Path planning method, vehicle and electronic equipment

Abstract

The invention discloses a path planning method, a vehicle and electronic equipment. The method comprises the following steps: acquiring entrance and exit information of a working area and spatial orientation information of a plurality of working positions; generating a plurality of candidate paths based on the entrance and exit information and the spatial orientation information; determining a first type of trunk road based on a first type of candidate lanes in the plurality of candidate paths, and determining a second type of trunk road based on a second type of candidate lanes in the plurality of candidate paths; combining at least part of the first type of candidate lanes in each candidate path into a first type of main road, combining at least part of the second type of candidate lanes in each candidate path into a second type of main road, and generating a target path; and sending the target path to the target vehicle. According to the invention, the technical problem of poor path planning effect in related technologies is solved.

Description

Technical Field

[0001] The present invention relates to the fields of vehicles and path planning, and in particular, to a path planning method, a vehicle, and an electronic device. Background Art

[0002] With the development of intelligent driving technology, path planning is the key to ensuring the efficient and safe operation of vehicles. Especially in industrial automation environments such as open-pit mines, in the open-pit mine operation area, not only are there a large number of vehicles, but also the space is limited and the obstacles are complex. Vehicles need to perform operation operations frequently, which puts higher requirements on the real-time performance, accuracy, and safety of path planning.

[0003] In related technologies, path planning decisions are made through mutual communication between vehicles to prompt vehicles to avoid or detour in a timely manner. However, this planning method has limited ability to handle complex driving scenarios, and as the number of vehicles increases, path intersections are likely to occur, making it difficult for vehicles to drive along the planned path, and it will also lead to congestion, insecurity and other situations, thus there are defects in the poor path planning effect.

[0004] In view of the above problems, no effective solution has been proposed yet. Summary of the Invention

[0005] Embodiments of the present invention provide a path planning method, a vehicle, and an electronic device to at least solve the technical problem of poor path planning effect in related technologies.

[0006] According to one aspect of the embodiments of the present invention, a path planning method is provided, including: obtaining entrance and exit information of an operation area, and spatial orientation information of a plurality of operation positions, where the entrance and exit information includes an entrance position and an exit position of the operation area; generating a plurality of candidate paths based on the entrance and exit information and the spatial orientation information, where each candidate path includes a first type of candidate lane for connecting the corresponding operation position to the entrance position and a second type of candidate lane for connecting the corresponding operation position to the exit position; determining a first type of main road based on the first type of candidate lanes in the plurality of candidate paths, and determining a second type of main road based on the second type of candidate lanes in the plurality of candidate paths; incorporating at least a part of the first type of candidate lanes in each candidate path into the first type of main road, incorporating at least a part of the second type of candidate lanes in each candidate path into the second type of main road, and generating a target path; sending the target path to a target vehicle.

[0007] According to another aspect of the embodiments of the present invention, there is also provided a path planning method, including: receiving a target path, where the target path is generated by incorporating at least part of the first-type candidate lanes in each of multiple candidate paths into the first-type main road, and incorporating at least part of the second-type candidate lanes in each of the candidate paths into the second-type main road. The first-type main road is determined based on the first-type candidate lanes in the multiple candidate paths, and the second-type main road is determined based on the second-type candidate lanes in the multiple candidate paths. The multiple candidate paths are generated based on the entrance and exit information of the operation area and the spatial orientation information of multiple operation positions. The entrance and exit information includes the entrance position and the exit position of the operation area. Each candidate path includes a first-type candidate lane for connecting the corresponding operation position to the entrance position and a second-type candidate lane for connecting the corresponding operation position to the exit position; driving according to the target path.

[0008] According to another aspect of the embodiments of the present invention, there is also provided a vehicle, including: a communication unit for communicating with a cloud server or a target vehicle; a memory storing an executable program; and a processor for running the program, where when the program runs, it executes the methods in the various embodiments of the present invention.

[0009] According to another aspect of the embodiments of the present invention, there is also provided an electronic device, including: a communication unit for communicating with a target vehicle; a memory storing an executable program; and a processor for running the program, where when the program runs, it executes the methods in the various embodiments of the present invention.

[0010] According to another aspect of the embodiments of the present invention, there is also provided a computer-readable storage medium, which includes a stored executable program, where when the executable program runs, it controls the device where the computer-readable storage medium is located to execute the methods in the various embodiments of the present invention.

[0011] According to another aspect of the embodiments of the present invention, there is also provided a computer program product, including a computer program, where when the computer program is executed by a processor, it implements the methods in the various embodiments of the present invention.

[0012] According to another aspect of the embodiments of the present invention, there is also provided a computer program product, including a non-volatile computer-readable storage medium, the non-volatile computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, it implements the methods in the various embodiments of the present invention.

[0013] According to another aspect of the embodiments of the present invention, there is also provided a computer program, where when the computer program is executed by a processor, it implements the methods in the various embodiments of the present invention.

[0014] In an embodiment of the present invention, entrance and exit information of a work area and spatial orientation information of multiple work positions are obtained; based on the entrance and exit information and the spatial orientation information, multiple candidate paths are generated; based on the first type of candidate lanes in the multiple candidate paths, a first type of main road is determined, and based on the second type of candidate lanes in the multiple candidate paths, a second type of main road is determined; at least part of the first type of candidate lanes in each candidate path is incorporated into the first type of main road, and at least part of the second type of candidate lanes in each candidate path is incorporated into the second type of main road to generate a target path; the target path is sent to a target vehicle. It is easy to notice that two types of candidate lanes are generated through the entrance and exit information and the spatial orientation information to cover the complete paths from the entrance to the work positions and from the work positions back to the exit. And in the scenario of multiple work positions, a path planning method of incorporating main roads and lanes is introduced. By planning and defining a main road, the paths can be planned with the main road as the core, and then at least part of the candidate lanes is incorporated into the main road to generate the target path, achieving the purpose of orderly planning the paths, thereby realizing the technical effect of improving the driving effect of path planning, and further solving the technical problem of poor path planning effect in the related art. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of this application. The schematic embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation to the present invention. In the drawings:

[0016] Figure 1 is a flowchart of a path planning method according to an embodiment of the present invention;

[0017] Figure 2 is a schematic diagram of an optional driving according to a planned path according to an embodiment of the present invention;

[0018] Figure 3 is a flowchart of an optional path planning method according to an embodiment of the present invention;

[0019] Figure 4 is a schematic diagram of an optional path planning device according to an embodiment of the present invention;

[0020] Figure 5 is a schematic diagram of an optional path planning device according to an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0021] To enable those skilled in the art to better understand the solution of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0022] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged under appropriate circumstances so that the embodiments of the present invention described herein can be implemented in an order different from those illustrated or described herein. In addition, the terms "comprising" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0023] According to an embodiment of the present invention, a method embodiment of a path planning method is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although the logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that here.

[0024] Figure 1 is a flowchart of a path planning method according to an embodiment of the present invention, as Figure 1 shown, the method includes the following steps:

[0025] Step S102, obtain the entrance and exit information of the operation area, and the spatial orientation information of multiple operation positions.

[0026] Among them, the entrance and exit information includes the entrance position and the exit position of the operation area.

[0027] The above-mentioned operation area can be a loading area for loading operations. The operation area can be in an open-pit mine or in a tunnel. The location of the operation area is not limited here and can be determined according to needs. The operation area can include multiple loading points so that vehicles such as haul trucks can leave the operation area for transportation activities after completing loading at the loading points. The operation area can be in the shape of a rectangle, an irregular polygon, etc. The shape of the operation area is not limited and can be determined according to needs. This specifically depends on the terrain of the mine and the operation layout. The operation area can also be divided into multiple sub-areas, and each sub-area corresponds to one or more excavators for operation. The operation area defines the boundaries and scope of the operation, helps to understand the specific environment of vehicle path planning and multi-vehicle collaborative operation, and provides a clear geographical space framework for the path planning method.

[0028] The above-mentioned operation position can be a specific position for loading operations within the operation area. In a multi-excavator scenario, each excavator has one or more corresponding operation positions for separate operations without interference. Vehicles such as haul trucks need to accurately approach these operation positions for loading. The operation position can be fixed, that is, the excavator position is fixed; the operation position can also be dynamic, that is, the excavator moves within the operation area according to operation requirements. In this case, vehicles such as haul trucks need to track the excavator position in real time. The operation position can be defined by position coordinates (X, Y, Z) and orientation angle (θ). The operation position can also be determined by pre-configured codes. The operation position is the end point and starting point of path planning, ensuring that the haul truck can efficiently and safely reach the loading position for loading. After loading is completed, it can leave along an adapted path to avoid conflicts with other haul trucks or equipment.

[0029] The above-mentioned entrance and exit information can include the entrance position and exit position of the operation area. The entrance and exit information can be the channels for haul trucks to enter and leave the operation area. The entrance and exit can be one-way, that is, the entrance and exit are separated, and vehicles such as haul trucks can only enter from the entrance and leave from the exit; the entrance and exit can also be two-way, that is, the same position can be used as both the entrance and the exit. The entrance and exit information is crucial for path planning. It determines the initial path direction and the final path direction of vehicles such as haul trucks, helps to design a conflict-free traffic flow line, and improves the operation efficiency and safety of the operation area.

[0030] The above spatial orientation information may refer to the relative position and orientation angle of the operation area. The spatial orientation information may include, but is not limited to, position, orientation, lane layout, obstacle distribution, etc. The content of the spatial orientation information is not limited here and can be determined according to needs. The spatial orientation information can be static, such as a fixed road layout; the spatial orientation information can also be dynamic, such as the real-time position information and relative angle of the mining truck and the excavator. The spatial orientation information is the basis for realizing the collaborative path planning of the mining truck, which helps to understand the relative position relationship of each element in the operation area and design a path to avoid collisions and improve efficiency. In addition, the dynamic spatial orientation information also supports the real-time scheduling and path adjustment between the mining truck and the excavator to cope with changes in the operation environment.

[0031] In an alternative embodiment, the entrance and exit information of the operation area and the spatial orientation information of multiple operation positions can be obtained by using the Global Positioning System (GPS for short) and lidar. A GPS receiver can be equipped on the vehicle to record its position information in real time. At the same time, GPS positioning reference points are set at the entrances and exits of the operation area and at each operation position to determine the precise coordinates of the entrances and exits and the operation positions. Then, the lidar of the vehicle can periodically scan the surrounding environment, including the environment at the entrances and exits and the terrain features near the operation positions. By analyzing the lidar data, a three-dimensional map of the operation area can be constructed, so as to determine the exact spatial relationship between the entrances and exits and the operation positions. Furthermore, the GPS data and the terrain feature data scanned by the lidar can be input into a computing platform for fusion processing to generate a digital map of the operation area, in which the entrance and exit information and the spatial orientation information of each operation position are clearly marked.

[0032] In another alternative embodiment, the entrance and exit information and the spatial orientation information of multiple operation positions can also be obtained from the data recorded by the unmanned aerial vehicle. The unmanned aerial vehicle is equipped with a camera and can take pictures over the operation area according to a pre-set flight route, capturing images of every corner of the operation area, especially near the entrances and exits and the operation positions. The images obtained by the unmanned aerial vehicle are processed by image recognition software to automatically identify the specific positions of the entrances and exits and the spatial orientation information of each operation position. In addition, through in-depth analysis of the images, detailed data on the terrain of the operation area can be obtained. The image data recorded by the unmanned aerial vehicle and the analysis results can be integrated into a database and transmitted to the autonomous driving system of the vehicle in real time, so as to consider the actual layout of the entrances and exits and the operation positions during path planning.

[0033] In yet another alternative embodiment, the entrance and exit information and the spatial orientation information of multiple working positions can be obtained from a pre-planned and configured data information management system. Satellite remote sensing and unmanned aerial vehicle (UAV) aerial photography technologies can be used to obtain a three-dimensional terrain model of the working area. The three-dimensional terrain model includes, but is not limited to, information such as surface undulations, vegetation cover, and geological structures. Then, in combination with topographic survey data, the positions of the entrances and exits within the working area are planned to ensure the traffic capacity and safety of the entrances and exits. At the same time, multiple working positions are reasonably arranged. At the planned positions of the entrances and exits and working positions, positioning markers such as radio frequency identification (RFID) tags, two-dimensional codes, GPS beacons, etc. are installed, and these positioning markers can provide accurate geographic coordinate information. Then, these positioning markers and the corresponding information are entered into the data information management system for future reference when needed. However, considering that the working area may change due to reasons such as temporary operation requirements, equipment maintenance, and weather impacts, the data information in the data information management system can be updated by uploading it through the handheld devices of on-site staff, or by using automated devices (such as UAVs) for real-time scanning and confirmation, and automatically synchronized to the data information management system.

[0034] Step S104, based on the entrance and exit information and the spatial orientation information, generate multiple candidate paths.

[0035] Among them, each candidate path includes a first type of candidate lane for connecting the corresponding working position to the entrance position and a second type of candidate lane for connecting the corresponding working position to the exit position.

[0036] The above-mentioned candidate paths can include the first type of candidate lane and the second type of candidate lane. The candidate paths can be possible driving route options planned for vehicles in an autonomous driving or intelligent transportation system. In the collaborative path planning of mining trucks in an open-pit mine loading area, the candidate paths can include the empty and heavy-load lanes corresponding to the excavators. The generation and function of the candidate paths are particularly crucial, as they not only affect the operating efficiency of the mining trucks but also directly relate to the safety of the working area. The candidate paths can be determined based on the entrance and exit information and the spatial orientation information to guide the vehicle from the entrance to the working position or from the working position to the exit.

[0037] The above-mentioned first type of candidate lanes can be used to connect the working positions with the entrance location. The first type of candidate lanes may have different attributes, such as lane width, slope, turning radius, load-bearing capacity, etc. The attributes of the first type of candidate lanes are not limited here and can be determined according to needs. These factors will vary according to the specific terrain and transportation requirements of the working area. In addition, according to the layout and quantity of the working positions, there can be various lane types such as straight, circular, branched, etc. The lane types are not limited here and can be determined according to needs. By generating the first type of candidate lanes, the time and energy consumption for vehicles to enter the working positions can be reduced, the ineffective driving of vehicles during the process of searching for working positions can be avoided, and at the same time, the driving safety of vehicles when entering the working area can be ensured, reducing the possibility of accidents. The generation of the first type of candidate lanes usually needs to consider factors such as the entrance information, the spatial orientation information of the working positions, as well as the minimum turning radius, sufficient safety distance, etc., and finally form the lane layout for entering the working positions. In addition, it is also necessary to verify the actual usability of the lanes through on-site surveys to ensure consistency with the planning model.

[0038] The above-mentioned second type of candidate lanes can be used to connect the working positions with the exit location. In contrast to the first type of lanes, the second type of candidate lanes can be used for path planning when vehicles leave the working area after completing their operations. The types of the second type of candidate lanes need to consider the load status of vehicles when leaving the working area, the exit distribution of the working area, and the possible traffic flow to ensure unobstructed paths for vehicles such as mining trucks to leave the working positions, avoiding congestion and collisions. By generating the second type of candidate lanes, the evacuation efficiency of mining trucks after completing their operations can be improved, the waiting and staying time can be reduced, and at the same time, the orderliness and safety of the traffic at the exits of the working area can be ensured, avoiding the tail effect, that is, the congestion at the exits caused by mining trucks staying in the working area for a long time after completing their operations. Similar to the first type of candidate lanes, the second type of candidate lanes can be determined based on the exit information and spatial orientation information. It is designed through traffic engineering software and adjusted in combination with the actual on-site situation. It should be particularly noted that since mining trucks may be in a fully loaded state when leaving the working positions, the lane design also needs to additionally consider the impact of the change in vehicle center of gravity on driving stability to ensure the bearing capacity and safety of the road.

[0039] In an alternative embodiment, the working area is regarded as a graph, where the nodes represent key locations such as entrances, exits, and working positions, the edges represent the paths connecting these locations, and specific weights (such as distance, required time, energy consumption, etc.) are assigned to each edge. Graph search algorithms, such as Dijkstra's algorithm or depth-first search, are used to explore the paths from any entrance or exit to one or more working positions. The principle of Dijkstra's algorithm is to find the shortest paths from a vertex to the remaining vertices in a weighted graph.

[0040] In another alternative embodiment, a machine learning model, such as a neural network or reinforcement learning, can be utilized and trained to identify the spatial orientations of the entrances and working positions and predict multiple possible candidate paths. The machine learning model can be trained through a large amount of historical driving data and path planning so that the machine learning model can learn effective paths. By learning historical behavior patterns and environmental features, the machine learning model can generate candidate paths that better meet the actual requirements. Through reinforcement learning, the path selection strategy can be continuously adjusted to adapt to the dynamic changes in the environment, such as changes in traffic flow or temporary adjustments to working positions.

[0041] In yet another alternative embodiment, virtual reality or augmented reality technology can be combined with the entrance and exit information and spatial orientation information of the physical world to create a virtual working area model for real-time generation and visualization of multiple candidate paths. This enables users to intuitively see the effects of each candidate path in the virtual environment and make adjustments or selections according to actual needs.

[0042] Step S106: Based on the first type of candidate lanes among the multiple candidate paths, determine the first type of main road, and based on the second type of candidate lanes among the multiple candidate paths, determine the second type of main road.

[0043] The above-mentioned first type of main road can be used to guide the vehicle to drive in the direction from the entrance to the working position. The first type of main road can be the necessary road for the vehicle to enter the working position. Specifically, in practical applications, the first type of main road can be the empty-load lane in front of the excavator reversing lane. It is usually the common section from the entrance to multiple working positions. By means of the first type of main road, a standardized driving path can be provided, reducing the uncertainty when the empty-load mining truck enters the working area to find a working position, thereby shortening the time to reach the working position; at the same time, by planning the first type of main road, potential traffic conflicts can be avoided, enhancing traffic safety within the working area.

[0044] The above-mentioned second type of main road can be used to guide the vehicle to drive in the direction from the working position to the exit. The second type of main road can be the necessary road for the vehicle to leave the working position. Specifically, in practical applications, the second type of main road can be the heavy-load lane in front of the excavator reversing lane. It is usually the common section from multiple working positions to the exit. By means of the second type of main road, a standardized driving path can be provided, reducing the uncertainty when the heavy-load mining truck leaves the working position, thereby shortening the departure time; at the same time, by planning the second type of main road, potential traffic conflicts can be avoided, enhancing traffic safety within the working area.

[0045] In an alternative embodiment, a first type of arterial road is determined based on the first type of candidate lanes among multiple candidate paths through a first preset rule. The first preset rule can be used to evaluate the priority of lanes. The preset rule can be set manually. The first preset rule, such as lane length, preferentially selecting straight sections, avoiding continuous turns, ensuring lane width, etc., is used to determine the first type of arterial road. Also, a second type of arterial road can be determined based on the second type of candidate lanes among multiple candidate paths through a second preset rule. The second preset rule, such as lane length, preferentially selecting straight sections, avoiding continuous turns, ensuring lane width, etc., is used to determine the second type of arterial road.

[0046] In another alternative embodiment, clustering analysis can be performed on all the first type of candidate lanes to identify the parts that are frequently shared by multiple paths. Based on these shared parts, a central point or a central path is determined as the first type of arterial road. For the first type of candidate lanes, clustering algorithms can be used for grouping to identify the sections with high usage frequency. From the clustering results, one or several central points are determined, and these points are connected to form the first type of candidate lanes. Also, clustering analysis can be performed on all the second type of candidate lanes to identify the parts that are frequently shared by multiple paths. Based on these shared parts, a central point or a central path is determined as the second type of arterial road. For the second type of candidate lanes, clustering algorithms can be used for grouping to identify the sections with high usage frequency. From the clustering results, one or several central points are determined, and these points are connected to form the second type of candidate lanes.

[0047] In yet another alternative embodiment, experts can determine the first type of arterial road from the first type of candidate lanes and the second type of arterial road from the second type of candidate lanes according to their experience. Expert planning can comprehensively apply professional knowledge in fields such as mine operation, safety management, and vehicle engineering to improve the overall quality of path planning. Facing the constantly changing mine operation environment, expert planning can respond in a timely manner, adjust the path planning strategy to address new challenges. Expert planning can also identify potential safety risks to take preventive measures during path planning and reduce the accident rate.

[0048] Step S108, at least part of the first type of candidate lanes in each candidate path is incorporated into the first type of arterial road, and at least part of the second type of candidate lanes in each candidate path is incorporated into the second type of arterial road to generate a target path.

[0049] The above-mentioned target path can refer to the driving route determined for the target vehicle in a multi-lane and multi-operation position layout. This path aims to ensure that the vehicle can efficiently and safely reach the operation position from the entrance or return to the exit from the operation position. By avoiding congestion, obstacles, and adjusting the turning radius, etc., the target path shortens the driving time and energy consumption of the vehicle and improves the operation efficiency. By generating the target path, it ensures that the vehicle follows safety rules during movement and reduces the risk of accidents. In a multi-vehicle operation environment, the target path helps to avoid the head-on phenomenon, that is, the path conflict between vehicles, and ensures smooth traffic flow.

[0050] In an alternative embodiment, artificial intelligence technologies such as deep learning or genetic algorithms can be used to intelligently analyze the first type of candidate lanes, automatically merge into the first type of main road, and intelligently analyze the second type of candidate lanes, automatically merge into the second type of main road to generate the target path. Specifically, use historical driving data, lane geometric information, and vehicle performance data to train a deep learning model to make it understand the impact of different candidate lanes on the target path. Analyze the candidate lanes through the deep learning model, predict the performance indicators (such as overall driving time, energy consumption) of the merged main road, so as to determine which first type of candidate lane in each candidate path should be merged into the first type of main road, and determine which second type of candidate lane in each candidate path should be merged into the second type of main road. Verify the lane merging scheme recommended by the deep learning model in a virtual simulation environment, and make manual adjustments if necessary to finally determine the target path.

[0051] In another alternative embodiment, a multi-objective algorithm can be adopted to transform the problem of lane merging into a mathematical problem containing multiple objectives (such as time, energy consumption, safety) to solve and generate the target path. For the first type of candidate lanes, generate a feature matrix containing its attributes (length, width, slope, turning radius, etc.) and their corresponding performance indicators (such as driving time, expected energy consumption). Establish a multi-objective function: set multiple objectives, such as minimizing the total driving time, reducing energy consumption, maximizing safety, etc., and convert these objectives into mathematical expressions. Apply the multi-objective algorithm to solve the first type of candidate lane scheme that can simultaneously meet or balance multiple objectives. Similarly, solve the second type of candidate lane scheme that can simultaneously meet or balance multiple objectives, and then combine the first type of candidate lane scheme and the second type of candidate lane scheme to obtain the target path.

[0052] In yet another alternative embodiment, the first type of candidate lanes and the second type of candidate lanes can be dynamically evaluated and merged according to real-time traffic flow and road conditions through an adaptive lane allocation and merging algorithm to generate a target path. Traffic flow, vehicle speed, road conditions and other information on each first type of candidate lane and second type of candidate lane can be collected in real time through on-vehicle sensors and vehicle-to-vehicle (V2V) communication. Based on the real-time information, each first type of candidate lane and second type of candidate lane is scored, considering its current driving efficiency, safety and contribution to the overall operation process. Then, according to the lane scores and vehicle types (empty or full load), the merging of the first type of candidate lanes and the second type of candidate lanes is dynamically selected to determine the target path.

[0053] Step S110, send the target path to the target vehicle.

[0054] The above-mentioned target vehicle can be a means of transportation that travels along the target path to complete an operation. In the cloud platform application scenario, the target vehicle can be all vehicles that execute operation tasks, the target vehicle can also be a single vehicle, and the target vehicle can also be other vehicles that execute operation tasks. The number of target vehicles is not limited here. That is, unified task allocation can be performed through the target vehicles associated with the cloud platform. The cloud platform can analyze the operation task execution status of all vehicles and send their respective paths to the corresponding vehicles. The target vehicle may be in an empty or full load state, and its state affects the path planning decision. An empty target vehicle can be a mining truck before unloading or waiting for loading. A full-load target vehicle can be a mining truck that has completed the loading task and is ready to return to the unloading point or exit. By monitoring the position and state of the target vehicle in real time, its path planning can be dynamically adjusted to handle emergencies or adjust the overall operation process.

[0055] In an alternative embodiment, the target path can be sent to the target vehicle through a wireless communication network. Specifically, the target path can be converted into a digital code or a coordinate sequence for easy network transmission. To prevent the data from being intercepted or tampered with by a third party, the digital code or coordinate sequence can also be encrypted and sent to the target vehicle in a secure manner through the wireless network. After receiving the encrypted data, the target vehicle uses the corresponding key for decoding and presents the target path in the driving assistance system or the automatic navigation system.

[0056] In another alternative embodiment, radio frequency identification technology and the global positioning system can also be combined. RFID readers are arranged at key positions. When a vehicle passes by, the identification information stored on the vehicle RFID tag is automatically read, and the vehicle position is determined through GPS positioning. Subsequently, the target path is sent to the vehicle.

[0057] In yet another alternative embodiment, the target vehicle can also receive the target path from the nearest vehicle or base station that knows its target path through vehicle-to-vehicle communication technology, especially applicable to operating areas with poor network signals. This requires each vehicle to be equipped with a V2V communication module capable of automatically establishing a communication link with surrounding vehicles. The target path is first sent to the vehicle closest to the target vehicle, and this vehicle then transmits the data step by step through the V2V network to the target vehicle. During the transmission of path information, if there are temporary environmental changes (such as obstacles, temporarily closed sections), the V2V network can quickly provide feedback, and the target vehicle can immediately adjust its driving route.

[0058] Furthermore, the target path can also be sent to the target vehicle through map data. The generated target path can be pre-structured to obtain map data so that the vehicle can achieve navigation based on the map data.

[0059] Through the above steps, it is possible to effectively solve the problems of low driving efficiency and safety caused by path conflicts in traditional path planning. Specifically, by comprehensively analyzing the entrance and exit information and spatial orientation information, the generated candidate paths can ensure that the vehicle can drive smoothly in the operating area, avoiding unnecessary collisions and waiting times, thus improving the operating efficiency. In addition, the strategy of merging candidate lanes into the main road not only adjusts the path structure but also reduces the number of vehicle turns, further enhancing the driving efficiency. This technical solution is applicable to various operating environments, including but not limited to mining areas, warehouses, factories, logistics centers, etc., and can be flexibly adjusted according to the specific requirements of different scenarios, having a wide range of application scenarios.

[0060] In the embodiment of the present invention, the entrance and exit information of the operating area and the spatial orientation information of multiple operating positions are obtained; based on the entrance and exit information and the spatial orientation information, multiple candidate paths are generated; based on the first type of candidate lanes in the multiple candidate paths, the first type of main road is determined, and based on the second type of candidate lanes in the multiple candidate paths, the second type of main road is determined; at least part of the first type of candidate lanes in each candidate path is merged into the first type of main road, and at least part of the second type of candidate lanes in each candidate path is merged into the second type of main road to generate the target path; the target path is sent to the target vehicle. It is easy to notice that two types of candidate lanes are generated through the entrance and exit information and the spatial orientation information to cover the complete path from the entrance to the operating position and from the operating position back to the exit. And in the scenario of multiple operating positions, a path planning method of introducing the main road and lane merging is adopted. By planning and defining a main road, the path can be planned around the main road, and then at least part of the candidate lanes is merged into the main road to generate the target path, achieving the purpose of orderly path planning, thus realizing the technical effect of improving the driving effect of path planning, and further solving the technical problem of poor path planning effect in the related technology.

[0061] Further, based on the entrance and exit information and the spatial orientation information, multiple candidate paths are generated, including: generating multiple initial paths based on the entrance and exit information, the spatial orientation information, and the kinematic characteristics of the operation vehicle, where each initial path includes a first-type initial lane for connecting the corresponding operation position to the entrance position and a second-type initial lane for connecting the corresponding operation position to the exit position; in the case where there is a conflict area between any two paths among the multiple initial paths, adjusting the multiple initial paths to obtain multiple candidate paths; in the case where there is no conflict area between any two paths among the multiple initial paths, determining the multiple initial paths as multiple candidate paths; where the conflict area is an area where different operation vehicles conflict during driving on any two paths.

[0062] The above-mentioned initial path may include a first-type initial lane and a second-type initial lane. The initial path may be a path initially generated based on the entrance and exit information, the spatial orientation information, and the kinematic characteristics of the operation vehicle, without considering the driving conditions of other vehicles.

[0063] The above-mentioned first-type initial lane may be used to connect the operation position to the entrance position. The first-type initial lane may be a path initially generated based on the entrance and exit information, the spatial orientation information, and the kinematic characteristics of the operation vehicle, without considering the driving conditions of other vehicles.

[0064] The above-mentioned second-type initial lane may be used to connect the operation position to the exit position. The second-type initial lane may be a path initially generated based on the entrance and exit information, the spatial orientation information, and the kinematic characteristics of the operation vehicle, without considering the driving conditions of other vehicles.

[0065] In an alternative embodiment, multiple initial paths can be predicted through a machine learning model, such as a neural network or reinforcement learning, by analyzing the entrance and exit information, the spatial orientation information, and the kinematic characteristics of the operation vehicle. Or, a virtual model can also be constructed based on the entrance and exit information, the spatial orientation information, and the kinematic characteristics of the operation vehicle for real-time generation and visualization of multiple initial paths. This enables the user to visually see the effects of each initial path in a virtual environment and make adjustments or selections according to actual needs.

[0066] Then, it is determined whether there is a conflict area between any two of the multiple initial paths. A conflict area is an area where different work vehicles conflict during the driving process on any two paths. Specifically, it can be judged whether there are intersection points or too close distances on two paths by analyzing the geometric shape and spatial position of the paths. Alternatively, dynamic analysis can be performed on the vehicle during the path driving process, and the driving state of the vehicle can be simulated through a model to predict the actual driving position of the vehicle on the initial path. By comparing the predicted behaviors of the vehicle on any two initial paths, if the positions of the vehicle are too close or overlapping at a specific time point in the prediction, there is a conflict. This helps to identify and adjust the path in advance before the vehicle actually reaches the conflict area. Or, a spatio-temporal network can also be used for conflict detection. Based on the driving speed and path of the vehicle, the spatio-temporal grid that the vehicle may occupy at each time point can be predicted. Check whether any two paths simultaneously occupy the same grid at the same time point; if they do, it indicates a conflict. In the case of a conflict area, it is necessary to adjust the multiple initial paths to obtain multiple candidate paths. If there is no conflict area, the multiple initial paths can be determined as multiple candidate paths for further generating the target path.

[0067] The kinematic characteristics can refer to the physical limitations of the vehicle during driving, such as turning radius, maximum speed, vehicle length and width, and safety bounding box. By considering the kinematic characteristics of the vehicle, the generated initial path can ensure the safe driving of the vehicle in the operation area and avoid the problem of infeasible paths caused by physical limitations. In addition, by identifying the conflict area, it can ensure that the multiple paths do not interfere with each other, avoid possible collisions during the driving process of the vehicle, and improve the operation safety. This technical solution is not only applicable to a single type of work vehicle, but also can be applied to a complex environment where multiple types of vehicles coexist, including but not limited to forklifts, drones, etc., and can adjust the path according to the characteristics of different types of vehicles.

[0068] Further, adjust multiple initial paths to obtain multiple candidate paths, including: matching any two of the multiple initial paths to determine at least one first conflict path combination among the multiple initial paths, where there is a conflict area between two paths in the same first conflict path combination; adjusting at least one first conflict path combination to obtain multiple adjusted paths; when there is a conflict area between any two of the multiple adjusted paths and the number of iterations has not reached the preset number, match any two of the multiple adjusted paths to determine at least one second conflict path combination among the multiple adjusted paths, and adjust at least one second conflict path combination to obtain the paths in the next adjustment process; when there is no conflict area between any two of the multiple adjusted paths or the number of iterations reaches the preset number, use the multiple adjusted paths as multiple candidate paths.

[0069] The above-mentioned first conflict path combination can be a combination of two initial paths. If there is a conflict between two initial paths, these two initial paths can be matched to obtain the first conflict path combination. The first conflict path combination can be one group or multiple groups. By identifying the first conflict path combination, measures can be taken in advance to avoid the collision risk between vehicles and ensure the safe operation of the operation area. By adjusting the paths determined as the first conflict path combination, the conflict area can be reduced, and the overall feasibility and efficiency of the paths can be improved. Identifying conflict paths also helps to adjust the driving order and speed of vehicles, adjust the traffic flow in the operation area, and reasonably allocate operation resources.

[0070] The above-mentioned second conflict path combination can be used to represent the situation where there is still a conflict area between any paths after adjusting the paths with conflict areas. The second conflict path combination can be one group or multiple groups. By identifying and processing the second conflict path combination, the path planning algorithm can continuously adjust the paths in multiple iterations until all path combinations do not interfere with each other in space, achieving a global solution. As the number of iterations increases, the accuracy of path planning also improves accordingly, and it can more accurately handle the subtle conflicts that may occur in complex scenarios, improving the refinement level of vehicle scheduling in the operation area. In the environment of multi-vehicle collaborative operation, by continuously identifying and reducing conflicts, not only the safety is improved and the manual intervention is reduced, but also the waiting and scheduling delays of vehicles such as mining trucks can be avoided to the greatest extent, and the operation efficiency is enhanced.

[0071] In an alternative embodiment, path adjustment can be based on a local conflict adjustment strategy. By identifying and locally adjusting the conflict points in the initial path, the conflict area is gradually reduced, and finally a candidate path is generated. The spatio-temporal grid technology can be used to identify any two paths with conflict areas in the initial path, forming a first conflict path combination. For each first conflict path combination, a smoothing algorithm is used to adjust the path to avoid the conflict area, obtaining multiple adjusted paths. Then, it is checked whether there are new conflict areas in the adjusted paths. If so, the conflict paths are determined as the second conflict path combination, and the adjustment step is repeated until there are no more conflict areas or the preset number of iterations is reached. Finally, the conflict-free path is used as the candidate path.

[0072] In another alternative embodiment, the interaction of paths can be analyzed from a global perspective to adjust the paths in a collaborative manner and reduce the conflict area. Each initial path can be regarded as an edge in a graph, and a network graph containing all paths is established. On the network graph, all path segments with conflicts are marked to form a conflict area network. Then, graph theory algorithms, such as the shortest path algorithm, minimum spanning tree, etc., are used. Here, the graph theory algorithm is not limited and can be determined according to needs. Next, the network graph is re-planned to ensure that all paths do not conflict at the global level. It is checked whether there are conflicts in the re-planned global paths. If so, the adjustment continues until the conflict-free condition is met or the number of iterations reaches the upper limit. The final conflict-free path is the candidate path.

[0073] In yet another alternative embodiment, path conflict prediction and adjustment based on deep learning can be performed. A deep learning model is used to predict path conflicts and adjust the paths accordingly to generate candidate paths. The deep learning model can be a convolutional neural network, a recurrent neural network, etc. Here, the deep learning model is not limited and can be determined according to needs. For each initial path, the deep learning model predicts the possibility of conflict with other paths, identifying the first conflict path combination. Based on the conflict prediction results, the first conflict path combination is adjusted, such as changing the path curve, changing the driving order, or adjusting the speed. The deep learning model can perform iterative learning based on the feedback of the adjusted paths to continuously improve the prediction accuracy. After multiple iterations, the paths with conflict prediction results lower than the threshold are used as candidate paths, or when the preset number of iterations is reached, the adjustment process is stopped, and the current path set is used as the candidate path.

[0074] By adjusting a single conflict combination and then traversing and analyzing any two initial paths with conflict areas until there are no more conflict areas in the paths or the maximum number of iterations set by the algorithm is reached. The path structure can be gradually adjusted, the conflict area can be reduced, and the driving efficiency and safety of the work vehicle in the work area can be improved. In addition, by setting an upper limit on the number of iterations, an infinite loop adjustment is avoided, ensuring the efficiency of path planning.

[0075] In the scenario of collaborative path planning for mining trucks in the loading area of an open-pit mine, multiple initial paths corresponding to an excavator can be generated according to the position and orientation of the excavator, with reference to the double-lane path planning algorithm. The multiple initial paths can include the reversing lane for entering the operation area and the oncoming lane for exiting the operation area. When there are multiple excavators in the operation area, check whether there are conflict areas in the initial paths of different excavators, and pair up the initial paths with conflict areas two by two to obtain the first conflict path combinations.

[0076] Furthermore, at least one first conflict path combination is adjusted to obtain multiple adjusted paths, including: obtaining a target conflict path combination according to the conflict path combination corresponding to the largest conflict area in at least one first conflict path combination; adjusting the two paths in the target conflict path combination respectively to obtain two initial adjustment results; determining a target adjustment result from the two initial adjustment results based on the evaluation indicators of the two initial adjustment results; and adjusting the two paths in the target conflict path combination based on the target adjustment result to obtain multiple adjusted paths.

[0077] In an alternative embodiment, path adjustment can be performed based on the conflict area density. By analyzing the density of the conflict areas, the path combination with the highest conflict density is preferentially adjusted. For each first conflict path combination, calculate the vehicle density per unit area of the conflict area to determine the target conflict path combination with the largest conflict density. Adjust the two paths in the target conflict path combination respectively, for example, by changing the path curve, adjusting the driving direction or speed, to obtain two initial adjustment results. Then, use evaluation indicators, such as path length, estimated driving time, energy consumption, etc. The evaluation indicators are not limited here and can be determined according to needs. Then, evaluate the two initial adjustment results, and the result that can reduce the conflict area the most or has the highest path performance can be selected as the target adjustment result. Based on the target adjustment result, modify the two paths in the target conflict path combination to generate adjusted paths and integrate these paths into the overall path planning.

[0078] In another alternative embodiment, the collision probability under different path adjustment schemes can be predicted through virtual simulation, and the scheme that can most effectively reduce the collision risk is selected for path adjustment. First, establish a virtual simulation environment, constructing a virtual environment similar to the actual operation area, including terrain, lane layout, vehicle characteristics, etc. In the virtual environment, perform multiple simulations on the target conflict path combination, trying different adjustment schemes, such as by fine-tuning the path curvature, changing the driving order, etc. Predict and analyze the collision risk for each adjustment scheme, and identify the initial adjustment result with the lowest collision probability as the target adjustment result for application in the actual path planning to effectively adjust the conflict path combination and generate multiple adjusted paths.

[0079] Furthermore, the evaluation metrics can be performance metrics for evaluating the result of path adjustment, such as path length, travel time, number of turns, etc. For example, when adjusting two conflicting paths, the adjustment effect can be evaluated by calculating the length and the number of turns of the adjusted path. By adjusting based on the target conflicting path combination with the largest conflict area, this technical solution can preferentially solve the most serious path conflict problem and improve the efficiency of path planning. In addition, by introducing the evaluation metrics, it can ensure the improvement of the path adjustment result and avoid the decrease in path passing efficiency caused by blind adjustment.

[0080] Further, the two paths in the target conflicting path combination are adjusted separately to obtain two initial adjustment results, including: for any one of the paths, the path is adjusted according to multiple adjustment step lengths to obtain multiple candidate adjustment results; determining the conflict area between different candidate adjustment results and the other path; and obtaining the initial adjustment result according to the candidate adjustment result corresponding to the smallest conflict area among the multiple candidate adjustment results.

[0081] In an alternative embodiment, the path can be adjusted based on the path adjustment technique of perturbation to obtain two initial adjustment results. By applying a small perturbation to the path, that is, adjusting the local or the whole of the path with different step lengths, the conflict area can be reduced. One of the paths in the target conflicting path combination can be selected, and a series of adjustment step lengths (for example, 1 meter, 2 meters, 3 meters) can be set, and small displacements are made at the key points of the path (such as near the starting point, midpoint, end point or conflict point) to generate multiple candidate adjustment results, and the values here are only examples. Then, for each candidate adjustment result, analyze the conflict area between it and the other path, which can be analyzed by the spatio-temporal grid technique. Then, according to the size of the conflict area, select the candidate adjustment result with the smallest conflict area to ensure the minimum interference between the adjusted path and the other path, and thus determine the initial adjustment result. And, the above steps are also repeated for the other path in the target conflicting path combination to finally obtain the initial adjustment result.

[0082] In another alternative embodiment, the principle of genetic algorithm can be applied to find a path adjustment scheme through operations such as crossover and mutation in multiple generations of evolution. An initial population containing multiple adjustment schemes is generated for each path, and each scheme represents a candidate adjustment result. Define an evaluation function to calculate the size of the conflict area between each path adjustment scheme and the other path, and use this as the basis for selecting the evolution direction. Based on the evaluation function, through the selection, crossover and mutation operations in the genetic algorithm, the population is evolved to find the path adjustment scheme with the smallest conflict area. After multiple generations of evolution, select the scheme with the smallest conflict area as the initial adjustment result, and execute this strategy for the two paths respectively to obtain two initial adjustment results.

[0083] The adjustment step size can refer to the incremental path length for each adjustment during the path adjustment process. For example, when adjusting a path, the adjustment step size can be set to 1 meter, and the path is adjusted step by step until the adjustment result of the minimum conflict area is found. The numerical value here is only for illustration. By adjusting the path according to multiple adjustment step sizes, this technical solution can finely adjust each path, reduce the conflict area, and improve the driving efficiency and safety of the work vehicle in the work area. In addition, by determining the initial adjustment result of the minimum conflict area, the result of the path adjustment can be ensured, and the expansion of the conflict area can be avoided.

[0084] In the case of specific application to mining trucks, the evaluation index can be the sum of the distances from the reversing completion point to the excavator. The smaller the value, the better the obtained solution. Select the combination with the largest conflict area from the first conflict path combinations, that is, push the first conflict path combinations along the direction of the reversing completion point, upward, left, and right directions. The stepping distance is generally set to 1 meter, and the maximum pushing distance is generally set to 5 meters. The numerical values here are only for illustration. Retain the pushing result with the smallest conflict area with another first conflict path combination, and calculate the evaluation index at this time as . Then, perform the same processing on another first conflict path combination to obtain the evaluation index , compare and , and take the pushing result corresponding to the minimum value as the adjustment plan.

[0085] Furthermore, the first type of candidate lanes includes driving lanes and reversing lanes. Based on the first type of candidate lanes in multiple candidate paths, determine the first type of main roads. Based on the second type of candidate lanes in multiple candidate paths, determine the second type of main roads, including: determine the driving lane with the longest length as the first type of main road from the driving lanes included in multiple candidate paths; determine the candidate lane with the longest length as the second type of main road from the second type of candidate lanes included in multiple candidate paths.

[0086] The above-mentioned first type of candidate lanes may include driving lanes and turning lanes. A driving lane may refer to a straight passageway for a vehicle to travel from an entrance position to an operation position. In particular, a driving lane may be an empty-load lane, which may refer to the driving path of a mining truck from a starting point to an operation position when the truck is not loaded with ore or other materials. The empty-load lane corresponds to the full-load lane (i.e., the driving path of the mining truck after loading). The empty-load lane mainly focuses on the driving requirements of the mining truck in the empty-load state, including but not limited to the directness of the route, driving speed, safety distance, etc. The planning of the driving lane needs to consider factors such as the terrain of the mine, spatial layout, size of the mining truck, driving speed, and operation efficiency. Its design goal is to enable the mining truck to complete the driving task in the shortest time with the least energy consumption while avoiding collisions with other vehicles or obstacles. A turning lane may refer to a lane used when a vehicle makes a turn, U-turn, or changes direction. For example, within the operation area of an open-pit mine, the mining truck needs to turn beside the excavator for loading operations, and in the unloading area, the mining truck needs to turn to safely leave. The turning lane needs to consider the turning radius of the vehicle, safety distance, and spatial limitations within the loading area to ensure that the mining truck does not conflict with other vehicles or fixed obstacles during turning, and at the same time, minimize the time and space required for turning to improve operation efficiency. By distinguishing between driving lanes and turning lanes, the rationality of the path planning can be ensured, avoiding unnecessary turning and waiting times.

[0087] In an alternative embodiment, based on the path length sorting, the driving lane with the longest length may be determined as the first type of main road from the driving lanes, and the candidate lane with the longest length may be determined as the second type of main road from the second type of candidate lanes. That is, sort the lengths of the driving lanes and select the driving lane with the longest length as the first type of main road. Sort the lengths of the second type of candidate lanes and select the second type of candidate lane with the longest length as the second type of main road.

[0088] In another alternative embodiment, the shortest path algorithm in graph theory can be utilized to convert it into a problem of finding the longest path. By constructing a lane network graph, the longest path from the starting point to the ending point is found. Consider the driving lanes of all candidate paths as the edges in the graph, and the connection points of adjacent lanes as the nodes, forming a complete lane network graph. Assign weights to each lane (edge), where the weight can be the negative length of the lane. In this way, the longest lane will be transformed into the shortest path problem in graph theory, that is, finding the longest path from the starting node to the ending node, and the lanes included in this path are the selected primary roads of the first type. Similarly, for the candidate lanes of the second type, construct a lane network graph and find the longest path from the starting point to the ending point. Consider the candidate lanes of the second type in all candidate paths as the edges in the graph, and the connection points of adjacent lanes as the nodes, forming a complete lane network graph. Assign weights to each lane (edge), where the weight can be the negative length of the lane. In this way, the longest lane will be transformed into the shortest path problem in graph theory, that is, finding the longest path from the starting node to the ending node, and the lanes included in this path are the selected primary roads of the second type.

[0089] By determining the driving lanes and candidate lanes with the longest lengths as the primary roads, this technical solution can construct a more efficient driving path within the operation area, reduce the vehicle driving distance, and improve the operation efficiency.

[0090] Furthermore, incorporating at least a part of the candidate lanes of the first type in each candidate path into the primary roads of the first type, and incorporating at least a part of the candidate lanes of the second type in each candidate path into the primary roads of the second type to generate a target path, includes: determining the first bifurcation point between the candidate lanes of the first type and the primary roads of the first type, and determining the second bifurcation point between the candidate lanes of the second type and the primary roads of the second type; sampling the candidate lanes of the first type to generate the first connection path between the first bifurcation point and the candidate lanes of the first type, and sampling the candidate lanes of the second type to generate the second connection path between the second bifurcation point and the candidate lanes of the second type, where the intersection point between the first connection path and the candidate lanes of the first type is the first target sampling point, and the intersection point between the second connection path and the candidate lanes of the second type is the second target sampling point; splicing the section of the primary roads of the first type before the first bifurcation point, the first connection path, and the section of the candidate lanes of the first type after the first target sampling point to generate the target lanes of the first type, and splicing the section of the primary roads of the second type before the second bifurcation point, the second connection path, and the section of the candidate lanes of the second type after the second target sampling point to generate the target lanes of the second type, where the target lanes of the first type and the target lanes of the second type meet the preset conditions; obtaining the target path based on the target lanes of the first type and the target lanes of the second type.

[0091] In an alternative embodiment, the first bifurcation point can be determined based on the relative position and orientation of the first type of candidate lane and the first type of main road. The first type of candidate lane can be a feeder road to merge each feeder road into the main road. The first bifurcation point can be manually marked or predicted through reinforcement learning. Similarly, the second bifurcation point can be manually marked, or the second bifurcation point between the second type of candidate lane and the second type of main road can be predicted through reinforcement learning.

[0092] Then, sample the first type of candidate lane to generate the first connection path between the first bifurcation point and the first type of candidate lane. Among them, the intersection point between the first connection path and the first type of candidate lane is the first target sampling point. Multiple first target sampling points can be determined by sampling the first type of candidate lane at a fixed interval or according to the safety distance condition starting from the first bifurcation point. Then, use the path planning algorithm to generate the first connection path. Similarly, the second type of candidate lane can be sampled at a fixed interval or according to the safety distance condition to determine the second target sampling point. Then, use the path planning algorithm to generate the second connection path between the second bifurcation point and the first type of candidate lane.

[0093] After obtaining the first connection path and the second connection path, a smooth transition can be generated between the endpoints of the section of the first type of main road before the first bifurcation point, the first connection path, and the section of the first type of candidate lane after the first target sampling point through linear interpolation, and stitched to generate the first type of target lane. Or, the endpoints of the three sections can be connected through smooth curve fitting technology, and the first type of target lane can be generated by adjusting the curve parameters. The smooth curve fitting can be a discrete point smoothing method, and the smooth curve fitting can also adopt the Gaussian process regression algorithm. Or, the smooth curve fitting can also be spline curve fitting, that is, fitting through polynomial segments. The objective function adopted by the smooth curve fitting can be the sum of the smoothness cost, the length cost, and the original point offset cost. Similarly, the section of the second type of main road before the second bifurcation point, the second connection path, and the section of the second type of candidate lane after the second target sampling point can be stitched through linear interpolation or smooth curve fitting technology to generate the second type of target lane. The generated first type of target lane and the second type of target lane meet the preset conditions. Specifically, through the above stitching process, the first type of target lane and the second type of target lane meet the position constraint, curvature constraint, and the orientation constraint conditions of the starting and ending points.

[0094] Finally, based on the first type of target lanes and the second type of target lanes, a target path can be obtained. The first type of target lanes and the second type of target lanes are fused through the lane priorities to obtain the target path. Alternatively, the first type of target lanes and the second type of target lanes can also be solved by constraint through mixed integer linear programming to obtain the target path. Or, the behavior of the vehicle driving on the first type of target lanes and the second type of target lanes can also be simulated through a multi-agent system, and a coordination algorithm is used to generate the target path for the vehicle to drive.

[0095] The bifurcation point is the connection point between the main road and the branch road, and sampling means selecting multiple points on the path to generate the path. By determining the bifurcation point and the connection path, this technical solution can ensure the coherence and rationality of the driving path of the vehicle in the operation area, avoid unnecessary turning and waiting time, and improve the operation efficiency. In addition, by splicing the path through the target lanes that meet the preset conditions, the efficiency of path planning can be ensured, and the degradation of path performance can be avoided.

[0096] Further, determining the first bifurcation point between the first type of candidate lanes and the first type of main road, and determining the second bifurcation point between the second type of candidate lanes and the second type of main road includes: based on the type of the operation vehicle, constructing a first buffer area corresponding to the first type of candidate lanes, and a second buffer area corresponding to the second type of candidate lanes; determining at least one first intersection point between the first type of main road and the first buffer area, and at least one second intersection point between the second type of main road and the second buffer area; determining the first bifurcation point from at least one first intersection point, and determining the second bifurcation point from at least one second intersection point, where the distance between the first bifurcation point and the first type of candidate lanes is greater than the distance between the intersection points other than the first bifurcation point and the first type of candidate lanes, and the distance between the second bifurcation point and the second type of candidate lanes is greater than the distance between the intersection points other than the second bifurcation point and the second type of candidate lanes.

[0097] The above-mentioned buffer area refers to the minimum safety distance area set in path planning to avoid collisions between vehicles. The buffer area can be a rectangular area on both sides of the lane line. For example, in a warehouse environment, the buffer area may be the minimum safety distance between the shelves. By constructing the first buffer area and the second buffer area, this technical solution can ensure the safety of the driving path of the vehicle in the operation area and avoid collisions between vehicles. In addition, by determining the intersection points between the bifurcation point and the buffer area, the rationality of path planning can be ensured, and the blind spots in path planning can be avoided.

[0098] In an alternative embodiment, a first buffer area corresponding to the first type of candidate lane and a second buffer area corresponding to the second type of candidate lane can be constructed based on the type of the work vehicle. Different types of work vehicles will have different buffer areas. The size information of the vehicle, such as width, length, and safety boundary, can be determined by the type of the work vehicle. Then, taking the center lines of the first type of candidate lane and the second type of candidate lane as references respectively, expand half of the vehicle width plus the safety boundary to the left and right to construct the corresponding first buffer area and second buffer area. Or, based on the type of the vehicle, determine the motion characteristics of the vehicle, such as braking distance and turning radius. By analyzing the motion characteristics, determine the first buffer area and the second buffer area.

[0099] Then, predict the first intersection point and the second intersection point through manual analysis or reinforcement learning. Then, determine the first bifurcation point from at least one first intersection point to ensure that the distance between the first bifurcation point and the first type of candidate lane is greater than the distance between the intersection points other than the first bifurcation point and the first type of candidate lane. And determine the second bifurcation point from at least one second intersection point to ensure that the distance between the second bifurcation point and the second type of candidate lane is greater than the distance between the intersection points other than the second bifurcation point and the second type of candidate lane.

[0100] Specifically, the distance between each first intersection point and the first type of candidate lane can be calculated, and the intersection point that is the farthest from the first type of candidate lane among the first intersection points of the first buffer area and the first type of candidate lane is selected as the first bifurcation point to reduce the conflict risk when the vehicle enters or leaves the candidate lane. Or, by setting a safety distance threshold according to the type and driving speed of the work vehicle, analyze all first intersection points, evaluate whether the path from the first intersection point to the first type of candidate lane can meet the set safety distance requirement, and select the intersection point that is the farthest from the first type of candidate lane and can provide a better view among the intersection points that meet the safety assessment as the first bifurcation point to improve the safety and smoothness of vehicle driving. Similarly, the distance between each second intersection point and the second type of candidate lane can also be calculated, and the intersection point that is the farthest from the second type of candidate lane is selected as the second bifurcation point.

[0101] Specifically, the buffer area can be constructed using the following formula:

[0102] F = ;

[0103] where F represents the expansion parameter of the buffer area, in meters, represents the vehicle width, represents the safety bounding box, represents the error value.

[0104] Further, sample the first type of candidate lanes to generate a first connection path between the first bifurcation point and the first type of candidate lanes, and sample the second type of candidate lanes to generate a second connection path between the second bifurcation point and the first type of candidate lanes, including: determining a first target path point closest to the first bifurcation point in the first type of candidate lanes and a second target path point closest to the second bifurcation point in the second type of candidate lanes; sampling on the first type of candidate lanes starting from the first target path point to obtain a plurality of first sampling points, and sampling on the second type of candidate lanes starting from the second target path point to obtain a plurality of second sampling points; based on the kinematic characteristics of the vehicle, sequentially connecting the first sampling points and the first bifurcation point to obtain a plurality of first reference paths, and sequentially connecting the second sampling points and the second bifurcation point to obtain a plurality of second reference paths; obtaining the first connection path at least according to the shortest path among the plurality of first reference paths, and obtaining the second connection path according to the shortest path among the plurality of second reference paths.

[0105] In an alternative embodiment, the first target path point closest to the first bifurcation point in the first type of candidate lanes and the second target path point closest to the second bifurcation point in the second type of candidate lanes can be determined manually. Here, the first type of candidate lanes can be a feeder road. Or, the first target path point with the minimum distance from the first bifurcation point can be calculated from the first type of candidate lanes through the Euclidean distance, and the second target path point closest to the second bifurcation point can be obtained from the second type of candidate lanes.

[0106] Then, a plurality of first sampling points can be obtained by sampling on the first type of candidate lanes starting from the first target path point according to a fixed distance or a sampling frequency determined by the kinematic characteristics of the vehicle. Similarly, a plurality of second sampling points can be obtained by sampling on the second type of candidate lanes starting from the second target path point according to a fixed distance or a sampling frequency determined by the kinematic characteristics of the vehicle.

[0107] Next, based on the kinematic characteristics of the vehicle, the first sampling points and the first bifurcation point can be sequentially connected to obtain a plurality of first reference paths. The calculation parameters of the path planning algorithm can be defined through the kinematic characteristics of the vehicle, and the first reference paths that can connect the first sampling points and the first bifurcation point can be selected through the path planning algorithm. Or, a vehicle dynamic model can also be constructed through the kinematic characteristics of the vehicle, and starting from the first bifurcation point, the paths connecting the first bifurcation point to the first sampling points can be predicted to obtain a plurality of first reference paths. Similarly, the second sampling points and the second bifurcation point can also be sequentially connected through the path planning algorithm or the vehicle dynamic model to obtain a plurality of second reference paths.

[0108] Finally, the shortest path in the first reference path can be determined as the first connection path by the straight-line distance, or the shortest path among multiple first reference paths can be determined as the first connection path by the shortest path algorithm in graph theory. Similarly, the shortest path among multiple second reference paths can be determined as the second connection path by the distance comparison method or the shortest path algorithm in graph theory.

[0109] The target path point can be the intersection point closest to the bifurcation point. The reference path is a path generated based on the sampling points and the bifurcation point. By determining the target path point and generating the reference path, this technical solution can ensure the coherence and rationality of the driving path of the vehicle within the operation area, avoid unnecessary turning and waiting times, and improve the operation efficiency. In addition, by connecting the paths based on the kinematic characteristics of the vehicle, the feasibility of the path planning can be ensured, and the problem of infeasible paths caused by physical limitations can be avoided.

[0110] Next, a technical solution proposed in this application will be described in conjunction with an optional embodiment, which relates to a method for generating a target path.

[0111] The nearest point on the first type of candidate lane to the first bifurcation point can be obtained. Starting from the nearest point, sampling is performed every 5 meters in sequence, and the maximum sampling distance is 30 meters to obtain a first sampling point set. The values here are only for illustration. Traverse the first sampling points in sequence, and use the curve smoothing algorithm to connect the first bifurcation point to the first sampling points to obtain a first reference path set. Select the shortest path in the reference path set to obtain the first connection path. Similarly, the nearest point on the second type of candidate lane to the second bifurcation point can be obtained, and starting from the nearest point, a second sampling point set can be obtained. Traverse the second sampling points in sequence, and use the curve smoothing algorithm to connect the second bifurcation point to the second sampling points to obtain a second reference path set. Select the shortest path in the reference path set to obtain the second connection path.

[0112] Then, the section of the first type of main road before the first bifurcation point, the first connection path, and the section of the first type of candidate lane after the first target sampling point are spliced to generate the first type of target lane, and the section of the second type of main road before the second bifurcation point, the second connection path, and the section of the second type of candidate lane after the second target sampling point are spliced to generate the second type of target lane. And a smoothing method is used to obtain the first type of target lane and the second type of target lane with the end position unchanged and satisfying the curvature constraint. Finally, the first type of target lane and the second type of target lane are merged to obtain the target path. In addition, when the first type of candidate lane is a driving lane, the first type of candidate lane and the reversing lane need to be spliced again to obtain the merged first type of target lane.

[0113] In the above embodiments, different types of main roads are introduced. Through path planning of lane merging, in a multi-excavator operation scenario, a main road is planned and defined in advance, and the paths of the mining trucks are planned with the main road as the core. Then, paths leading to different excavators are branched out from the main road to ensure the orderliness and coordination of the overall path. And the path adjustment and separation of multi-excavator commutation are considered. When generating the commutation path, the spatial factors of the operation area and the path interference among multiple excavators are fully considered. The commutation paths of different excavators are separated in advance by the algorithm to ensure that the paths do not interfere with each other and avoid vehicle conflicts caused by path crossing. Global collaborative path planning is carried out to avoid the "butting" phenomenon. When planning the path, the driving requirements of all excavators and mining trucks are comprehensively considered, and a reasonable main road and branched paths are defined in advance, so as to avoid path conflicts during the driving of mining trucks. By planning the path in advance, multi-vehicle collaborative operation is realized, avoiding post-event path conflict handling and vehicle waiting, and improving the operation efficiency of vehicles such as mining trucks in the operation area.

[0114] As Figure 2 shown, it shows the situation of driving according to the planned path. Figure 2 It includes 2 working positions 202, 2 first-type candidate lanes 2061 and 2 second-type candidate lanes 2062, and 2 first-type main roads 208. Among them, there is a first intersection 2001 between the 2 first-type candidate lanes, and there is a second intersection 2002 between the 2 second-type candidate lanes.

[0115] According to an embodiment of the present invention, there is also provided a method embodiment of an optional path planning method. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. And although the logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in a different order than here.

[0116] Figure 3 is a flowchart of an optional path planning method according to an embodiment of the present invention. As Figure 3 shown, the method includes the following steps:

[0117] Step S302, receive the target path.

[0118] Among them, the target path is generated by incorporating at least a part of the first - type candidate lanes in each of the multiple candidate paths into the first - type main road and incorporating at least a part of the second - type candidate lanes in each of the candidate paths into the second - type main road. The first - type main road is determined based on the first - type candidate lanes in the multiple candidate paths, and the second - type main road is determined based on the second - type candidate lanes in the multiple candidate paths. The multiple candidate paths are generated based on the entrance and exit information of the operation area and the spatial orientation information of multiple operation positions. The entrance and exit information includes the entrance position and the exit position of the operation area. Each candidate path includes a first - type candidate lane for connecting the corresponding operation position to the entrance position and a second - type candidate lane for connecting the corresponding operation position to the exit position.

[0119] In an alternative embodiment, the vehicle can receive the target path through V2V communication technology. Or, the vehicle can also utilize on - vehicle vision sensors and simultaneous localization and mapping technology, and the vehicle can autonomously identify and understand the target path without receiving path information through wireless communication.

[0120] The target path can be generated by incorporating at least a part of the first - type candidate lanes in each of the multiple candidate paths into the first - type main road and incorporating at least a part of the second - type candidate lanes in each of the candidate paths into the second - type main road. The generation of the target path can adopt the method in the above - mentioned embodiment.

[0121] The first - type main road is determined based on the first - type candidate lanes in the multiple candidate paths. The determination of the first - type main road can adopt the method in the above - mentioned embodiment. The second - type main road can be determined based on the second - type candidate lanes in the multiple candidate paths. The determination of the second - type main road can adopt the method in the above - mentioned embodiment.

[0122] The multiple candidate paths can be generated based on the entrance and exit information of the operation area and the spatial orientation information of multiple operation positions. The generation of the candidate paths can adopt the method in the above - mentioned embodiment. Among them, the entrance and exit information can include the entrance position and the exit position of the operation area. Each candidate path can include a first - type candidate lane for connecting the corresponding operation position to the entrance position and a second - type candidate lane for connecting the corresponding operation position to the exit position.

[0123] Step S304, drive according to the target path.

[0124] In an alternative embodiment, a control signal can be generated by a controller to control the vehicle to drive according to the target path. Or, model predictive control (MPC) can also be used to predict the future driving state using the vehicle dynamics model and minimize the deviation from the target path through a control sequence.

[0125] In this embodiment, at least a part of the first type of candidate lanes among multiple candidate paths can be merged into the first type of main road, and at least a part of the second type of candidate lanes can be merged into the second type of main road. Such a merging strategy can effectively reduce path conflicts and improve the overall smoothness and efficiency of vehicle driving. The main road is determined based on the information of all the first type of candidate lanes, and the branch lanes are paths separated from the main road and can be used for direct driving to specific operation positions. This distinction helps to enhance the hierarchical structure of the paths and facilitates vehicle operation scheduling. The entrance and exit information of the operation area and the spatial orientation information of multiple operation positions are used to generate candidate paths. This method ensures the rationality of path planning, enabling the generated paths to adapt to the actual layout and requirements of the operation environment. By adopting this method, through merging paths and distinguishing between the main road and branch lanes, path conflicts of vehicles in the operation area are effectively avoided, the risk of vehicle collisions is reduced, and the safety level of operations is improved. In path merging, considering the actual needs of the operation positions and the convenience of vehicle entry and exit, the path planning becomes more flexible, can quickly adapt to changing operation conditions, and improves the driving efficiency of vehicles and the operation productivity of the operation area.

[0126] According to an embodiment of the present invention, there is provided an apparatus embodiment of a path planning apparatus. It should be noted that this apparatus can be used to execute the above path planning method.

[0127] Figure 4 is a schematic diagram of an optional path planning apparatus according to an embodiment of the present invention, as Figure 4 shown, the apparatus includes:

[0128] An acquisition module 40, configured to acquire the entrance and exit information of the operation area, as well as the spatial orientation information of multiple operation positions, where the entrance and exit information includes the entrance position and the exit position of the operation area.

[0129] A generation module 42, configured to generate multiple candidate paths based on the entrance and exit information and the spatial orientation information, where each candidate path includes a first type of candidate lane for connecting the corresponding operation position to the entrance position and a second type of candidate lane for connecting the corresponding operation position to the exit position.

[0130] A determination module 44, configured to determine a first type of main road based on the first type of candidate lanes among the multiple candidate paths, and determine a second type of main road based on the second type of candidate lanes among the multiple candidate paths;

[0131] A merging module 46, configured to merge at least a part of the first type of candidate lanes in each candidate path into the first type of main road, and merge at least a part of the second type of candidate lanes in each candidate path into the second type of main road to generate a target path.

[0132] A sending module 48, configured to send the target path to the target vehicle.

[0133] Further, the generation module 42 is further configured to generate multiple initial paths based on the entrance and exit information, the spatial orientation information, and the kinematic characteristics of the operating vehicle, where each initial path includes a first-type initial lane for connecting the corresponding operation position to the entrance position and a second-type initial lane for connecting the corresponding operation position to the exit position; in the case that there is a conflict area between any two paths among the multiple initial paths, adjust the multiple initial paths to obtain multiple candidate paths; in the case that there is no conflict area between any two paths among the multiple initial paths, determine the multiple initial paths as multiple candidate paths; where the conflict area is an area where different operating vehicles conflict during the driving process on any two paths.

[0134] Further, the generation module 42 is further configured to match any two initial paths among the multiple initial paths, determine at least one first conflict path combination among the multiple initial paths, where there is a conflict area between the two paths in the same first conflict path combination; adjust at least one first conflict path combination to obtain multiple adjusted paths; in the case that there is a conflict area between any two paths among the multiple adjusted paths and the number of iterations has not reached the preset number of times, match any two initial paths among the multiple adjusted paths, determine at least one second conflict path combination among the multiple adjusted paths, and adjust at least one second conflict path combination to obtain the paths in the next adjustment process; in the case that there is no conflict area between any two paths among the multiple adjusted paths, or the number of iterations reaches the preset number of times, use the multiple adjusted paths as multiple candidate paths.

[0135] Further, the generation module 42 is further configured to obtain a target conflict path combination according to the conflict path combination corresponding to the largest conflict area in at least one first conflict path combination; respectively adjust the two paths in the target conflict path combination to obtain two initial adjustment results; based on the evaluation indexes of the two initial adjustment results, determine a target adjustment result from the two initial adjustment results; and adjust the two paths in the target conflict path combination based on the target adjustment result to obtain multiple adjusted paths.

[0136] Further, the generation module 42 is further configured to, for any one path, adjust the path according to multiple adjustment step lengths to obtain multiple candidate adjustment results; determine the conflict area between different candidate adjustment results and another path; and obtain an initial adjustment result according to the candidate adjustment result corresponding to the smallest conflict area among the multiple candidate adjustment results.

[0137] Further, the first type of candidate lanes includes driving lanes and turning lanes. The determination module 44 is further configured to determine the driving lane with the longest length among the driving lanes included in multiple candidate paths as the first type of main road; and determine the candidate lane with the longest length among the second type of candidate lanes included in multiple candidate paths as the second type of main road.

[0138] Further, the merging module 46 is further configured to determine a first bifurcation point between the first type of candidate lanes and the first type of main road, and determine a second bifurcation point between the second type of candidate lanes and the second type of main road; sample the first type of candidate lanes to generate a first connection path between the first bifurcation point and the first type of candidate lanes, and sample the second type of candidate lanes to generate a second connection path between the second bifurcation point and the second type of candidate lanes. Among them, the intersection point between the first connection path and the first type of candidate lanes is the first target sampling point, and the intersection point between the second connection path and the second type of candidate lanes is the second target sampling point; splice the section of the first type of main road before the first bifurcation point, the first connection path, and the section of the first type of candidate lanes after the first target sampling point to generate the first type of target lane, and splice the section of the second type of main road before the second bifurcation point, the second connection path, and the section of the second type of candidate lanes after the second target sampling point to generate the second type of target lane, where the first type of target lane and the second type of target lane meet the preset conditions; based on the first type of target lane and the second type of target lane, obtain the target path.

[0139] Further, the merging module 46 is further configured to construct a first buffer area corresponding to the first type of candidate lanes and a second buffer area corresponding to the second type of candidate lanes based on the type of the operation vehicle; determine at least one first intersection point between the first type of main road and the first buffer area, and at least one second intersection point between the second type of main road and the second buffer area; determine the first bifurcation point from at least one first intersection point, and determine the second bifurcation point from at least one second intersection point. Among them, the distance between the first bifurcation point and the first type of candidate lanes is greater than the distance between the intersection points other than the first bifurcation point and the first type of candidate lanes, and the distance between the second bifurcation point and the second type of candidate lanes is greater than the distance between the intersection points other than the second bifurcation point and the second type of candidate lanes.

[0140] Further, the merging module 46 is further configured to determine a first target path point closest to the first fork point in the first type of candidate lane and a second target path point closest to the second fork point in the second type of candidate lane; sample on the first type of candidate lane starting from the first target path point to obtain a plurality of first sampling points, and sample on the second type of candidate lane starting from the second target path point to obtain a plurality of second sampling points; based on the kinematic characteristics of the vehicle, sequentially connect the first sampling points to the first fork point to obtain a plurality of first reference paths, and sequentially connect the second sampling points to the second fork point to obtain a plurality of second reference paths; obtain a first connection path at least according to the shortest path among the plurality of first reference paths, and obtain a second connection path according to the shortest path among the plurality of second reference paths.

[0141] According to an embodiment of the present invention, there is provided an apparatus embodiment of an optional path planning apparatus. It should be noted that this apparatus can be used to execute the above path planning method.

[0142] Figure 5 is a schematic diagram of an optional path planning apparatus according to an embodiment of the present invention, as Figure 5 shown, the apparatus includes:

[0143] A receiving module 50, configured to receive a target path, where the target path is generated by incorporating at least a part of the first type of candidate lane in each candidate path among multiple candidate paths into the first type of main road, and incorporating at least a part of the second type of candidate lane in each candidate path into the second type of main road. The first type of main road is determined based on the first type of candidate lane among multiple candidate paths, and the second type of main road is determined based on the second type of candidate lane among multiple candidate paths. The multiple candidate paths are generated based on the entrance and exit information of the operation area and the spatial orientation information of multiple operation positions. The entrance and exit information includes the entrance position and the exit position of the operation area. Each candidate path includes a first type of candidate lane for connecting the corresponding operation position to the entrance position and a second type of candidate lane for connecting the corresponding operation position to the exit position;

[0144] A driving module 52, configured to drive according to the target path.

[0145] According to another aspect of the embodiments of the present invention, there is also provided a vehicle, including: a communication unit, configured to communicate with a cloud server or a target vehicle; a memory, storing an executable program; a processor, configured to run the program, where when the program runs, it executes the methods in the various embodiments of the present invention.

[0146] The above-mentioned communication unit can be used to ensure the effective and reliable transmission of data between the vehicle executing the path planning method proposed in this application and the cloud server or the target vehicle. For example, when the above-mentioned path planning method proposed in this application is executed by the vehicle, the vehicle may include the above-mentioned communication unit. Furthermore, the vehicle can communicate with the target vehicle based on the above-mentioned communication unit and send the generated target path to the target vehicle, etc. When the above-mentioned vehicle is the target vehicle, it can also receive the target path sent by the cloud server or other vehicles.

[0147] The above-mentioned memory can refer to the device inside the computer for storing data and programs, which can include memory, hard disk, etc. Among them, the memory can be used to temporarily store the running programs and data, and the hard disk can be used to store programs and data in the long term. The memory can be used to enable the computer to read and write data and execute programs. The above-mentioned processor can be responsible for executing the instructions in the computer program and performing data processing, and can be responsible for controlling and executing various operations, including arithmetic operations, logical operations, data transmission, etc.

[0148] An embodiment of this application also provides an electronic device, including: a communication unit for communicating with the target vehicle; a memory storing an executable program; a processor for running the program, wherein when the program runs, it executes the methods in various embodiments of the present invention.

[0149] The above-mentioned communication unit can be a basic module or component for transmitting, receiving, and processing information, which can be a hardware module, a software module, or a combination of a hardware module and a software module, and is used to ensure the effective and reliable transmission of data between the platform or vehicle executing the path planning method proposed in this application and the target vehicle. For example, when the above-mentioned path planning method proposed in this application is executed by the platform, the platform may include the above-mentioned electronic device. Furthermore, the platform can communicate with the target vehicle based on the above-mentioned communication unit and send the generated target path to the target vehicle, etc. Another example is when the above-mentioned path planning method proposed in this application is executed by the vehicle, the vehicle may include the above-mentioned electronic device. Furthermore, the vehicle can communicate with the target vehicle based on the above-mentioned communication unit and send the generated target path to the target vehicle, etc. The above-mentioned communication unit can include modules such as a transmitter, a receiver, a modulator, a demodulator, an encoder, a decoder, etc. The specific structure of the communication unit can be determined according to actual needs and is not limited here.

[0150] An embodiment of this application also provides a computer-readable storage medium, which includes a stored executable program, wherein when the executable program runs, it controls the device where the computer-readable storage medium is located to execute the methods in various embodiments of the present invention.

[0151] An embodiment of the present application also provides a computer program product, including a computer program, which implements the methods in various embodiments of the present invention when executed by a processor.

[0152] An embodiment of the present application also provides a computer program product, including a non-volatile computer-readable storage medium for storing a computer program, which implements the methods in various embodiments of the present invention when executed by a processor.

[0153] An embodiment of the present application also provides a computer program, which implements the methods in various embodiments of the present invention when executed by a processor.

[0154] In the above embodiments of the present invention, the descriptions of the various embodiments have their own emphases. For parts not detailed in a certain embodiment, reference may be made to the relevant descriptions of other embodiments.

[0155] In the several embodiments provided by the present application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are merely illustrative. For example, the division of the units can be a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed couplings or direct couplings or communication connections to each other can be through some interfaces. The indirect couplings or communication connections of the units or modules can be in electrical or other forms.

[0156] The units described as separate components may or may not be physically separated. The components displayed as units may or may not be physical units, that is, they can be located in one place or distributed to multiple units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0157] In addition, the functional units in various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically alone, or two or more units can be integrated into one unit. The above integrated units can be implemented in the form of hardware or in the form of software functional units.

[0158] When the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The foregoing storage medium includes: various media that can store program codes, such as USB flash drives, read-only memories (ROMs), random access memories (RAMs), mobile hard disks, magnetic disks, or optical discs.

[0159] The foregoing are only the preferred embodiments of the present invention. It should be noted that for those of ordinary skill in the art, without departing from the principle of the present invention, several improvements and refinements can be made, and these improvements and refinements should also be regarded as the protection scope of the present invention.

Claims

1. A path planning method, characterized in that: include: Acquire entrance and exit information of the working area and spatial position information of multiple working positions, wherein the entrance and exit information includes the entrance position and exit position of the working area; Based on the entrance and exit information and the spatial orientation information, a plurality of candidate paths are generated, wherein each candidate path includes a first type candidate lane for connecting a corresponding work position with the entrance position and a second type candidate lane for connecting a corresponding work position with the exit position; Determine a first type of main road based on the first type of candidate lanes in the plurality of candidate paths, and determine a second type of main road based on the second type of candidate lanes in the plurality of candidate paths; Merging at least part of the first type candidate lanes in each candidate path into the first type main road, and merging at least part of the second type candidate lanes in each candidate path into the second type main road, to generate a target path; The target path is sent to the target vehicle.

2. The method according to claim 1, characterized in that The generating a plurality of candidate paths based on the entrance and exit information and the spatial orientation information comprises: Based on the entrance and exit information, the spatial orientation information and the kinematic characteristics of the work vehicle, a plurality of initial paths are generated, wherein each initial path includes a first type of initial lane for connecting a corresponding work position with the entrance position and a second type of initial lane for connecting a corresponding work position with the exit position; In the case where there is a conflicting area between any two paths among the multiple initial paths, adjusting the multiple initial paths to obtain the multiple candidate paths; In a case where there is no conflicting area between any two paths among the multiple initial paths, determining the multiple initial paths as the multiple candidate paths; The conflict area is an area where different working vehicles conflict when traveling on any two paths.

3. The method according to claim 2, characterized in that The adjusting the multiple initial paths to obtain the multiple candidate paths includes: Matching any two of the multiple initial paths to determine at least one first conflicting path combination among the multiple initial paths, wherein a conflicting region exists between two paths in the same first conflicting path combination; Adjusting the at least one first conflicting path combination to obtain a plurality of adjusted paths; If there is a conflicting area between any two paths among the multiple adjusted paths and the number of iterations does not reach a preset number, match any two initial paths among the multiple adjusted paths, determine at least one second conflicting path combination among the multiple adjusted paths, and adjust the at least one second conflicting path combination to obtain a path in the next adjustment process; When there is no conflicting area between any two of the multiple adjusted paths, or when the number of iterations reaches the preset number, the multiple adjusted paths are used as the multiple candidate paths.

4. The method according to claim 3, characterized in that The adjusting the at least one first conflicting path combination to obtain a plurality of adjusted paths includes: Obtaining a target conflicting path combination according to the conflicting path combination corresponding to the largest conflicting area in the at least one first conflicting path combination; Adjusting two paths in the target conflicting path combination respectively to obtain two initial adjustment results; Determining a target adjustment result from the two initial adjustment results based on the evaluation indicators of the two initial adjustment results; Two paths in the target conflict path combination are adjusted based on the target adjustment result to obtain the plurality of adjusted paths.

5. The method according to claim 4, characterized in that The two paths in the target conflicting path combination are adjusted respectively to obtain two initial adjustment results, including: For any path, the path is adjusted according to a plurality of adjustment steps to obtain a plurality of candidate adjustment results; determining conflict areas between different candidate adjustment results and another path; The initial adjustment result is obtained according to the candidate adjustment result corresponding to the smallest conflicting area among the multiple candidate adjustment results.

6. The method according to any one of claims 1 to 5, characterized in that The first type of candidate lanes include driving lanes and turning lanes, and the first type of main road is determined based on the first type of candidate lanes in the plurality of candidate paths, and the second type of main road is determined based on the second type of candidate lanes in the plurality of candidate paths, including: Determine, from the driving lanes included in the plurality of candidate paths, a driving lane with the longest length as the first type of main road; From the second type candidate lanes included in the plurality of candidate paths, determine the candidate lane with the longest length as the second type main road.

7. The method according to claim 6, characterized in that The step of merging at least a portion of the first type candidate lanes in each candidate path into the first type main road, and merging at least a portion of the second type candidate lanes in each candidate path into the second type main road, to generate a target path includes: Determine a first bifurcation point between the first type candidate lane and the first type main road, and determine a second bifurcation point between the second type candidate lane and the second type main road; Sampling the first type candidate lane to generate a first connecting path from the first bifurcation point to the first type candidate lane, and sampling the second type candidate lane to generate a second connecting path from the second bifurcation point to the first type candidate lane, wherein the intersection point between the first connecting path and the first type candidate lane is a first target sampling point, and the intersection point between the second connecting path and the second type candidate lane is a second target sampling point; Splicing the road section before the first bifurcation point, the first connecting path, and the road section after the first target sampling point in the first type candidate lane of the first type to generate a first type target lane, and splicing the road section before the second bifurcation point, the second connecting path, and the road section after the second target sampling point in the second type candidate lane of the second type to generate a second type target lane, wherein the first type target lane and the second type target lane meet a preset condition; The target path is obtained based on the first type target lane and the second type target lane.

8. The method according to claim 7, characterized in that The determining of a first bifurcation point between the first type candidate lane and the first type main road, and the determining of a second bifurcation point between the second type candidate lane and the second type main road, comprises: Based on the type of the working vehicle, construct a first buffer area corresponding to the first type of candidate lane and a second buffer area corresponding to the second type of candidate lane; Determine at least one first intersection point between the first type of main road and the first buffer area, and at least one second intersection point between the second type of main road and the second buffer area; The first bifurcation point is determined from the at least one first intersection, and the second bifurcation point is determined from the at least one second intersection, wherein a distance between the first bifurcation point and the first type of candidate lane is greater than a distance between an intersection other than the first bifurcation point and the first type of candidate lane, and a distance between the second bifurcation point and the second type of candidate lane is greater than a distance between an intersection other than the second bifurcation point and the second type of candidate lane.

9. The method according to claim 7, characterized in that: The step of sampling the first type candidate lane to generate a first connecting path from the first bifurcation point to the first type candidate lane, and sampling the second type candidate lane to generate a second connecting path from the second bifurcation point to the first type candidate lane includes: Determine a first target path point in the first type candidate lane that is closest to the first bifurcation point, and a second target path point in the second type candidate lane that is closest to the second bifurcation point; Sampling is performed on the first type candidate lane starting from the first target path point to obtain a plurality of first sampling points, and sampling is performed on the second type candidate lane starting from the second target path point to obtain a plurality of second sampling points; Based on the kinematic characteristics of the vehicle, sequentially connect the first sampling point and the first bifurcation point to obtain a plurality of first reference paths, and sequentially connect the second sampling point and the second bifurcation point to obtain a plurality of second reference paths; The first connecting path is obtained at least according to the shortest path among the plurality of first reference paths, and the second connecting path is obtained according to the shortest path among the plurality of second reference paths.

10. A path planning method, characterized in that: include: Receive a target path, wherein the target path is generated by merging at least a portion of a first type candidate lane in each candidate path among a plurality of candidate paths into a first type main road, and merging at least a portion of a second type candidate lane in each candidate path into a second type main road, the first type main road is determined based on the first type candidate lane in the plurality of candidate paths, the second type main road is determined based on the second type candidate lane in the plurality of candidate paths, the plurality of candidate paths are generated based on entrance and exit information of a work area and spatial orientation information of a plurality of work positions, the entrance and exit information includes an entrance position and an exit position of the work area, and each candidate path includes a first type candidate lane for connecting a corresponding work position with the entrance position and a second type candidate lane for connecting a corresponding work position with the exit position; Follow the target path as described.

11. A vehicle, characterized in that: include: A communication unit, used for communicating with a cloud server or a target vehicle; A memory storing an executable program; A processor, configured to run the program, wherein the program executes the method according to any one of claims 1 to 10 when running.

12. An electronic device, characterized in that: include: a communication unit, for communicating with a target vehicle; A memory storing an executable program; A processor, configured to run the program, wherein the program executes the method according to any one of claims 1 to 9 when running.

Images