Robot path planning method and system, medium, product and computer equipment

Abstract

The invention belongs to the technical field of robot path planning. According to the robot path planning method and system, the medium, the product and the computer equipment, global path searching is carried out on a robot from a starting point to a target point, jump points are extracted, key nodes are screened, and an optimal global path is determined; taking a key node of the global optimal path as a sub-target point, performing speed sampling on the sub-target point, and generating a plurality of simulated motion tracks according to a speed sampling result; scoring each simulated motion trajectory based on the weighted sum of the course deviation, the obstacle distance, the motion speed and the trajectory smoothness, and selecting the trajectory with the maximum score as the optimal local trajectory; determining a walking path of the robot according to each optimal local track on the global optimal path; according to the invention, the accuracy and environmental adaptability of path planning are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of robot path planning, and in particular to a robot path planning method, system, medium, product and computer equipment. Background Art

[0002] The statements in this section merely provide background art related to the present invention and do not necessarily constitute prior art.

[0003] As the core technology of robot autonomous navigation, robot path planning has always been a hot topic in academia and industry. Although traditional path planning methods have achieved certain results, there are still many problems that need to be solved in practical applications: (1) Traditional algorithms rely on discrete map modeling, and maps need to be rebuilt or paths adjusted when the environment changes. For example, in logistics warehousing scenarios, the movement of goods or personnel will change the distribution of obstacles in real time, but the algorithm cannot dynamically perceive these changes, resulting in path planning failure or falling into local optimality. This static modeling method is inherently inconsistent with the dynamic environment, limiting the robot's autonomous navigation capabilities in actual scenarios. (2) Algorithms such as A* and Dijkstra need to traverse a large number of nodes to search for the optimal path. The computational complexity increases exponentially with the scale of the environment. In large and complex scenarios (such as urban delivery robots), there is a sharp contradiction between real-time requirements and the amount of algorithm computation. When the environment changes suddenly, traditional methods cannot respond quickly, resulting in path planning delays or even decision failures, which seriously affects the efficiency of task execution. Summary of the invention

[0004] In order to address the deficiencies in the prior art, the present invention provides a robot path planning method, system, medium, product and computer equipment, which deeply integrate global path planning with local dynamic obstacle avoidance technology, and overcome the challenges faced by traditional methods in dynamic and complex environments with the help of an adaptive parameter adjustment mechanism, thereby improving the accuracy and environmental adaptability of path planning.

[0005] In order to achieve the above object, the present invention adopts the following technical solution: In a first aspect, the present invention provides a robot path planning method.

[0006] A robot path planning method includes the following processes: Perform a global path search from the starting point to the target point of the robot, extract jump points and screen key nodes to determine the optimal global path; Taking the key nodes of the global optimal path as sub-target points, performing speed sampling on the sub-target points, and generating multiple simulated motion trajectories according to the speed sampling results; Each simulated motion trajectory is scored based on the weighted sum of heading deviation, obstacle distance, motion speed and trajectory smoothness, and the trajectory with the largest score is selected as the optimal local trajectory; The walking path of the robot is determined according to each optimal local trajectory on the global optimal path.

[0007] As a further limitation of the first aspect of the present invention, the node A critical jump point is determined when at least one of the following conditions is met: is the starting point or destination point; There is at least one mandatory neighbor; When the parent node of is expanded along the diagonal direction, Search for new jump points in horizontal or vertical directions; Horizontal or vertical jump search, including: if the current node is extended along the horizontal or vertical direction, the search continues to jump along the direction until an obstacle is encountered or a jump point is found. If the node ahead is unreachable during the jump process, the current node is the jump point; Diagonal jump search includes: if the current node is extended along the diagonal direction, the search is decomposed into horizontal and vertical jumps. During the search process, if the horizontal or vertical jump encounters an obstacle, the current node is determined to be a jump point and the expansion in this direction is terminated.

[0008] As a further limitation of the first aspect of the present invention, scoring is performed based on a weighted sum of heading deviation, obstacle distance, motion speed, and trajectory smoothness, including: ; in, is the heading deviation function; is the obstacle distance evaluation function; Indicates the sampling speed; is the trajectory smoothness function, which is used to quantify the robot's linear velocity at a given speed. and angular velocity The smoothness of the generated trajectory is is the heading weight, is the obstacle avoidance weight, is the speed weight, is the smoothness weight.

[0009] As a further limitation of the first aspect of the present invention, the obstacle distance evaluation function adopts an elliptical safety distance, and the obstacle distance evaluation function expression is: ; in,( , )、( , represent the current coordinates of the robot and the obstacle respectively. is the position after time ( , ) relative to the elliptical projection position of the robot, , , is the angle coefficient of the ellipse equation, is the angular position direction of the obstacle relative to the robot, is the forward direction angle of the dynamic obstacle. Within time, .

[0010] As a further limitation of the first aspect of the present invention, the trajectory smoothness function is: , where represents the instantaneous curvature of the trajectory generated under the given linear velocity and angular velocity , is a constant greater than zero.

[0011] As a further limitation of the first aspect of the present invention, the motion state of the robot, environmental information, elliptical safety distance parameters, and historical trajectory scores are obtained as input features of the multi-layer perceptron network to dynamically adjust the heading weight , obstacle avoidance weight , speed weight and smoothness weight .

[0012] In the second aspect, the present invention provides a robot path planning system.

[0013] A robot path planning system includes: A global path optimization unit, configured to: perform a global path search for the robot from the starting point to the target point, extract jump points and screen key nodes, and determine the optimal global path; A local path generation unit, configured to: use the key nodes of the global optimal path as sub-goal points, perform speed sampling on the sub-goal points, and generate multiple simulated motion trajectories according to the speed sampling results; A local path optimization unit, configured to: score each simulated motion trajectory based on the weighted sum of heading deviation, obstacle distance, motion speed, and trajectory smoothness, and select the trajectory with the maximum score as the optimal local trajectory; A walking path determination unit, configured to: determine the walking path of the robot according to each optimal local trajectory on the global optimal path.

[0014] In a third aspect, the present invention provides a computer device, comprising: a processor and a computer-readable storage medium; a processor adapted to execute a computer program; A computer-readable storage medium, wherein a computer program is stored in the computer-readable storage medium, and when the computer program is executed by the processor, the robot path planning method as described in the first aspect of the present invention is implemented.

[0015] In a fourth aspect, the present invention provides a computer-readable storage medium storing a computer program, wherein the computer program is suitable for being loaded by a processor and executing the robot path planning method as described in the first aspect of the present invention.

[0016] In a fifth aspect, the present invention provides a computer program product, which includes a computer program. When the computer program is executed by a processor, it implements the robot path planning method as described in the first aspect of the present invention.

[0017] Compared with the prior art, the present invention has the following beneficial effects: 1. The global path search efficiency of the present invention is high. It adopts the improved JPSA* algorithm and introduces the JPS jump point screening mechanism to expand only the key nodes, thereby greatly reducing the search space and being able to quickly generate the global optimal path.

[0018] 2. The present invention realizes real-time dynamic local obstacle avoidance and safety distance calculation. In the local obstacle avoidance stage, candidate trajectories are sampled in real time based on the improved DWA algorithm, and scored through a comprehensive evaluation function (including heading deviation, obstacle distance, movement speed and trajectory smoothness). Among them, the elliptical safety distance model is used to calculate the safety distance between the robot and the obstacle, and the major axis of the ellipse is determined according to the dynamic state of the obstacle, so that the robot can accurately judge the safety distance when avoiding obstacles and effectively avoid collisions.

[0019] 3. The present invention realizes the optimization of path smoothness. In the process of local dynamic obstacle avoidance, by adding a smoothness evaluation item based on trajectory curvature into the evaluation function, it is ensured that when selecting the obstacle avoidance trajectory, the generated motion trajectory has high smoothness, thereby improving motion stability and energy efficiency.

[0020] 4. The present invention realizes adaptive parameter tuning and adopts a multi-layer perceptron (MLP) network structure to perform online adaptive parameter tuning on various weights in the evaluation function (such as heading deviation weight, obstacle distance weight, speed weight and smoothness weight). It can dynamically adjust parameters according to the real-time environment and robot status, improve local planning and obstacle avoidance performance, and enhance system adaptability and robustness.

[0021] 5. The present invention organically integrates global path planning (based on the improved JPSA* algorithm) and local dynamic obstacle avoidance (improved DWA algorithm) to form a closed-loop planning process. First, the global algorithm is used to quickly generate the optimal path, and then the local algorithm is used to perform real-time obstacle avoidance and path optimization during the movement, thereby achieving efficient and safe overall path planning.

[0022] Advantages of additional aspects of the present invention will be given in part in the following description, and in part will become obvious from the following description, or will be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] The accompanying drawings in the specification, which constitute a part of the present invention, are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute improper limitations on the present invention.

[0024] Figure 1 A schematic diagram of a flow chart of a robot path planning method provided in Embodiment 1 of the present invention; Figure 2 Schematic diagram of JPS jump provided in Example 1 of the present invention; wherein (a) is a schematic diagram of the first search process, (b) is a schematic diagram of the second search process, (c) is a schematic diagram of the third search process, and (d) is a schematic diagram of the fourth search process; Figure 3 A schematic diagram of the elliptical distance between the robot and the obstacle provided in Example 1 of the present invention; Figure 4 A schematic diagram of the network architecture of a multi-layer perceptron provided in Example 1 of the present invention; Figure 5 A schematic diagram of a robot path planning system provided in Embodiment 2 of the present invention; Figure 6 A schematic diagram of a computer device provided in Example 3 of the present invention. DETAILED DESCRIPTION

[0025] The present invention will be further described below in conjunction with the accompanying drawings and embodiments.

[0026] It should be noted that the following detailed descriptions are exemplary and are intended to provide further explanation of the present invention. Unless otherwise specified, all technical and scientific terms used herein have the same meanings as those commonly understood by those skilled in the art to which the present invention belongs.

[0027] In the absence of conflict, the embodiments of the present invention and the features of the embodiments may be combined with each other.

[0028] Embodiment 1: This implementation proposes a robot path planning method, which aims to improve the path planning efficiency and obstacle avoidance ability of the robot in a complex dynamic environment. This method optimizes the A* algorithm by introducing the Jump Point Search algorithm (JPS) to achieve efficient global path search, and performs local dynamic obstacle avoidance and path smoothing optimization through the improved Dynamic Window Approach (DWA) to ensure the safety and feasibility of the path. At the same time, the elliptical safety distance model is introduced to improve the obstacle avoidance accuracy, and the Multi-Layer Perceptron (MLP) is used to achieve adaptive optimization of local planning parameters to enhance the real-time and adaptability of the algorithm. This method is suitable for robot navigation and path planning tasks in the fields of intelligent warehousing, logistics and distribution, and improves the operating efficiency and safety of the system.

[0029] In this implementation, the Jump Point Search (JPS) algorithm is used as an effective optimization of the A* algorithm. By only expanding the jump points with key significance, the search space and the amount of calculation are greatly reduced, thereby achieving faster global path planning. This method not only retains the advantage of the A* algorithm in solving the global optimal path, but also improves the search speed in large-scale or high-resolution maps. The Dynamic Window Approach (DWA) algorithm shows good adaptability in local obstacle avoidance and real-time path correction. The DWA algorithm achieves real-time response to dynamic obstacles during movement by sampling candidate trajectories in the robot's velocity space and combining comprehensive evaluation of indicators such as obstacle distance, heading deviation, speed, and trajectory smoothness. Although the DWA method alone can achieve effective obstacle avoidance in a local environment, it often falls into the local optimal solution and cannot guarantee the optimality of the overall path.

[0030] This implementation method first uses the A* algorithm improved based on JPS to perform global path search to quickly obtain the approximate optimal path, and then uses the improved DWA algorithm to perform local path smoothing and dynamic obstacle avoidance during the movement, and finally forms a closed-loop path planning process. Such a fusion strategy not only ensures the rationality of the global path, but also improves the real-time performance and safety in a dynamic environment. In addition, in order to cope with the uncertainty of environmental changes and dynamic obstacles, it has become a trend to introduce adaptive parameter adjustment methods based on machine learning. By collecting environmental and historical trajectory information online and using a lightweight multi-layer perceptron (MLP) to perform real-time tuning of the weight parameters in the evaluation function, the system can automatically adjust the decision-making strategy in different scenarios, thereby further improving the adaptability and robustness of path planning.

[0031] The path planning method based on the JPSA*-improved DWA fusion algorithm adopted in this implementation is to deeply integrate the global path planning and local dynamic obstacle avoidance technology on the basis of existing technologies, and overcome the challenges faced by traditional methods in dynamic and complex environments with the help of an adaptive parameter adjustment mechanism. It has significant practical application value and development prospects.

[0032] More specifically, the robot path planning method is as follows: Figure 1 As shown, the following process is included: Initialize the map environment, starting point and target point information, initialize the first list and the second list, add the starting point to the first list, expand the neighbors of the current node, filter the jump point as the key node, add the current key node to the second list, and determine whether the target point is in the first list. If not, return to continue expanding the neighbors of the current node. If so, the global optimal path is obtained; according to the global optimal path, extract the key nodes (i.e., jump points) in the global optimal path as sub-target points of the improved dynamic window algorithm, perform speed sampling, generate simulated motion trajectories, and select the optimal local trajectory according to the evaluation function (each time taking one of the adjacent jump points as the starting point and the next adjacent jump point as the sub-target point for local planning), determine whether the target point is reached, if not, return to continue speed sampling, and if so, the optimal local path is obtained.

[0033] More specifically, in the global path planning process, under the heuristic search framework of the A* algorithm, the jump point screening mechanism of the JPS algorithm is introduced to reduce the search space and improve the path search efficiency, including the mandatory neighbor rule and the jump point determination rule.

[0034] When the current node is expanded to multiple jump points through the jump rule of the JPS algorithm, each jump point will be evaluated according to the evaluation function Calculate the corresponding Value; among them, From the starting point to the node The actual cost, i.e. the cost of the path already taken; Represents a slave node The estimated cost to the target node, i.e. the heuristic function, is used to estimate the cost of the remaining path.

[0035] By comparing these jump points Value, select The smallest jump point is used as the key node to ensure that the algorithm prioritizes expanding the node with the best cost during the search process and ultimately finds the global optimal path.

[0036] In this implementation, preferably, the mandatory neighbor rule includes: setting the current extended node to , one of its neighbor nodes is blocked by an obstacle and is removed from its parent node arrive The shortest path must pass through ,but called The mandatory neighbor, Defined as a jump point.

[0037] In this implementation, preferably, the jump point determination rule includes: node When at least one of the following conditions is met, it is considered a jump point: ① is the starting point or target point; There is at least one mandatory neighbor; ③ When the parent node of is expanded along the diagonal direction, Search for a new jump point in the horizontal or vertical direction.

[0038] The jump search rules of this implementation mainly include horizontal or vertical jump search and diagonal jump, such as Figure 2 shown.

[0039] Jump search in horizontal or vertical direction, including: If the current node expands the search in the horizontal or vertical direction, the search continues to jump in that direction until an obstacle is encountered or a jump point is found. If the node ahead is unreachable during the jump, the current node is the jump point.

[0040] Diagonal jumps, specifically, include: If the current node is extended along the diagonal direction, the search is decomposed into horizontal and vertical jumps. During the search process, if the horizontal or vertical jump encounters an obstacle, the current node is determined to be a jump point and the expansion in that direction is terminated.

[0041] In this implementation, if Figure 2 As shown in the figure, taking the starting point (4, 0) and the target point (3, 6) as an example, the black grids are obstacles in the map, and their coordinates are (2, 5), (3, 5), and (4, 5). The dark gray grids are the starting point and the target point, and their coordinates are (4, 0) and (3, 6) respectively. The light gray grids are the jump points searched from the starting point to the target point. Figure 2 (a), (b), (c) and (d) are schematic diagrams of the first search process, the second search process, the third search process and the fourth search process, respectively.

[0042] like Figure 2 As shown in (a) in the figure, starting from the starting point (4, 0), searching in the horizontal and vertical directions respectively, due to encountering obstacles and boundaries, no jumping point is found, so moving along the diagonal.

[0043] like Figure 2 As shown in (b), the diagonal movement extends to node (1, 3). When searching vertically, the map boundary is encountered. When searching horizontally, a node (1, 5) with a mandatory neighbor is found, so the search ends. Node (1, 3) is called a jump point. like Figure 2 As shown in (c), node (1, 5) encounters a boundary when searching horizontally and an obstacle in the vertical direction. It moves diagonally and finds a mandatory neighbor node (2, 6). At this time, node (1, 5) is called a jump point. like Figure 2 As shown in (d) in the figure, the search from node (2, 6) in the horizontal direction hits the boundary, and the target node (3, 6) is found when searching in the vertical direction. Therefore, the search of the entire path is completed, and the optimal path is found by backtracking (the black bold solid arrow in the figure).

[0044] In this implementation, the DWA (Dynamic Window Approach) algorithm samples candidate trajectories in the velocity space when avoiding local obstacles, and scores the trajectories in combination with an improved evaluation function. It includes four indicators: heading deviation, obstacle distance, movement speed and trajectory smoothness. The formula is as follows: (1); in, It is the heading deviation function, which indicates the degree of alignment of the robot toward the target direction and is used to guide the robot's movement direction to prevent it from deviating from the target point; It is the obstacle distance evaluation function, which indicates the minimum braking distance between the robot and the obstacle at a certain linear velocity and angular velocity, so that the robot maintains a safe distance from the obstacle to prevent collision; Indicates the sampling speed, which enables the robot to move quickly to the target point; is the trajectory smoothness function, which is used to quantify the robot's linear velocity at a given speed. and angular velocity The smoothness of the generated trajectory is is the heading weight, is the obstacle avoidance weight, is the speed weight, These weight factors are adaptively adjusted to optimize the final planning path, allowing the algorithm to adapt to more different scenarios.

[0045] In this implementation, in order to deal with sudden dynamic obstacles such as pedestrians or vehicles, the elliptical safety distance model is used for the obstacle distance term in the evaluation function, such as Figure 3As shown, the major axis of the ellipse is dynamically determined according to the motion state of the obstacle to ensure that the robot maintains a safe distance from the obstacle. The obstacle distance evaluation function expression is: (2); in,( , )、( , ) represent the current coordinates of the robot and the obstacle respectively, For The position after time ( , ) The distance relative to the robot's elliptical projection position: (3); (4); in, is the angular coefficient of the ellipse equation, is the angular position of the obstacle relative to the robot, is the forward direction angle of the dynamic obstacle, and the major axis is determined according to the speed space of the obstacle The distance, in Within time, The length of is defined as: (5).

[0046] In this implementation, the DWA algorithm adds a trajectory smoothness evaluation item to the comprehensive evaluation function of the candidate trajectory during the local obstacle avoidance process, thereby optimizing the trajectory smoothness while ensuring the obstacle avoidance effect. Specifically, for each set of sampled linear speeds and angular velocity In addition to calculating indicators such as heading deviation, obstacle distance and movement speed, the smoothness of the trajectory is also calculated. The evaluation formula of the smoothness indicator based on the instantaneous curvature of the trajectory is: (6); in, Indicates that in a given ( , ) under the generated trajectory instantaneous curvature ( The smaller the value, the smoother the path). is a very small constant (it is desirable ), so that the denominator is not zero. In this way, in the process of local dynamic obstacle avoidance, the system not only selects candidate trajectories that can avoid obstacles, but also ensures that the selected trajectory has good smoothness through the smoothness evaluation item, thereby improving the stability and safety of the robot's motion.

[0047] In this implementation, preferably, the weight parameters of the evaluation function in the DWA algorithm are adjusted in an adaptive manner to avoid the uncertainty of manually set parameters and provide adaptability, safety and efficiency of path planning. Specifically, the parameter adaptive tuning method includes the following steps: Feature collection: Get the robot's motion state (position ( , ),speed( , ), target direction), environmental information (obstacle coordinates ( , ), speed, direction ( )), ellipse safety distance parameter (major axis , Angle coefficient ,direction ) and historical trajectory scores as model input features; Model calculation: Based on a lightweight multi-layer perceptron (MLP) network, dynamically adjust the heading weight of the DWA evaluation function , obstacle avoidance weight , speed weight and smoothness weight ; Trajectory optimization: Substitute the adjusted weight parameters into the DWA evaluation function (i.e., formula (1)), calculate the scores of different trajectories, and select the optimal trajectory as the final motion path.

[0048] It is implemented using a lightweight multi-layer perceptron (MLP) network architecture, as shown in the following figure. Figure 4 As shown, it includes the following structures: Input layer: The input dimension is 14 dimensions, including environmental information, robot status, elliptical safety distance parameters and historical trajectory scores; Hidden layer: HiddenLayer1 contains 64 neurons, uses ReLU activation function, and introduces dropout (probability 0.2) to prevent overfitting; HiddenLayer2 contains 32 neurons, uses ReLU activation function, and adds L1 regularization (λ=10-4) to enhance weight sparsity and improve generalization ability.

[0049] Output layer: uses a fully connected structure and outputs 4 weight parameters (heading weight , obstacle avoidance weight , speed weight and smoothness weight ), the activation function is linear regression, which is used to adaptively adjust the DWA evaluation function.

[0050] The present invention organically integrates global path planning (based on the improved JPSA* algorithm) with local dynamic obstacle avoidance (improved DWA algorithm) to form a closed-loop planning process. The system first uses the global algorithm to quickly generate the optimal path, and then uses the local algorithm to perform real-time obstacle avoidance and path optimization during the movement process, achieving efficient and safe overall path planning.

[0051] Embodiment 2: like Figure 5 As shown, this implementation provides a robot path planning system, including: The global path optimization unit is configured to: search for a global path from the starting point to the target point of the robot, extract jump points and screen key nodes, and determine the optimal global path; The local path generation unit is configured to: take the key nodes of the global optimal path as sub-target points, perform speed sampling on the sub-target points, and generate multiple simulated motion trajectories according to the speed sampling results; The local path optimization unit is configured to: score each simulated motion trajectory based on a weighted sum of heading deviation, obstacle distance, motion speed, and trajectory smoothness, and select the trajectory with the largest score as the optimal local trajectory; The walking path determination unit is configured to determine the walking path of the robot according to each optimal local trajectory on the global optimal path.

[0052] The specific working process of each of the above units is described in Example 1 and will not be repeated here.

[0053] It is understandable that the above-mentioned units can be separately or completely combined into one or several other units to constitute, or one (some) of the units can be further divided into multiple smaller units in function to constitute, which can achieve the same operation without affecting the realization of the technical effects of the embodiments of the present application. The above-mentioned units are divided based on logical functions. In practical applications, the functions of one unit can also be implemented by multiple units, or the functions of multiple units can be implemented by one unit. In other embodiments of the present application, the system may also include other units. In practical applications, these functions can also be implemented with the assistance of other units, and can be implemented by the collaboration of multiple units.

[0054] According to another embodiment of the present application, the system described in this embodiment can be constructed, and the method of Example 1 of the present application can be implemented by running a computer program (including program code) capable of executing the steps involved in the corresponding method described in Example 1 on a general computing device such as a computer including processing elements and storage elements such as a central processing unit (CPU), random access memory (RAM), and read-only memory (ROM). The computer program can be recorded on, for example, a computer-readable recording medium, and loaded into the above-mentioned computing device through the computer-readable recording medium and run therein.

[0055] Embodiment 3: like Figure 6 As shown, this implementation provides an electronic device, which includes a processor 1001, a communication interface 1002, and a computer-readable storage medium 1003. The processor 1001, the communication interface 1002, and the computer-readable storage medium 1003 may be connected via a bus or other means.

[0056] Among them, the communication interface 1002 is used to receive and send data, the computer-readable storage medium 1003 can be stored in the memory of the electronic device, the computer-readable storage medium 1003 is used to store a computer program, the computer program includes program instructions, and the processor 1001 is used to execute the program instructions stored in the computer-readable storage medium 1003.

[0057] The processor 1001 (or CPU (Central Processing Unit)) is the computing core and control core of the electronic device, which is suitable for implementing one or more instructions, and specifically suitable for loading and executing one or more instructions to implement corresponding method processes or corresponding functions.

[0058] The processor 1001 is configured to execute the following process: Perform a global path search from the starting point to the target point of the robot, extract jump points and screen key nodes to determine the optimal global path; Taking the key nodes of the global optimal path as sub-target points, performing speed sampling on the sub-target points, and generating multiple simulated motion trajectories according to the speed sampling results; Each simulated motion trajectory is scored based on the weighted sum of heading deviation, obstacle distance, motion speed and trajectory smoothness, and the trajectory with the largest score is selected as the optimal local trajectory; The walking path of the robot is determined according to each optimal local trajectory on the global optimal path.

[0059] The specific working process is described in Example 1 and will not be repeated here.

[0060] Embodiment 4: This implementation provides a computer-readable storage medium (Memory), which is a memory device in an electronic device for storing programs and data. It is understandable that the computer-readable storage medium here can include both built-in storage media in the electronic device and, of course, extended storage media supported by the electronic device. The computer-readable storage medium provides a storage space that stores the processing system of the electronic device.

[0061] In addition, the storage space also stores one or more instructions suitable for being loaded and executed by the processor, and these instructions may be one or more computer programs (including program codes). It should be noted that the computer-readable storage medium here may be a high-speed RAM memory, or a non-volatile memory, such as at least one disk memory; optionally, it may also be at least one computer-readable storage medium located away from the aforementioned processor.

[0062] In one embodiment, the computer-readable storage medium stores one or more instructions; the processor loads and executes the one or more instructions stored in the computer-readable storage medium to implement the following process: Perform a global path search from the starting point to the target point of the robot, extract jump points and screen key nodes to determine the optimal global path; Taking the key nodes of the global optimal path as sub-target points, performing speed sampling on the sub-target points, and generating multiple simulated motion trajectories according to the speed sampling results; Each simulated motion trajectory is scored based on the weighted sum of heading deviation, obstacle distance, motion speed and trajectory smoothness, and the trajectory with the largest score is selected as the optimal local trajectory; The walking path of the robot is determined according to each optimal local trajectory on the global optimal path.

[0063] The specific working process is described in Example 1 and will not be repeated here.

[0064] Embodiment 5: The present implementation provides a computer program product or a computer program, which includes computer instructions stored in a computer-readable storage medium. A processor of an electronic device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the electronic device performs the following process: Perform a global path search from the starting point to the target point of the robot, extract jump points and screen key nodes to determine the optimal global path; Taking the key nodes of the global optimal path as sub-target points, performing speed sampling on the sub-target points, and generating multiple simulated motion trajectories according to the speed sampling results; Each simulated motion trajectory is scored based on the weighted sum of heading deviation, obstacle distance, motion speed and trajectory smoothness, and the trajectory with the largest score is selected as the optimal local trajectory; The walking path of the robot is determined according to each optimal local trajectory on the global optimal path.

[0065] The specific working process is described in Example 1 and will not be repeated here.

[0066] A person skilled in the art can appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed in this application can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. A person skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0067] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on the computer, the process or function according to the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium or transmitted through a computer-readable storage medium. The computer instructions can be transmitted from a website site, a computer, a server or a data center to another website site, a computer, a server or a data center by wired (e.g., coaxial cable, optical fiber, digital line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that can be accessed by the computer or a data processing device such as a server, a data center, etc. that includes one or more available media integrated. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid-state drive (SSD)), etc.

[0068] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A robot path planning method, characterized in that: The process includes: Perform a global path search from the starting point to the target point of the robot, extract jump points and screen key nodes to determine the optimal global path; Taking the key nodes of the global optimal path as sub-target points, performing speed sampling on the sub-target points, and generating multiple simulated motion trajectories according to the speed sampling results; Each simulated motion trajectory is scored based on the weighted sum of heading deviation, obstacle distance, motion speed and trajectory smoothness, and the trajectory with the largest score is selected as the optimal local trajectory; The walking path of the robot is determined according to each optimal local trajectory on the global optimal path.

2. The robot path planning method according to claim 1, characterized in that: node A critical jump point is determined when at least one of the following conditions is met: is the starting point or destination point; There is at least one mandatory neighbor; When the parent node of is expanded along the diagonal direction, Search for new jump points in horizontal or vertical directions; Horizontal or vertical jump search, including: if the current node is extended along the horizontal or vertical direction, the search continues to jump along the direction until an obstacle is encountered or a jump point is found. If the node ahead is unreachable during the jump process, the current node is the jump point; Diagonal jump search includes: if the current node is extended along the diagonal direction, the search is decomposed into horizontal and vertical jumps. During the search process, if the horizontal or vertical jump encounters an obstacle, the current node is determined to be a jump point and the expansion in this direction is terminated.

3. The robot path planning method according to claim 1, characterized in that: Scoring is based on a weighted sum of heading deviation, obstacle distance, speed, and trajectory smoothness, including: ； in, is the heading deviation function; is the obstacle distance evaluation function; Indicates the sampling speed; is the trajectory smoothness function, which is used to quantify the robot's linear velocity at a given speed. and angular velocity The smoothness of the generated trajectory is is the heading weight, is the obstacle avoidance weight, is the speed weight, is the smoothness weight.

4. The robot path planning method according to claim 3, characterized in that: The obstacle distance evaluation function adopts the elliptical safety distance, and the obstacle distance evaluation function expression is: ； in,( , )、( , ) represent the current coordinates of the robot and the obstacle respectively, For The position after time ( , ) relative to the robot's elliptical projection position, , , is the angular coefficient of the ellipse equation, is the angular position of the obstacle relative to the robot, is the forward direction angle of the dynamic obstacle. Within time, .

5. The robot path planning method according to claim 3, characterized in that: The trajectory smoothness function is: ,in, Indicates that at a given line speed and angular velocity The instantaneous curvature of the trajectory generated under is a constant greater than zero.

6. The robot path planning method according to any one of claims 3 to 5, characterized in that: Obtain the robot's motion state, environmental information, elliptical safety distance parameters, and historical trajectory scores as input features of the multi-layer perceptron network to dynamically adjust the heading weight. , obstacle avoidance weight , speed weight and smoothness weight .

7. A robot path planning system, characterized in that: include: The global path optimization unit is configured to: search for a global path from the starting point to the target point of the robot, extract jump points and screen key nodes, and determine the optimal global path; The local path generation unit is configured to: take the key nodes of the global optimal path as sub-target points, perform speed sampling on the sub-target points, and generate multiple simulated motion trajectories according to the speed sampling results; The local path optimization unit is configured to: score each simulated motion trajectory based on a weighted sum of heading deviation, obstacle distance, motion speed, and trajectory smoothness, and select the trajectory with the largest score as the optimal local trajectory; The walking path determination unit is configured to determine the walking path of the robot according to each optimal local trajectory on the global optimal path.

8. A computer device, characterized in that: include: a processor and a computer readable storage medium; a processor adapted to execute a computer program; A computer-readable storage medium, wherein a computer program is stored in the computer-readable storage medium, and when the computer program is executed by the processor, the robot path planning method according to any one of claims 1 to 6 is implemented.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and the computer program is suitable for being loaded by a processor and executing the robot path planning method according to any one of claims 1 to 6.

10. A computer program product, characterized in that The computer program product comprises a computer program, and when the computer program is executed by a processor, the robot path planning method according to any one of claims 1 to 6 is implemented.

Images