Path planning method, system and equipment and storage medium

Abstract

The invention discloses a path planning method, system and device and a storage medium, and belongs to the technical field of mobile robots. The method comprises the following steps: acquiring navigation information and a local map of a plurality of robots in the same scene, wherein the navigation information at least comprises ontology attribute parameters; performing fusion processing on the plurality of local maps to generate a scene fusion map, and establishing a coordinate mapping relationship between the local map of each robot and the scene fusion map; based on the scene fusion map, the coordinate mapping relation and the navigation information, an initial path set is obtained through calculation; performing space-time conflict analysis on the initial path set based on the ontology attribute parameters; when a path conflict is detected, determining a target conflict-free path scheme based on the plurality of path adjustment branches; and converting the target conflict-free path scheme into a local path of each robot through the coordinate mapping relation. According to the embodiment of the invention, the overall execution efficiency of multiple robots can be effectively improved.

Description

Technical Field

[0001] This application relates to the field of mobile robot technology, and in particular to a path planning method, system, device and storage medium. Background Technology

[0002] Path planning is a key technology in the field of mobile robotics. It is defined as the process of autonomously calculating the optimal or feasible trajectory from a starting position to a target position in an environment with obstacles, based on a predetermined task objective. When multiple robots perform collaborative work in the same shared environment, this technology expands to multi-robot path planning. In this case, the goal is not only to plan collision-free paths for each robot, but also to optimize overall operational efficiency through system-level motion coordination, and effectively avoid system deadlock and path congestion. Currently, existing multi-robot path planning schemes generally suffer from low execution efficiency. Summary of the Invention

[0003] The purpose of this application is to provide a path planning method, system, device, and storage medium that can effectively improve the overall execution efficiency of multiple robots.

[0004] To achieve the above objectives, a first aspect of this application provides a path planning method, comprising: Obtain navigation information and local map of multiple robots in the same scene, wherein the navigation information includes at least ontological attribute parameters; Multiple local maps are fused to generate a scene fusion map, and a coordinate mapping relationship is established between each robot's local map and the scene fusion map; Based on the scene fusion map, the coordinate mapping relationship, and the navigation information, an initial path set is calculated; Spatiotemporal conflict analysis is performed on the initial path set based on the ontology attribute parameters; When a path conflict is detected, the branches are adjusted based on multiple paths to determine a conflict-free path to the target. The target conflict-free path scheme is converted into the local path for each robot through the coordinate mapping relationship.

[0005] Compared with existing technologies, the path planning method provided in this application has the following advantages: by integrating local maps of multiple robots to establish a unified scene fusion map and constructing accurate coordinate mapping relationships, the problem of inconsistent environmental representation caused by perceptual differences among heterogeneous robots is effectively solved; by introducing a dynamic path adjustment mechanism based on spatiotemporal conflict analysis, a multi-branch evaluation strategy is adopted when path conflicts are detected, which significantly improves the collaborative efficiency and robustness of path planning of the multi-robot system; finally, by using the reverse transformation of coordinate mapping, the planned path can be accurately distributed to each robot for execution, thereby expanding the applicable scenarios of the system while optimizing the overall navigation efficiency and reducing the risk of congestion.

[0006] In some embodiments, the step of fusing multiple local maps to generate a scene fused map and establishing a coordinate mapping relationship between each robot's local map and the scene fused map includes: The resolution of multiple local maps is standardized, and the resolution adjustment parameters are recorded to obtain a candidate map set. A reference map is selected from the candidate map set, and the coordinate transformation matrix between each candidate map and the reference map is calculated; Based on the coordinate transformation matrix, each local map is mapped to the same coordinate system to form a set of mapped maps; Image fusion is performed on the mapped map set to generate the scene fusion map; Based on the coordinate transformation matrix and the resolution adjustment parameters, the coordinate mapping relationship between the local map of each robot and the scene fusion map is calculated.

[0007] In some embodiments, calculating the coordinate transformation matrix between each candidate map and the reference map includes: Feature matching algorithms are used to extract and match environmental structure features in each candidate map; Based on the matched environmental structure features, the coordinate transformation matrix of each candidate map relative to the reference map is obtained.

[0008] In some embodiments, performing spatiotemporal conflict analysis on the initial path set based on the ontology attribute parameters includes: Based on the body attribute parameters of each robot, calculate its motion trajectory when moving along the initial path; Based on the motion trajectory and the body size model in the body attribute parameters, calculate the spatial occupancy area of ​​the robot at each future moment; Detect whether the spatial occupancy areas of any different robots overlap at the same time.

[0009] In some embodiments, determining a target conflict-free path scheme based on adjusting branches of multiple paths when a path conflict is detected includes: When a path conflict is detected, multiple path adjustment branches are generated, with different path adjustment branches assigning different path passing priorities to the robot involved in the conflict. For each path, adjust the branch and, based on the path priority, establish a branching scheme that keeps the initial path of the high-priority robot unchanged and replans the path for the low-priority robot to eliminate path conflicts. Evaluate the total cost for all robots to complete the navigation task under each path adjustment branch; Choose the branch with the lowest total cost as the target conflict-free path.

[0010] In some embodiments, after obtaining navigation information and local maps of multiple robots in the same scene, the method further includes: Determine whether a scene fusion map has been constructed for the given scene; If it does not exist, a new scene record is created, and the local map of the first robot is directly used as the scene fusion map; If they exist, the multiple local maps are merged to update the scene fusion map.

[0011] In some embodiments, the method further includes: During the robot's navigation according to the local path, the following operations are performed on each robot: Regularly obtain the real-time location information of each robot; Based on the real-time location information and the scene fusion map, the initial path set of each robot is recalculated; Based on the aforementioned ontology attribute parameters, a new spatiotemporal conflict analysis is performed on the recalculated initial path set. When a path conflict is detected, the branches are adjusted based on multiple paths to redetermine a conflict-free target path solution; Based on the coordinate mapping relationship, the newly determined target conflict-free path scheme is converted into an updated local path and sent to the corresponding robot.

[0012] To achieve the above objectives, a second aspect of this application provides a path planning system that applies the path planning method described in the first aspect above. The path planning system includes a path planning module, an information processing and display module, and a map fusion module. The information processing and display module is connected to the path planning module and the map fusion module, respectively. The information processing and display module is used to acquire navigation information and local maps of multiple robots in the same scene. The navigation information includes at least ontological attribute parameters. The map fusion module is used to fuse multiple local maps to generate a scene fusion map and establish a coordinate mapping relationship between each robot's local map and the scene fusion map. The path planning module is used to calculate an initial path set based on the scene fusion map, the coordinate mapping relationship, and the navigation information; perform spatiotemporal conflict analysis on the initial path set based on the ontology attribute parameters; and determine a conflict-free target path scheme based on multiple path adjustment branches when a path conflict is detected. The information processing and display module is also used to convert the target conflict-free path scheme into the local path of each robot through the coordinate mapping relationship.

[0013] To achieve the above objectives, a third aspect of this application provides an electronic device, the electronic device including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein the processor executes the computer program to implement the method described in the first aspect.

[0014] To achieve the above objectives, a fourth aspect of the present application provides a computer-readable storage medium comprising a stored computer program, wherein the computer program, when executed, controls the device containing the computer-readable storage medium to perform the method described in the first aspect. Attached Figure Description

[0015] Figure 1 This is a flowchart of a path planning method provided in an embodiment of this application; Figure 2 This is a schematic diagram of the path planning system provided in an embodiment of this application; Figure 3 This is a schematic diagram of the hardware structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0016] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0017] In the description of this application, it should be understood that the terms "center", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this application.

[0018] The terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this application, unless otherwise stated, "a plurality of" means two or more.

[0019] In the description of this application, it should be noted that, unless otherwise expressly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection between two components. Those skilled in the art can understand the specific meaning of the above terms in this application based on the specific circumstances.

[0020] First, let's analyze some of the terms used in this application: Simultaneous Localization and Mapping (SLAM) is a technology that enables mobile devices to acquire environmental data through their own sensors in unknown environments, while simultaneously estimating their own position and orientation and building a map of the environment. This technology is widely used in mobile robots, autonomous driving, drones, and other fields, providing fundamental support for autonomous navigation in scenarios without pre-defined maps.

[0021] Automated Guided Vehicle (AGV): An automated transportation device that does not require human driver and relies on a preset guidance path (such as electromagnetic track, optical navigation, visual navigation, etc.) to autonomously complete tasks such as cargo handling and material transfer.

[0022] Path planning is a key technology in the field of mobile robotics. It refers to the technology of autonomously calculating the optimal or feasible trajectory from the starting position to the target position in an environment with obstacles, based on the task objective. Mobile robots typically use SLAM technology to complete localization, map building, and navigation, thereby achieving path planning. When multiple robots perform collaborative operations in the same shared environment, this technology extends to multi-robot path planning. In this case, a multi-robot path planning system is needed to coordinate the navigation routes of each robot, reduce conflicts and waiting time, and thus improve the overall navigation efficiency. A typical example is in the scenario of goods sorting in a logistics center, where reasonable path planning can ensure that goods are delivered to the designated location quickly and accurately.

[0023] Current multi-robot path planning has developed into various technical solutions. One approach involves setting up identification nodes along the navigation path to delineate traffic control zones and adjacent areas. Robots upload node control information in real time during their journey, which is then analyzed by a host computer to issue scheduling commands to avoid congestion and collisions. However, this approach relies on pre-defined fixed scenarios and a unified traffic map, resulting in weak generalization capabilities and an inability to adapt to heterogeneous robots with different path planning methods and traffic demands. Another approach calculates the shortest path and total travel time for each robot, using the total time to determine the movement priority when they meet. Priority adjustment is then used to achieve avoidance. However, this approach uses a fixed shortest path and static priority avoidance mechanism, lacking dynamic flexibility in path planning. Frequent robot encounters can easily lead to task blocking, making it difficult to achieve global optimization. A third approach utilizes SLAM technology to construct an initial global grid map. After preprocessing, image filtering, and topology node optimization, it combines specific algorithms to perform global path planning to mitigate path conflicts. However, the map construction stage does not consider differences in robot type and size, only supporting robots of the same type and size, and cannot meet the needs of heterogeneous scenarios.

[0024] Based on this, embodiments of this application provide a path planning method, system, device, and storage medium, which can effectively improve the overall execution efficiency of multiple robots.

[0025] It should be noted that the path planning method provided in this application is applied to a path planning system. This system includes a map fusion module, a path planning module, and an information processing and display module. The information processing and display module externally interfaces with various robots in the same scene, handling information collaboration among multiple robots; internally, it coordinates the map fusion module and the path planning module, completing the sending, receiving, conversion, and storage of information. The map fusion module is mainly responsible for processing grid map-related information, fusing the grid maps of all robots in the scene, and recording the coordinate transformation relationship between the scene fusion map and each robot's local grid map. The path planning module focuses on robot navigation information processing, relying on the aforementioned scene fusion map and coordinate transformation relationship, and uses an overall path planning algorithm to plan the optimal navigation route for robots in the same scene, while ensuring that collisions do not occur during robot operation.

[0026] Please see Figure 1 , Figure 1 This is an optional flowchart of the path planning method provided in the embodiments of this application. Figure 1 The method may include, but is not limited to, steps S101 to S106.

[0027] Step S101: Obtain navigation information and local map of multiple robots in the same scene. The navigation information shall include at least the body attribute parameters. Step S102: Perform fusion processing on multiple local maps to generate a scene fusion map, and establish the coordinate mapping relationship between the local maps of each robot and the scene fusion map; Step S103: Calculate the initial path set based on the scene fusion map, coordinate mapping relationship, and navigation information; Step S104: Perform spatiotemporal conflict analysis on the initial path set based on ontology attribute parameters; Step S105: When a path conflict is detected, determine the target conflict-free path scheme based on adjusting branches of multiple paths. Step S106: Convert the target conflict-free path scheme into the local path of each robot through coordinate mapping relationship.

[0028] Steps S101 to S106 as illustrated in this embodiment of the application establish a unified scene fusion map by fusing local maps of multiple robots and constructing precise coordinate mapping relationships, effectively solving the problem of inconsistent environmental representation caused by perceptual differences among heterogeneous robots. By introducing a dynamic path adjustment mechanism based on spatiotemporal conflict analysis, a multi-branch evaluation strategy is adopted when path conflicts are detected, significantly improving the collaborative efficiency and robustness of path planning of the multi-robot system. Finally, the reverse transformation of coordinate mapping ensures that the planned path can be accurately distributed to each robot for execution, thereby expanding the applicable scenarios of the system while optimizing the overall navigation efficiency and reducing the risk of congestion.

[0029] In step S101 of some embodiments, the same scene can be a unified physical or virtual space for multiple robots to work collaboratively (such as a logistics warehouse, workshop, or a quantum computing task scheduling virtual environment), with consistent environmental features within the space (such as walls, obstacles, and virtual task nodes). Navigation information can be the data required for the robot to complete navigation, including at least ontological attribute parameters, and may also include the current scene number, robot identification code, starting position, target position, real-time position, motion parameters, etc., which may include velocity, acceleration, etc. The local map is the robot's ontological grid map, an environmental map adapted to its own coordinate system constructed by the robot using its own sensors (such as LiDAR) combined with SLAM technology, recording scene obstacles and passable areas. In a virtual robot scene, the local map represents task nodes and execution links. Ontological attribute parameters can be the robot's inherent key attributes, including size model, compatible computing device type, and external structure, where the size model covers geometric parameters such as the robot's length and width, and the compatible computing device type includes classical computers and quantum computers.

[0030] It should be noted that the type of robot referred to in the embodiments of this application is not limited to physical robots, but also includes virtual robots. In some embodiments, the robot is specifically applied to the path planning scheme of quantum computers in the form of a virtual robot. In this case, the size model in the ontology attribute parameters corresponds to the computing resource quota occupied by the virtual task, and the real-time position in the navigation information corresponds to the execution progress node of the virtual task.

[0031] Specifically, in step S101, the robot constructs a local map using its own sensors combined with SLAM technology and collects navigation information. It then uploads the local map and navigation information to the system's information processing and display module via wireless transmission and other communication modules. The information processing and display module receives this data, ensuring its real-time performance and integrity, and simultaneously reflects the robot's heterogeneous characteristics through its ontological attribute parameters.

[0032] In step S102 of some embodiments, the fusion process can be the process of unifying the local maps of multiple heterogeneous robots into a globally consistent map, including operations such as resolution unification, feature matching, coordinate calibration, and image fusion. The scene fusion map can be a unified global map generated based on the fusion of the local maps of all robots, eliminating differences in coordinate systems and resolutions between different local maps, and providing a unified environmental benchmark for multi-robot collaborative planning. The coordinate mapping relationship can be a mathematical relationship (such as a coordinate transformation matrix and its inverse matrix) describing the transformation rules between the robot's local map coordinate system and the scene fusion map coordinate system, enabling interoperability of positional data between the two types of maps.

[0033] In an optional embodiment, after obtaining navigation information and a local map of multiple robots in the same scene, the method further includes: Determine if a scene fusion map already constructed exists in the scene; If it does not exist, a new scene record will be created, and the local map of the first robot will be used directly as the scene fusion map; If present, merge multiple local maps to update the scene fusion map.

[0034] In this embodiment, taking a robot (e.g., robot R1) as an example, the navigation information and local map it uploads include the robot's identification number (e.g., SN code), the current work scene number (e.g., office number 1 is W1), and its own constructed ontological grid map M1 (containing grayscale image files and map parameter files). The grayscale image file can be in map.pgm format, and the map parameter file can be in map.yaml format. This map parameter file records information such as map resolution, map origin, and grayscale thresholds for obstacles and free space. After receiving this information, the information processing and display module first determines whether a scene fusion map has been constructed for the scene, i.e., it checks whether a record corresponding to the scene number is stored in the system.

[0035] If the inspection determines that there is no existing scene fusion map for the scene, the system will create a new record for the scene. At the same time, the map fusion algorithm will not be enabled, and the robot R1's body grid map M1 will be used directly as the scene fusion map for the scene. The coordinate transformation matrix will not be generated for the time being.

[0036] If the inspection confirms that a scene fusion map already exists for the scene, the system will add the relevant information of robot R1 to the scene's record, and then send the local map information of all existing robots in the scene to the map fusion module. The map fusion algorithm will then be used to fuse these local maps, ultimately completing the update of the current scene fusion map.

[0037] In an optional embodiment, multiple local maps are fused to generate a scene fused map, and a coordinate mapping relationship between each robot's local map and the scene fused map is established, including: Multiple local maps are subjected to resolution normalization, and the resolution adjustment parameters are recorded to obtain a set of candidate maps; Select a reference map from the candidate map set and calculate the coordinate transformation matrix between each candidate map and the reference map; Based on the coordinate transformation matrix, each local map is mapped to the same coordinate system to form a set of mapped maps; Image fusion is performed on the mapped map set to generate a scene fusion map; Based on the coordinate transformation matrix and resolution adjustment parameters, the coordinate mapping relationship between the local map of each robot and the scene fusion map is calculated.

[0038] Specifically, when the system determines that a scene fusion map already exists in the current scene, the map fusion module receives the local map (i.e., the body grid map of each robot) information of all robots in the scene sent by the information processing and display module, and then executes the fusion processing flow: First, resolution normalization processing is performed, and the grid map M with the lowest resolution is determined from all local maps. min A cubic convolution interpolation scheme was adopted to uniformly adjust the resolution of the body grid map of all robots to be consistent with M. min The resolution adjustment parameters are recorded to form a candidate map set.

[0039] Subsequently, the largest grid map in the candidate map set is selected as the reference map M0. Then, the edge detection algorithm is used to extract the contour features of environmental structures such as door frames, walls, and pillars in each candidate map. Finally, the feature matching algorithm is used to match these features and establish the correspondence between different candidate maps.

[0040] To quantify the coordinate transformation relationships corresponding to these matching features, this embodiment will employ a general affine transformation model, the expression of which is: ; Among them, the model is modeled after the general affine transformation. The relevant definition, T= These are the target coordinates in the M0 coordinate system of the transformed reference map; X= These are the source coordinates in the original candidate map coordinate system; A= It is a 2×2 affine transformation matrix, describing operations such as rotation, scaling, and shearing; b= It is a 2×1 translation vector that describes the translation operation.

[0041] However, directly using this general model for solving the problem may lead to overfitting due to small deviations in feature points, resulting in extreme values ​​in the transformation parameters and affecting the smoothness of map fusion. Therefore, an L2 regularization term is introduced into the objective function, specifically: ; in, It is the regularization coefficient. It is the Frobennius norm, used to measure the relationship between an affine transformation matrix and the identity matrix. The distance between them ensures that the transformation is as close as possible to a translation. It is the L2 norm of translation to prevent excessive translation.

[0042] Finally, the optimized objective function integrates the fitting error of feature points and the regularization constraint, and its expression is: ; The objective function is then optimized using the least squares method to obtain the coordinate transformation matrix of each candidate map relative to the reference map M0.

[0043] Next, based on the calculated coordinate transformation matrix, a coordinate transformation operation is performed on each point on each candidate map to map all local maps to the same coordinate system corresponding to the reference map M0, forming a mapped map set. Then, image fusion is performed on this mapped map set using a pixel-value-based weighted average method to output a scene fusion map M. x .

[0044] Finally, combining the coordinate transformation matrix obtained earlier with the resolution adjustment parameters (such as interpolation scaling ratio) recorded during resolution standardization, the inverse operation of resolution adjustment is embedded into the coordinate transformation inverse matrix. The final coordinate transformation matrix and corresponding inverse matrix of each robot's original local map to the scene fusion map are calculated. These matrices represent the coordinate mapping relationship between each robot's local map and the scene fusion map, and are sent to the information processing and display module for storage.

[0045] In an optional embodiment, calculating the coordinate transformation matrix between each candidate map and the reference map includes: Feature matching algorithms are used to extract and match environmental structure features in each candidate map; Based on the matched environmental structure features, the coordinate transformation matrix of each candidate map relative to the reference map is obtained.

[0046] Specifically, a feature matching algorithm is first used to extract and match environmental structural features. First, an edge detection algorithm is used to extract the contour features of fixed environmental structures such as door frames, walls, and pillars in each candidate map. Then, the feature matching algorithm is used to match these cross-map structural features, establishing feature associations between each candidate map and the reference map. Subsequently, based on these matched environmental structural features, using the coordinate system of the reference map as a reference, an affine transformation model with L2 regularization is used to solve for the coordinate transformation matrix. The objective function, including the regularization term, is optimized using the least squares method, ensuring the accuracy of the transformation while avoiding extreme values ​​of transformation parameters caused by overfitting. Finally, the coordinate transformation matrix of each candidate map relative to the reference map is obtained, providing a mathematical basis for subsequently mapping all candidate maps to the same coordinate system.

[0047] In this embodiment, the system performs resolution unification, feature matching, and smooth constraint fusion on the local maps of each heterogeneous robot to generate a globally consistent scene fusion map, while establishing coordinate mapping relationships; this solves the problem of inconsistent local maps of heterogeneous robots, provides a unified environmental benchmark for subsequent unified path planning, and improves planning compatibility.

[0048] In step S103 of some embodiments, the initial path set can be a set of initial motion trajectories planned in the scene fusion map for each robot without considering multi-robot conflicts, with each robot corresponding to an initial path.

[0049] Specifically, the starting position, target position, and real-time position in the navigation information of each robot are first converted into a scene fusion map M through coordinate mapping relationships. x Coordinates in a coordinate system; combined with the dimensional model in the robot's body attribute parameters in M x A safe passage space is reserved in the middle, and dynamic constraints on the path are set with reference to motion parameters. Finally, path planning algorithms such as heuristic search are used to calculate the initial path from the starting position to the target position for each robot, forming an initial path set.

[0050] In step S104 of some embodiments, spatiotemporal conflict analysis can be used to perform cross-verification of the motion trajectories of each robot in the initial path set in terms of time and space dimensions to determine whether there is a conflict of "different robots occupying the same area at the same time".

[0051] Based on the size model in the robot's body attribute parameters, combined with the time scale of the initial path (such as the position at each moment), the area occupied by each robot at different times is estimated. By comparing the occupied areas of all robots at each moment, if there is overlap, it is determined to be a path conflict. This provides an accurate basis for subsequent conflict handling and avoids collisions or congestion during robot operation.

[0052] In an optional embodiment, spatiotemporal conflict analysis is performed on the initial path set based on ontology attribute parameters, including: Based on the body attribute parameters of each robot, calculate its motion trajectory when moving along the initial path; Based on the motion trajectory and the body size model in the body attribute parameters, calculate the space occupied by the robot at each future time. Detect whether the spatial occupancy areas of any different robots overlap at the same time.

[0053] Specifically, when performing spatiotemporal conflict analysis, the path planning module first receives the scene fusion map, coordinate mapping relationship, and navigation information of each robot output by the map fusion module, and conducts analysis based on the initial path set: First, combining the motion parameters in the navigation information with the path point list corresponding to the initial path, which is a list of continuous point data from the robot's current position to the target position calculated by a heuristic search algorithm, the system will estimate the robot's motion capability in units of time, and transform the discrete path points into a continuous motion trajectory that changes dynamically over time, ensuring that the trajectory matches the robot's actual motion capability.

[0054] Subsequently, based on the size model in the motion trajectory and body attribute parameters, the spatial occupancy area of ​​the robot at each future moment is calculated: Since the robot has an actual volume, the system will take the position of the robot on the motion trajectory at each moment as the center, and combine its length, width and other size parameters to outline the rectangular or irregular space area occupied by the robot at that moment, forming the correspondence data of "time-space occupancy area".

[0055] Finally, the system extracts the "time-space occupied area" data of all robots in the same scene, compares the occupied areas of any two different robots at the same time, and determines whether there is any overlap. If the occupied areas of all robots do not overlap at any time, it means that there is no spatiotemporal conflict in the initial path set, and the path point list of each robot can be directly output. If it detects that the occupied areas of at least two robots overlap at a certain time, it is determined that there is a path conflict, and the subsequent path conflict handling process needs to be triggered. The entire process is based on the robot's own characteristics and dynamic motion state, realizing accurate and dynamic identification of path conflicts, and providing a clear basis for subsequent conflict resolution.

[0056] In step S105 of some embodiments, the path adjustment branches can be different path modification schemes designed for the conflict robot, with each branch corresponding to a conflict resolution strategy. The target conflict-free path scheme can be the final path scheme selected from multiple path adjustment branches that satisfies the requirements of no spatiotemporal conflict and takes into account the efficiency of global planning.

[0057] After detecting a path conflict, the system replans the path and selects the optimal solution through a multi-branch adjustment strategy. This resolves the spatiotemporal conflict problem of the initial path and ensures global planning efficiency through multi-branch selection, avoiding task blocking caused by single avoidance.

[0058] In an optional embodiment, when a path conflict is detected, determining a target conflict-free path solution based on adjusting branches of multiple paths includes: When a path conflict is detected, multiple path adjustment branches are generated, with different path adjustment branches assigning different path passing priorities to the robot involved in the conflict. For each path, adjust the branches according to the path priority. By establishing a branching scheme that keeps the initial path of high-priority robots unchanged and replans the path for low-priority robots, path conflicts are eliminated. Evaluate the total cost for all robots to complete the navigation task under each path adjustment branch; Choose the branch with the lowest total cost as the target conflict-free path.

[0059] Specifically, when the path planning module detects a path conflict, such as when the spatial occupancy regions Q1(t1) and Q2(t1) of robots R1 and R2 overlap at time t1, it generates multiple path adjustment branches to assign different path passage priorities to the robots involved in the conflict. For example, it can be split into two branches: one sets R1 to high priority and R2 to low priority, and the other sets R2 to high priority and R1 to low priority.

[0060] For each path adjustment branch, a branching scheme is executed to eliminate conflicts by maintaining the initial path of high-priority robots and replanning the path of low-priority robots: For the high-priority branch R1, its initial path point list P1 remains unchanged, and in the scene fusion map M... x In the process, the occupied area Q1(t1) of R1 at time t1 is marked as an obstacle and a new scene fusion map M is generated. x2 Then, replan the path for R2 to obtain a new list of path points P. x2 For the high-priority branch R2, its initial path P2 remains unchanged, and the area Q2(t1) occupied by R2 at time t1 is set as the obstacle to generate a new scene fusion map M. x1 Replanning the path of R1 yields P. x1 .

[0061] The new paths of each branch then need to be checked for conflicts: the list of path points in the branch is converted into the space occupied area at each time, and it is determined whether there is still overlap; if there is a conflict, the new priority branches are split and the above replanning operation is repeated until the occupied areas of all robots in the branch are not overlapping at any time, and a conflict-free path scheme under the branch is obtained.

[0062] After generating conflict-free path schemes for each branch, evaluate the total cost of all robots completing the navigation task under each branch. The total cost is evaluated using the total running time of each robot from its current position to the target position as the evaluation index, and the total time consumed by all robots within the branch is calculated.

[0063] Finally, the branch with the lowest total cost (total running time) is selected as the target conflict-free path solution, and the corresponding path point list is sent to the information processing and display module. Subsequently, the information processing and display module will use the saved coordinate transformation inverse matrix to convert the path into a path in the local coordinate system of each robot and send it down for execution, which not only eliminates path conflicts, but also ensures the optimal global efficiency of multi-robot navigation.

[0064] In step S106 of some embodiments, the local path can be a motion trajectory adapted to the robot's own local map coordinate system, which the robot can directly recognize and execute.

[0065] After the path planning module determines the conflict-free path scheme with the lowest total cost, the corresponding path, a list of path points based on the global coordinate system of the scene fusion map, is synchronized to the information processing and display module. Upon receiving this list, the information processing and display module uses the robot's identification code as an index to retrieve the coordinate mapping relationship of the corresponding robot stored in the previous map fusion stage, i.e., the inverse coordinate transformation matrix between the robot's local map and the scene fusion map.

[0066] Subsequently, the information processing and display module substitutes each global path point in the target conflict-free path into the inverse matrix for calculation, converting the points in the global coordinate system into coordinates adapted to the robot's local grid map coordinate system; these converted coordinates are integrated in the order of the path to form the robot's local path, that is, the list of path points in the local map coordinate system.

[0067] After the conversion is complete, the information processing and display module will send the local path to the corresponding robot. Upon receiving it, the robot will set it as its global path planning benchmark and start navigation in conjunction with its own local path planning capabilities. While the robot is not navigating, the system will continuously update the local path by repeatedly calling the path planning algorithm based on its periodically uploaded current position, until the robot sends a stop message indicating that navigation is complete. This ensures the accuracy and dynamic adaptability of the navigation process.

[0068] In an optional embodiment, the path planning method further includes: During the robot's navigation based on the local path, the following operations are performed on each robot: Regularly obtain the real-time location information of each robot; Based on real-time location information and scene fusion map, the initial path set of each robot is recalculated; Based on the ontology attribute parameters, a new spatiotemporal conflict analysis is performed on the recalculated initial path set; When a path conflict is detected, the branches are adjusted based on multiple paths to redetermine a conflict-free target path solution; Based on the coordinate mapping relationship, the newly determined target conflict-free path scheme is converted into an updated local path and sent to the corresponding robot.

[0069] Specifically, after the robot starts navigation based on the local path, the system will continuously perform periodic path update operations for each robot: First, the system will periodically obtain the real-time location information uploaded by each robot, and combine it with the previously established coordinate mapping relationship to convert the real-time location into coordinates under the global coordinate system of the scene fusion map, thereby correcting the path starting point offset caused by the actual movement deviation of the robot.

[0070] It should be noted that the default period for acquiring real-time location information for each robot is 5 seconds. This period can be flexibly configured according to the actual needs of the scenario and is not limited to the above default value.

[0071] Subsequently, based on the corrected real-time position, the original target position, and the scene fusion map, a new set of initial paths is generated for each robot. Next, the system re-performs spatiotemporal conflict analysis on the new set of initial paths based on the ontological attribute parameters: according to the size model of each robot, the spatial occupancy area at each moment when it moves along the new initial path is calculated, and it is detected whether the areas of any robot overlap at the same moment.

[0072] If a new path conflict is detected, the target conflict-free path scheme is re-determined according to the logic of generating multiple priority branches → maintaining high priority paths and replanning low priority paths → evaluating the total cost of branches → selecting the optimal solution.

[0073] Finally, the system uses coordinate mapping to convert the newly determined conflict-free target path into an updated local path in each robot's local coordinate system and sends it to the corresponding robot. The robot will then replace its original local path with the updated path and continue navigation. This periodic update process will continue until the robot sends a stop message indicating navigation completion, thus ensuring the real-time and conflict-free nature of the path during navigation and adapting to robot motion deviations and dynamic changes in the scene.

[0074] In summary, the embodiments of this application achieve the following technical effects through the above process: they can adapt to the differences in size, computing devices, etc. of heterogeneous robots, break the application limitations of fixed scenarios to improve the generalization ability of the solution; they can also avoid task blocking through dynamic multi-branch conflict adjustment strategies to achieve the optimal effect of global path planning; at the same time, relying on a unified fusion map and coordinate mapping relationship, they can effectively ensure the accuracy and efficiency of multi-robot collaborative navigation, and adapt to the needs of heterogeneous robot collaborative operation in multiple scenarios such as logistics and industry.

[0075] Please see Figure 2 This application embodiment also provides a route planning system that can implement the above-mentioned route planning method. The route planning system includes a route planning module 100, an information processing and display module 200, and a map fusion module 300. The information processing and display module 200 is connected to the route planning module 100 and the map fusion module 300, respectively. The information processing and display module 200 is used to acquire navigation information and local maps of multiple robots in the same scene. The navigation information includes at least the body attribute parameters. The map fusion module 300 is used to fuse multiple local maps, generate a scene fusion map, and establish the coordinate mapping relationship between the local maps of each robot and the scene fusion map. The path planning module 100 is used to calculate an initial path set based on the scene fusion map, coordinate mapping relationship and navigation information; perform spatiotemporal conflict analysis on the initial path set based on ontology attribute parameters; and determine a conflict-free target path scheme based on multiple path adjustment branches when a path conflict is detected. The information processing and display module 200 is also used to convert the target conflict-free path scheme into the local path of each robot through coordinate mapping relationship.

[0076] The specific implementation of this path planning system is basically the same as the specific implementation of the path planning method described above, and will not be repeated here.

[0077] Thirdly, embodiments of this application provide an electronic device, see [link to relevant documentation]. Figure 3 The diagram shown is a structural schematic of an electronic device provided in this application.

[0078] like Figure 3 As shown, the device includes: Memory 31 is used to store computer programs; Processor 32 is used to execute computer programs; When the processor 32 executes a computer program, it implements the path planning method as described in any of the above embodiments.

[0079] For example, a computer program may be divided into one or more modules / units, one or more of which are stored in memory 31 and executed by processor 32 to complete this application. One or more modules / units may be a series of computer program instruction segments capable of performing a specific function, which describe the execution process of the computer program in an electronic device.

[0080] The processor 32 may be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor.

[0081] The memory 31 can be used to store computer programs and / or modules. The processor 32 implements various functions of the electronic device by running or executing the computer programs and / or modules stored in the memory 31 and calling the data stored in the memory 31. The memory 31 may mainly include a program storage area and a data storage area. The program storage area may store the operating system, application programs required for at least one function (such as sound playback function, image playback function, etc.), etc.; the data storage area may store data created according to the use of the mobile phone (such as audio data, phonebook, etc.). In addition, the memory 31 may include high-speed random access memory, and may also include non-volatile memory, such as hard disk, RAM, plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, at least one disk storage device, flash memory device, or other volatile solid-state storage device.

[0082] It should be noted that the aforementioned electronic devices include, but are not limited to, processors and memory, as will be understood by those skilled in the art. Figure 3 The structural diagram is merely an example of the electronic device described above and does not constitute a limitation on the electronic device. It may include more components than shown in the diagram, or combine certain components, or use different components.

[0083] Fourthly, embodiments of this application also provide a computer-readable storage medium storing a computer program that, when executed, implements the path planning method of any of the above embodiments.

[0084] It should be understood that the implementation of all or part of the process in the above-described path planning method can also be accomplished by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the above-described path planning method. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or device capable of carrying computer program code, recording media, USB flash drives, portable hard drives, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc. It should be noted that the content contained in the computer-readable medium can be appropriately added or removed according to the requirements of legislation and patent practice in the relevant jurisdiction. For example, in some relevant jurisdictions, according to legislation and patent practice, computer-readable media do not include electrical carrier signals and telecommunication signals.

[0085] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. The storage medium can be a magnetic disk, optical disk, read-only memory (ROM), or random access memory (RAM), etc.

[0086] The above description is the preferred embodiment of this application. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of this application, and these improvements and modifications are also considered to be within the scope of protection of this application.

Claims

1. A path planning method, characterized in that, include: Obtain navigation information and local map of multiple robots in the same scene, wherein the navigation information includes at least ontological attribute parameters; Multiple local maps are fused to generate a scene fusion map, and a coordinate mapping relationship is established between each robot's local map and the scene fusion map; Based on the scene fusion map, the coordinate mapping relationship, and the navigation information, an initial path set is calculated; Spatiotemporal conflict analysis is performed on the initial path set based on the ontology attribute parameters; When a path conflict is detected, the branches are adjusted based on multiple paths to determine a conflict-free path to the target. The target conflict-free path scheme is converted into the local path for each robot through the coordinate mapping relationship.

2. The path planning method as described in claim 1, characterized in that, The process of fusing multiple local maps to generate a scene fusion map and establishing a coordinate mapping relationship between each robot's local map and the scene fusion map includes: The resolution of multiple local maps is standardized, and the resolution adjustment parameters are recorded to obtain a candidate map set. A reference map is selected from the candidate map set, and the coordinate transformation matrix between each candidate map and the reference map is calculated; Based on the coordinate transformation matrix, each local map is mapped to the same coordinate system to form a set of mapped maps; Image fusion is performed on the mapped map set to generate the scene fusion map; Based on the coordinate transformation matrix and the resolution adjustment parameters, the coordinate mapping relationship between the local map of each robot and the scene fusion map is calculated.

3. The path planning method as described in claim 2, characterized in that, The calculation of the coordinate transformation matrix between each candidate map and the reference map includes: Feature matching algorithms are used to extract and match environmental structure features in each candidate map; Based on the matched environmental structure features, the coordinate transformation matrix of each candidate map relative to the reference map is obtained.

4. The path planning method as described in claim 1, characterized in that, The spatiotemporal conflict analysis of the initial path set based on the ontology attribute parameters includes: Based on the body attribute parameters of each robot, calculate its motion trajectory when moving along the initial path; Based on the motion trajectory and the body size model in the body attribute parameters, calculate the spatial occupancy area of ​​the robot at each future moment; Detect whether the spatial occupancy areas of any different robots overlap at the same time.

5. The path planning method as described in claim 1, characterized in that, When a path conflict is detected, determining a conflict-free target path solution based on adjusting branches from multiple paths includes: When a path conflict is detected, multiple path adjustment branches are generated, with different path adjustment branches assigning different path passing priorities to the robot involved in the conflict. For each path, adjust the branch and, based on the path priority, establish a branching scheme that keeps the initial path of the high-priority robot unchanged and replans the path for the low-priority robot to eliminate path conflicts. Evaluate the total cost for all robots to complete the navigation task under each path adjustment branch; Choose the branch with the lowest total cost as the target conflict-free path.

6. The path planning method as described in claim 1, characterized in that, After obtaining navigation information and local maps of multiple robots in the same scene, the process also includes: Determine whether a scene fusion map has been constructed for the given scene; If it does not exist, a new scene record is created, and the local map of the first robot is directly used as the scene fusion map; If they exist, the multiple local maps are merged to update the scene fusion map.

7. The path planning method as described in claim 1, characterized in that, The method further includes: During the robot's navigation according to the local path, the following operations are performed on each robot: Regularly obtain the real-time location information of each robot; Based on the real-time location information and the scene fusion map, the initial path set of each robot is recalculated; Based on the aforementioned ontology attribute parameters, a new spatiotemporal conflict analysis is performed on the recalculated initial path set. When a path conflict is detected, the branches are adjusted based on multiple paths to redetermine a conflict-free target path solution; Based on the coordinate mapping relationship, the newly determined target conflict-free path scheme is converted into an updated local path and sent to the corresponding robot.

8. A path planning system, employing the path planning method as described in any one of claims 1 to 7, characterized in that, The route planning system includes a route planning module, an information processing and display module, and a map fusion module. The information processing and display module is connected to the route planning module and the map fusion module, respectively. The information processing and display module is used to acquire navigation information and local maps of multiple robots in the same scene. The navigation information includes at least ontological attribute parameters. The map fusion module is used to fuse multiple local maps to generate a scene fusion map and establish a coordinate mapping relationship between each robot's local map and the scene fusion map. The path planning module is used to calculate an initial path set based on the scene fusion map, the coordinate mapping relationship, and the navigation information; perform spatiotemporal conflict analysis on the initial path set based on the ontology attribute parameters; and determine a conflict-free target path scheme based on multiple path adjustment branches when a path conflict is detected. The information processing and display module is also used to convert the target conflict-free path scheme into the local path of each robot through the coordinate mapping relationship.

9. An electronic device, characterized in that, It includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein the processor, when executing the computer program, implements the path planning method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a stored computer program, wherein, when the computer program is executed, it controls the device on which the computer-readable storage medium is located to perform the path planning method as described in any one of claims 1 to 7.

Images