manufacture including instruction means that implement the functions specified in Figure 1 one or more of the flows Figure 1 or blocks or combinations of blocks.

[0049] These computer program instructions can also be loaded onto a computer or other programmable data processing device, such that a series of operation steps are executed on the computer or other programmable device to produce a computer-implemented process, and thus the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in Figure 1 one or more of the flows Figure 1 or blocks or combinations of blocks.

[0050] In the present invention, specific embodiments are used to elaborate on the principles and implementation manners of the present invention. The description of the above embodiments is only for helping to understand the method and its core idea of the present invention; at the same time, for those of ordinary skill in the art, according to the idea of the present invention, there will be changes in the specific implementation manners and application scopes. In summary, the content of this specification should not be construed as a limitation to the present invention.

[0051] Those of ordinary skill in the art will realize that the embodiments described herein are for helping the reader understand the principles of the present invention, and it should be understood that the protection scope of the present invention is not limited to such specific statements and embodiments. Those of ordinary skill in the art can make various other specific deformations and combinations without departing from the essence of the present invention according to these technical revelations disclosed by the present invention, and these deformations and combinations are still within the protection scope of the present invention.

Claims

1. A path planning method for autonomous driving of mining trucks, characterized in that, It includes the following steps: Obtain mine environment data and vehicle operation data; Perform data fusion on the mine environment data and vehicle operation data to obtain environment perception data; Construct a three-dimensional digital map of the mine environment based on the environment perception data; Construct a virtual potential field environment on the three-dimensional digital map of the mine environment, and combine with the vehicle operation data to dynamically plan the driving path based on the virtual force mechanism.

2. The method for path planning of autonomous driving of a mining truck according to claim 1, wherein Obtaining mine environment data and vehicle operation data includes: Use lidar to obtain mine environment point cloud data; Use millimeter-wave radar to detect the speed and position of distant targets; Use cameras to obtain visual data of the mine environment; Use an inertial navigation system to obtain the attitude, speed, and position information of the vehicle.

3. A method for path planning of autonomous driving of mining trucks according to claim 1, characterized in that, The three-dimensional digital map of the mine environment includes terrain, obstacles, road network, and work area information.

4. A method for path planning of an autonomous driving of a mining truck according to claim 1, characterized in that, Constructing a virtual potential field environment on the three-dimensional digital map of the mine environment using the artificial potential field method includes: Set the docking point of the mining truck as the gravitational source, and calculate the gravitational value of each position point according to the distance from the center of the gravitational source; Set obstacles and restricted areas as repulsive sources, and calculate the repulsive value of each position point according to the distance from the center of the repulsive source and the obstacle attributes; According to the resultant force of gravity and repulsion received by the mining truck at the current position, select the direction of the resultant force as the moving direction, and gradually update the position of the mining truck until it reaches the target point or the preset number of iterations is reached.

5. A method for autonomous driving path planning of a mining truck according to claim 1, characterized in that, When calculating the repulsive value of each position point according to the distance from the center of the repulsive source and the obstacle attributes, set a repulsive decay factor to dynamically adjust the magnitude and direction of the repulsion.

6. A method for path planning of autonomous driving of mining trucks according to claim 5, characterized in that, Specifically, setting the repulsive decay factor to dynamically adjust the magnitude and direction of the repulsion is: ， Among them, is the repulsive force, is the repulsive force gain coefficient, is the repulsive force attenuation factor, d is the distance between the mining truck and the obstacle, is the influence radius of the obstacle, is the position of the obstacle, is the speed of the mining truck, is the control parameter for adjusting the influence degree of the speed difference on the direction of the repulsive force, is the speed of the obstacle.

7. A method for autonomous driving path planning of a mining truck according to claim 1, characterized in that, When dynamically planning the driving path based on the virtual force mechanism, compare the resultant force of gravity and repulsion received by the mining truck at the current position with the preset resultant force threshold, and generate a virtual thrust pointing to the target point according to the comparison result.

8. A method for path planning of autonomous driving of mining trucks according to claim 7, characterized in that, Specifically, generating a virtual thrust pointing to the target point according to the comparison result is: ， Among them, is the virtual thrust, is the virtual thrust gain coefficient, is the target point position, is the current position of the mining truck, is the resultant force of gravitational and repulsive forces acting on the mining truck at the current position, is the preset resultant force threshold, is the preset distance threshold, is the Euclidean norm.

Images