Robot path planning method based on adaptive fuzzy and hybrid strategy
By improving the RRT-Connect algorithm and combining fuzzy control and hybrid extension strategies, the efficiency and robustness issues of robot path planning in complex environments are solved, achieving efficient and reliable path planning, and intelligent decision-making and optimization adaptable to different environments.
Patent Information
- Application Number
- CN202610023102.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-01-08
- Publication Date
- 2026-05-01
AI Technical Summary
Existing robot path planning methods suffer from poor environmental adaptability, slow convergence speed, low planning efficiency, poor path quality, and are prone to stalling and lacking robustness when facing complex and ever-changing environments.
An improved RRT-Connect algorithm based on adaptive fuzzy and hybrid strategies is adopted, which combines fuzzy control theory, hierarchical hybrid extension mechanism and dynamic path optimization idea. The extension strategy parameters are dynamically generated by the fuzzy logic controller. Combined with a three-level hybrid extension module and tree management optimization module, the adaptive and efficient path planning is achieved.
It significantly improves the success rate, efficiency, and path quality of robot path planning in complex three-dimensional spaces, providing an efficient, reliable, and intelligent path planning solution.
Smart Images

Figure CN121957003A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of robot motion planning technology, and in particular to a robot path planning method based on adaptive fuzzy and hybrid strategies. Background Technology
[0002] Path planning, a core technology of autonomous robot systems, directly determines the efficiency and safety of robots in complex scenarios such as industrial manufacturing, medical services, and warehousing and logistics. In particular, for multi-jointed robots with five, six, or even more degrees of freedom (DOF), motion planning needs to be solved in a high-dimensional configuration space (C-space). How to quickly generate a collision-free feasible path in a three-dimensional environment full of obstacles has always been a key research focus and technical bottleneck in robotics. Among these, sampling-based planning algorithms, especially Rapidly-exploring RandomTree (RRT), have become one of the mainstream methods in this field because they can effectively avoid explicit modeling of complex spaces and demonstrate strong exploration capabilities in high-dimensional C-space. However, existing RRT-based path planning techniques still have significant shortcomings in several aspects.
[0003] Existing algorithms struggle to achieve an effective balance between convergence efficiency and path quality. While the basic RRT algorithm possesses probabilistic completeness, its stochastic exploration mechanism leads to slow convergence, and the generated paths are typically redundant, non-smooth, and far from optimal. Therefore, researchers have proposed... The algorithm, by introducing a choose-parent mechanism and a rewire mechanism, can asymptotically converge to the optimal path. However, this pursuit of optimality comes at the cost of computational efficiency; its numerous neighborhood searches and rewire operations result in slow convergence, making it difficult to meet the real-time requirements of robot operations.
[0004] In terms of environmental adaptability and robustness, existing algorithms generally lack the ability to dynamically adapt to complex environments. Most RRT variants (such as RRT-Connect) employ fixed hyperparameters such as expansion step size and target bias rate. This static strategy is generally ineffective when environmental characteristics are varied: in open areas, a small step size limits exploration efficiency; while in narrow or obstacle-filled areas, a large step size easily leads to collisions, reducing the planning success rate. On the other hand, the robustness of algorithms is severely challenged when facing "trap regions" such as concave obstacles and narrow passages. For example, the RRT-Connect algorithm accelerates convergence by expanding two trees simultaneously, but in trap regions, the expansion of both trees easily stagnates, leading to a sharp increase in planning time or even failure. Existing algorithms lack a mechanism for intelligently sensing the environment and dynamically adjusting exploration strategies.
[0005] Regarding the limitations of exploration strategies, hybrid guidance strategies often introduce new inherent defects. To overcome the blind exploration of Random Tree Exploration (RRT), some studies have attempted to combine RRT with other deterministic methods (such as the Artificial Potential Field (APF) method). These hybrid methods utilize gradient information generated by APF to guide the random tree towards the target point, which can accelerate convergence to some extent. However, this strategy inherits the inherent defect of the APF method, namely the "local minima" problem. When the exploration node of the random tree gets stuck in a local minimum of the potential field, the attraction and repulsion forces reach equilibrium, the guidance fails, and the robot's exploration stagnates, making it unable to escape. This indicates that simply fusing different strategies cannot fundamentally solve the problem; instead, it may introduce new technical dilemmas.
[0006] In summary, existing robot path planning methods, due to their use of fixed parameters, suffer from poor environmental adaptability, slow convergence speed, low planning efficiency, and poor path quality when facing complex and ever-changing environments. Furthermore, they are prone to stalling and lack robustness when encountering trap areas such as concave obstacles. Summary of the Invention
[0007] This invention provides a robot path planning method based on adaptive fuzzy and hybrid strategies, which at least partially solves the aforementioned technical problems of existing robot path planning methods.
[0008] To solve the above-mentioned technical problems, the present invention provides the following technical solution: On one hand, the present invention provides a robot path planning method based on adaptive fuzzy and hybrid strategies, the robot path planning method based on adaptive fuzzy and hybrid strategies including: The robot senses its surroundings and obtains environmental information. Based on the RRT-Connect algorithm framework, the improved RRT-Connect algorithm is obtained by integrating fuzzy control theory, hierarchical hybrid extension mechanism and dynamic path optimization idea; The improved RRT-Connect algorithm is used to realize robot path planning based on environmental information.
[0009] Furthermore, the environmental information includes: the Euclidean distance between the node in the current tree to be expanded that is closest to the other tree and the other tree. and the openness of the environment FSO .
[0010] Furthermore, the openness of the environment FSO The acquisition method is as follows: starting from the end node of the current tree, emit evenly to the surrounding space. N A detection ray is used to count the number of rays that do not collide with obstacles. ,by and N The ratio of the two values is used as the degree of openness of the environment. FSO ;in, N This is a preset integer value.
[0011] Furthermore, the improved RRT-Connect algorithm, based on the RRT-Connect algorithm framework, adds a fuzzy logic controller, a three-level hybrid extension module, and a tree management and optimization module; among them, The fuzzy logic controller is used to dynamically generate macroscopic policy parameters to guide the current expansion based on fuzzy control theory and environmental information; among these, the macroscopic policy parameters include: dynamic target bias rate. Dynamically expanding step size and the radius of the dynamic reconnection neighborhood ; The three-level hybrid extension module is used for a hierarchical hybrid extension mechanism to perform node generation operations at the micro level based on the macro-strategy parameters output by the fuzzy logic controller. The tree management and optimization module is used to connect newly generated nodes to a random tree and perform local optimization on the path.
[0012] Furthermore, the fuzzy logic controller is specifically used for: take over and FSO And according to the preset membership function, and FSO These are respectively mapped to the membership degrees of different fuzzy sets; where, The fuzzy set is: {Near, Medium, Far}. FSOThe fuzzy set is: {Congested, Normal, Open}; Reasoning is performed according to a preset fuzzy rule base, and the fuzzy conclusions of each output variable obtained from the fuzzy reasoning are converted back into a numerical value through a defuzzification method, and finally a set of macroscopic policy parameters for guiding the current expansion are output; among them, the fuzzy rule base includes the following rules: Rule 1: IF ( is far) AND ( FSO is open) THEN ( is high) AND ( is large) AND ( is large); Rule 2: IF ( is near) AND ( FSO is congested) THEN ( is low) AND ( is small) AND ( is small); Rule 3: IF ( is medium) AND ( FSO is normal) THEN ( is medium) AND ( is medium) AND ( is medium).
[0013] Furthermore, the three - level hybrid expansion module is specifically used for: First, execute the first - level strategy. Taking the output by the fuzzy logic controller as the probability, generate a random number within the range of [0, 1] . If is less than , then determine that the expansion direction is towards the other tree; otherwise, abandon the target - biased growth and directly trigger the second - level strategy. Select the node closest to the expansion direction from the current tree . Subsequently, extend along the expansion direction with the output by the fuzzy logic controller to generate a main candidate node ; perform collision detection on the connection path from to ; if there is no collision, then regard as the valid new node for this iteration ; if a collision occurs, it is considered that the first - level strategy fails, and at this time, the second - level strategy is triggered;
[0014] When the first - level strategy fails, at the end node of the current expansion Calculate an escape force ; and outputs the fuzzy logic controller along the escape force direction. Extend the process to generate an escape candidate node. ;right Perform collision detection; if no collision occurs, then... Considered as a valid new node in this iteration If a collision still occurs, the secondary strategy is considered to have failed, and the tertiary strategy is triggered. When both the primary and secondary strategies fail, then... Centered on, with a radius of Select a node within the sphere as a backup candidate node. ; then on arrive Collision detection is performed on the connection path; if no collision is found, then... Considered as a valid new node in this iteration If a collision still occurs, the current iteration is considered to have failed, and no new node will be generated in this iteration; the iteration will wait for the next iteration.
[0015] Furthermore, escape force The calculation method is as follows: First, calculate the reinforcing repulsion force. : ; in, It is the repulsive force gain coefficient; yes Distance to the nearest obstacle; It is the range of the repulsive force; Then the rotational gravity was calculated. : ; in, It is the gravitational gain coefficient; It is the standard gravitational force pointing towards the target; It is a virtual gravitational force pointing in a random direction; Finally, the total escape force is synthesized. : .
[0016] Furthermore, the tree management and optimization module is specifically used for: Will The algorithm's concepts of reselecting parent nodes and reconnecting, and the output of the fuzzy logic controller. By combining these methods, a parent node reselection and reconnection mechanism is implemented within the RRT-Connect algorithm framework. This mechanism is followed by a process that occurs after a new node is successfully added. Path reconstruction is performed within a defined neighborhood to adaptively optimize the path.
[0017] In another aspect, the present invention also provides an electronic device comprising a processor and a memory; wherein the memory stores at least one instruction, which is loaded and executed by the processor to implement the above-described method.
[0018] In another aspect, the present invention also provides a computer-readable storage medium storing at least one instruction, which is loaded and executed by a processor to implement the above method.
[0019] The beneficial effects of the technical solution provided by this invention include at least the following: This invention utilizes fuzzy control theory to endow the path planning process with macroscopic intelligent decision-making capabilities, and provides microscopic precise execution and fault tolerance capabilities through hierarchical hybrid strategies. It is an efficient, robust, and universal solution that significantly improves the success rate, efficiency, and path quality of robot path planning in complex high-dimensional spaces. It can provide robots with an efficient, reliable, and intelligent decision-making path planning solution in complex three-dimensional spaces, achieving more efficient and better collision-free path generation. Attached Figure Description
[0020] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0021] Figure 1 This is a framework diagram of the AFA-RRT-Connect algorithm provided in this embodiment of the invention; Figure 2 This is a flowchart of macroscopic adaptive control (working principle of fuzzy logic controller) provided in an embodiment of the present invention; Figure 3 This is a three-level hybrid extended decision-making flowchart provided in an embodiment of the present invention; Figure 4 This is a schematic diagram of the escape strategy (hybrid artificial potential field method) provided in an embodiment of the present invention; Figure 5 This is a system block diagram of the electronic device provided in the embodiments of the present invention. Detailed Implementation
[0022] To make the objectives, technical solutions, and advantages of the present invention clearer, the embodiments of the present invention will be described in further detail below with reference to the accompanying drawings.
[0023] First, it should be noted that in the embodiments of the present invention, the words "exemplarily," "for example," etc., are used to indicate that they are examples, illustrations, or descriptions. Any embodiment or design scheme described as "exemplary" in the present invention should not be construed as being more preferred or advantageous than other embodiments or design schemes. Specifically, the use of the term "exemplarily" is intended to present the concept in a specific manner. Furthermore, in the embodiments of the present invention, the meaning expressed by "and / or" can be both, or it can be either one or the other.
[0024] First Embodiment For multi-degree-of-freedom robots in high-dimensional configuration space (C-Space), this embodiment provides a robot path planning method based on adaptive fuzzy and hybrid strategies, aiming to provide robots with an efficient, reliable and intelligent decision-making path planning solution in complex three-dimensional space.
[0025] This method provides a systematic solution from aspects such as adaptive adjustment of planning strategies, trap escape mechanisms, and path optimization, so as to significantly improve the efficiency and quality of path planning while ensuring the success rate and robustness of planning, so as to meet the needs of robots for efficient and reliable autonomous navigation in complex three-dimensional space.
[0026] At the macro-control level, this embodiment introduces a fuzzy logic controller as the "intelligent brain" of the algorithm. This controller can perceive environmental changes in real time (the distance from the current expansion node to the target and the openness of the surrounding environment). Through fuzzification, rule-based fuzzy inference, and defuzzification processes, it dynamically outputs a set of control parameters to guide expansion and optimization, including dynamic target bias rate, dynamic expansion step size, and dynamic reconnection neighborhood radius. This enables the algorithm to have global adaptive capabilities, allowing it to make the most appropriate decisions based on "distance from the target" and "environmental congestion." This solves the problem of insufficient efficiency and robustness in existing robot path planning algorithms due to the use of fixed parameters.
[0027] At the micro-execution level, this embodiment constructs a hierarchical three-level hybrid expansion mechanism of "main strategy - escape strategy - backup strategy". This mechanism can provide progressive solutions for different expansion results based on the success or failure status during the expansion process, and efficiently cope with normal environments and complex trap areas. First, an efficient target bias expansion (main strategy) is used for rapid exploration. When this strategy fails due to collision, a specially designed Hybrid APF method (escape strategy) is immediately activated to get rid of local traps. This Hybrid APF method is designed for the escape problem in trap areas. It achieves efficient escape by combining an "enhanced repulsive force" adjusted by the environmental crowding degree and a "rotational gravity" that integrates the target direction and random direction. This method can not only forcefully push away obstacles, but also jump out of the local minimum point of the traditional APF by introducing random perturbations, thus achieving efficient escape. If the escape strategy still fails, it reverts to pure random expansion (safety net strategy). This hybrid expansion mechanism enables efficient exploration of normal environments and intelligent escape from concave and other trap regions, ensuring the probabilistic completeness of the algorithm in extremely complex environments.
[0028] At the path optimization level, a dynamic path optimization mechanism is proposed, which will... The concept of reselecting parent nodes and reconnecting is combined with the dynamic reconnection neighborhood radius output by the fuzzy logic controller. The path optimization is performed using the dynamic neighborhood radius output by the fuzzy logic controller. After a new node is successfully added, the path is reconstructed within the neighborhood determined by the dynamic radius. This allows for a larger radius for broader optimization in open environments and a smaller radius to reduce computational overhead in congested environments. This balances optimization cost and path quality in environments with different levels of congestion, achieving an adaptive optimization process.
[0029] Based on the above, the implementation process of the robot path planning method in this embodiment is as follows: S1, perceives the robot's surrounding environment and obtains environmental information; S2, based on the RRT-Connect algorithm framework, integrates fuzzy control theory, hierarchical hybrid extension mechanism and dynamic path optimization idea to obtain the improved RRT-Connect algorithm; S3 utilizes the improved RRT-Connect algorithm to achieve robot path planning based on environmental information.
[0030] Specifically, this embodiment, based on the classic RRT-Connect algorithm framework, deeply integrates fuzzy control theory, hybrid artificial potential field method, and dynamic path optimization ideas to construct an efficient and intelligent adaptive fuzzy and hybrid strategy robot path planning algorithm system (AFA-RRT-Connect). It provides a highly robust and efficient solution for robot path planning in complex, dynamic 3D environments containing trap areas. The path planning system framework is as follows: Figure 1 As shown, its core idea revolves around a bidirectional, rapidly expanding random tree (RRT-Connect) framework. In each iterative expansion, the planning process achieves intelligence and adaptability through the collaborative work of three core modules. These three modules are: a Fuzzy Logic Controller (FLC), serving as the macro-control center; a three-level hybrid expansion module, serving as the micro-execution engine; and a tree management and optimization module, serving as the path quality assurance mechanism.
[0031] 1. System Participants: The core participant in this system is the starting tree. and target tree The starting tree begins from the robot's initial configuration. Growth begins, the target tree starts from the target configuration Growth begins. The two trees explore towards each other in the configuration space (C-Space) until they connect, forming a complete collision-free path.
[0032] 2. Core Algorithm Framework: The AFA-RRT-Connect algorithm acts as the intelligent decision-making and execution engine throughout the planning process. In each iteration, the algorithm alternately expands the initial tree. and target tree Before expanding a tree, the fuzzy logic controller (FLC) is first invoked to evaluate the current environment in real time, dynamically generating macroscopic strategy parameters to guide the expansion. Subsequently, the three-level hybrid expansion module executes specific node generation operations at the micro level based on these parameters, intelligently responding to different local environments. Finally, if a new node is successfully generated, the tree management and optimization module is responsible for effectively integrating it into the random tree and performing local path optimization. This entire process is repeated until the two trees are successfully connected.
[0033] 3. Core Functional Modules: The Fuzzy Logic Controller (FLC), acting as the system's "intelligent brain," is primarily responsible for the adaptive adjustment of macro-level strategies. It fuzzifies precise environmental perception data (such as distance to the target and environmental openness), simulates expert decision-making processes through a fuzzy inference engine, and ultimately defuzzifies and outputs a set of optimal extended control parameters. This allows the algorithm to make the most appropriate strategy adjustments based on states such as "distance from the target" and "environmental congestion." The three-level hybrid extension module, acting as the system's "executive limbs," is responsible for specific exploration steps. It incorporates three different levels of extension strategies, forming a progressive execution logic of "main strategy - escape strategy - safety strategy." It can seamlessly switch between efficient target bias exploration, intelligent escape from traps, and random exploration ensuring completeness, depending on the success or failure of the previous round of extension. The tree management and optimization module, acting as the system's "maintenance steward," is responsible for maintaining the structural health and path quality of the random tree. Building upon RRT-Connect, it integrates... The core optimization operations in the algorithm are reselection of parent nodes and reconnection. It innovatively uses the dynamic neighborhood radius output by FLC to balance the optimization effect and computational cost, thus achieving adaptive path optimization.
[0034] The following sections introduce the macro-adaptive control process, the three-level hybrid expansion process, and the dynamic path optimization process.
[0035] Macroeconomic adaptive control process: At the start of each iteration expansion, the system first invokes the Fuzzy Logic Controller (FLC) for macroscopic adaptive control. This process mainly involves three entities: the end node of the tree to be expanded, the environment perception module, and the Fuzzy Logic Controller (FLC). The FLC, as the decision-making core, dynamically generates expansion control parameters to guide this iteration based on information provided by the environment perception module. The macroscopic adaptive control process is as follows: Figure 2 As shown, the specific steps are as follows:
[0036] 1. Environmental Information Acquisition and Quantification The environment awareness module acquires the two key input variables required for the current iteration and quantifies them: The first input variable is the distance to the target. Calculate the Euclidean distance between the node in the current tree to be expanded that is closest to the target tree and the target tree; this distance reflects the progress of the algorithm in the global search. and These are the node configurations in the two trees, respectively.
[0037]
[0038] The second input variable is the ambient openness (FSO), which is emitted uniformly into the surrounding space starting from the end node of the current tree. N A detection ray is used to count the proportion of rays that do not collide with obstacles within a short distance; this proportion reflects the congestion level of the local space where the node is located. The number of rays that do not collide. The range of FSO is [0, 1], where a value close to 1 indicates an open environment, and a value close to 0 indicates a crowded environment. FSO is defined as:
[0039]
[0040] 2. Input fuzzification FLC receiver The two precise values, FSO and FSO, are mapped to the membership degrees of different fuzzy sets according to a preset membership function (MFs). In this embodiment, The fuzzy set of FSO is {Near, Medium, Far}, and the fuzzy set of FSO is {Congested, Normal, Open}.
[0041] 3. Fuzzy reasoning FLC performs inference based on an internally pre-defined fuzzy rule base. The rule base is a collection of "IF-THEN" rules that allow FLC to simulate expert experience and establish a complex mapping between environmental states and optimal policy parameters. Some rule examples are shown below:
[0042] Rule 1: IF ( (is far) AND (FSO is open) THEN ( is high) AND ( is big) AND ( is big) Rule 2: IF ( (is near) AND (FSO is crowded) THEN ( is low) AND ( is small) AND ( is small) Rule 3: IF ( is medium) AND (FSO is ordinary) THEN ( is in the middle) AND ( is in the middle) AND ( is in the middle).
[0043] 4. Output defuzzification The fuzzy conclusions of each output variable obtained from fuzzy inference are transformed back into a precise, executable value using a defuzzification method. Finally, FLC outputs a set of dynamic control parameters tailored to this iteration: the dynamic target bias rate. Dynamically expanding step size and the radius of the neighborhood of dynamic reconnection These parameters will be directly used in the subsequent three-level hybrid expansion and path optimization process.
[0044] Three-level hybrid expansion process: After obtaining the dynamic control parameters output by FLC, the system executes a three-level hybrid expansion process to generate new valid nodes. This process involves entities such as the current expansion tree, obstacle space, and the three-level hybrid expansion module. This module generates nodes in a progressive and fault-tolerant manner, ensuring the most effective expansion attempts are made in different local environments. The decision-making process for the three-level hybrid expansion is as follows: Figure 3 As shown, the specific steps are as follows:
[0045] 1. Primary Strategy (Main Strategy) Attempt The module first executes the most efficient main policy, using the dynamic target bias rate output by FLC. Generate a random number in the range [0, 1] with probability. ,like Less than If the target is not found, the expansion direction is set towards the other tree; otherwise, the target bias growth is abandoned, and the secondary strategy is triggered directly. The node closest to this direction is selected from the current tree. Then, the dynamically expanded step size is output by FLC along the selected direction. Extend the process to generate a primary candidate node. ;right arrive Collision detection is performed on the connection path. If no collision is found, then... Considered a valid new node in this iteration If the process is successful, the node will be handed over to the tree management and optimization module for processing. If a collision occurs, the main strategy is considered to have failed, and a secondary strategy will be triggered immediately.
[0046] 2. Secondary strategy (escape strategy) attempt The module initiates a specially designed escape strategy—the Hybrid APF—only when the primary strategy fails. Figure 4 As shown, at the current extended end node Calculate an escape force The calculation method is as follows: First, calculate the reinforcing repulsion force. The magnitude of the repulsive force is negatively correlated with the openness (FSO) of the environment (i.e., 1-FSO); the more crowded the environment, the stronger the repulsive force, providing a stronger pushing effect. It is the repulsive force gain coefficient. yes Distance to the nearest obstacle It is the range of the repulsive force.
[0047]
[0048] Subsequently, rotational gravity is calculated, with the gravitational direction composed of a weighted mixture of "gravity toward the target" and "a random direction vector." The weight of the random component is negatively correlated with the Free-Solution (FSO). The more crowded the environment, the stronger the random exploration, helping to escape the local minima of the traditional APF. It is the gravitational gain coefficient. It is the standard gravitational force pointing towards the target. It is a virtual gravitational force pointing in a random direction.
[0049]
[0050] It is worth noting that the repulsion gain coefficient With gravitational gain coefficient These are all preset positive constants, and their values are determined based on the robot's motion space scale and simulation experiments. Typically, It mainly affects the speed at which the robot approaches the target. If the value is too large, it will cause the robot to overshoot the target point, and if the value is too small, the convergence will be slow. The main factor affecting obstacle avoidance response strength is the value set to ensure the robot generates sufficient thrust to avoid collisions when approaching obstacles. Much larger (For example, in a map with meter-level units, you can set...) , This ensures that repulsive forces dominate near obstacles, while attractive forces dominate away from obstacles.
[0051] Finally, the total escape force is synthesized. And dynamically expand the step size along the direction of the force. Extend this process to generate an escape candidate node. .
[0052]
[0053] right Perform a collision detection. If no collision is found, the node is considered a valid new node. The process ends successfully. If a collision still occurs, the escape strategy is considered to have failed, and a third-level strategy is triggered.
[0054] 3. Trial of three strategies (safety net strategy) When both the first and second level strategies fail, this strategy is executed to ensure the probabilistic completeness of the algorithm. Firstly, in the case of... Centered on, with a radius of Within a high-dimensional sphere, select a completely new and random node as a backup candidate node. Then, the validity of the node is verified, that is, the node is... arrive Collision detection is performed on the connection path. If no collision is found, then... Considered a valid new node in this iteration If a collision still occurs, the expansion of this iteration fails, no new node is generated, and the process waits for the next iteration.
[0055] Dynamic path optimization process: When a new node is successfully generated using any of the above strategies Then, the tree management and optimization module starts to perform access and path optimization. This is achieved by using the dynamic reconnection neighborhood radius generated by FLC. The optimization process of this invention is also adaptive: when the environment is open, a larger radius is used for more extensive and better optimization; when the environment is crowded, a smaller radius is used to reduce unnecessary collision detection and computational overhead.
[0056] 1. Select the parent node In Centered on, with radius Within the neighborhood, search Calculate all neighboring nodes. Calculate the distance to each neighboring node. The path cost (calculated from the root node of the tree), choose to make The neighboring node with the lowest total cost is chosen as its parent node.
[0057] 2. Reconnect Similarly in Centered on, with radius Within the neighborhood, checking for the existence of neighboring nodes can be done through... Find a shorter path. If a neighboring node exists, disconnect its parent node and reconnect to it. superior.
[0058] In summary, this embodiment provides a robot path planning method based on adaptive fuzzy and hybrid strategies. It utilizes fuzzy control theory to endow the path planning process with macroscopic intelligent decision-making capabilities, and provides microscopic precise execution and fault tolerance capabilities through hierarchical hybrid strategies. It is an efficient, robust, and universal solution that significantly improves the success rate, efficiency, and path quality of robot path planning in complex high-dimensional spaces. It can provide robots with an efficient, reliable, and intelligent decision-making path planning solution in complex three-dimensional spaces, achieving more efficient and better collision-free path generation.
[0059] Second Embodiment This embodiment provides an electronic device, such as... Figure 5 As shown, the electronic device includes a processor and a memory; wherein the processor and the memory can be connected via a communication bus; the memory stores at least one instruction, which is loaded and executed by the processor to implement the method of the first embodiment described above. Furthermore, the electronic device may also include a transceiver, the processor and the transceiver can be connected via a communication bus, and the transceiver is used to communicate with other devices.
[0060] Below, in conjunction with Figure 5 A detailed introduction to each component of this electronic device is provided below: The processor is the control center of the electronic device. The electronic device may include multiple processors, each of which can be a single-core processor (single-CPU) or a multi-core processor (multi-CPU). The term "processor" can refer to a single processor or a collective term for multiple processing elements. For example, a processor can be one or more central processing units (CPUs), other general-purpose processors, application-specific integrated circuits (ASICs), or one or more integrated circuits configured to implement embodiments of the present invention, such as one or more digital signal processors (DSPs), one or more field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor or any conventional processor. The processor can perform various functions of the electronic device by running or executing software programs stored in memory and by calling data stored in memory.
[0061] In a specific implementation, as one example, the processor may include one or more CPUs, for example... Figure 5 CPU0 and CPU1 shown are, of course, merely illustrative examples.
[0062] The memory is used to store the software program that executes the solution of the present invention, and the processor controls its execution. For specific implementation methods, please refer to the above method embodiments, which will not be repeated here.
[0063] Optionally, the memory may be a read-only memory (ROM) or other type of static storage device capable of storing static information and instructions, random access memory (RAM) or other type of dynamic storage device capable of storing information and instructions, or electrically erasable programmable read-only memory (EEPROM), compact disc read-only memory (CD-ROM) or other optical disc storage, optical disc storage (including compressed optical discs, laser discs, optical discs, digital universal optical discs, Blu-ray discs, etc.), magnetic disk storage media or other magnetic storage devices, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and accessible by a computer, but not limited thereto. The memory may be integrated with the processor or exist independently, and may be accessed through the interface circuit of the electronic device ( Figure 5 (Not shown in the image) is coupled to the processor; however, this embodiment of the invention does not impose specific limitations on this.
[0064] The transceiver may include a receiver and a transmitter. Figure 5 (Not shown separately). The receiver is used to implement the receiving function, and the transmitter is used to implement the transmitting function. The transceiver can be integrated with the processor or exist independently, and can be connected through the interface circuit of the electronic device (…). Figure 5 (Not shown in the image) is coupled to the processor, and this embodiment of the invention does not specifically limit this.
[0065] In addition, it should be noted that, Figure 5 The structure of the electronic device shown is not intended to limit the device. Actual devices may include more or fewer components than shown, or combine certain components, or have different component arrangements. Furthermore, the technical effects achieved by this electronic device when performing the method of the first embodiment described above can be referenced to the technical effects described in the first embodiment; therefore, they will not be repeated here.
[0066] Third Embodiment This embodiment provides a computer-readable storage medium storing at least one instruction, which is loaded and executed by a processor to implement the method of the first embodiment described above. The computer-readable storage medium may be a ROM, random access memory, CD-ROM, magnetic tape, floppy disk, or optical data storage device, etc. The instruction stored therein can be loaded and executed by a processor in a terminal.
[0067] Furthermore, it should be noted that the present invention can be provided as a method, apparatus, or computer program product. Therefore, embodiments of the present invention can take the form of a completely or partially hardware embodiment, a completely or partially software embodiment, or an embodiment combining software and hardware aspects. Moreover, when implemented in software, embodiments of the present invention can take the form of a computer program product implemented on one or more computer-usable storage media containing computer-usable program code. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer program are loaded or executed on a computer, all or part of the processes or functions described in the embodiments of the present invention are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any usable medium accessible to a computer or a data storage device such as a server or data center containing one or more sets of usable media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium. A semiconductor medium can be a solid-state drive (SSD).
[0068] Embodiments of the present invention are described with reference to flowchart illustrations and / or block diagrams of methods, terminal devices (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, embedded processor, or other programmable data processing terminal device to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing terminal device, generate instructions for implementing the flowchart illustrations. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.
[0069] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing terminal device to operate in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The functions specified in one or more boxes. These computer program instructions may also be loaded onto a computer or other programmable data processing terminal equipment to cause a series of operational steps to be performed on the computer or other programmable terminal equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable terminal equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.
[0070] It should also be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. The terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or terminal device that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or terminal device. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or terminal device that includes said element. Furthermore, the term "and / or" is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, and B alone, where A and B can be singular or plural. Additionally, the character " / " in this text generally indicates an "or" relationship between the preceding and following objects, but it can also indicate an "AND / OR" relationship. Please refer to the context for specific interpretations. "At least one" refers to one or more items, while "more than" refers to two or more items. "At least one of the following" or similar expressions refer to any combination of these items, including any combination of single or multiple items. For example, at least one of a, b, or c can be represented as: a, b, c, ab, ac, bc, or abc, where a, b, and c can be single or multiple.
[0071] Furthermore, it is understood that in various embodiments of the present invention, the order of the above-mentioned process numbers does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.
[0072] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.
[0073] In the several embodiments provided by this invention, it should be understood that the disclosed devices, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative. For instance, the division of functional modules / units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another device, or some features may be ignored or not executed. Furthermore, the shown or discussed mutual couplings or direct couplings or communication connections may be through some interfaces; indirect couplings or communication connections between devices or units may be electrical, mechanical, or other forms. Units described as separate components may or may not be physically separate, and components shown as units may or may not be physical units, i.e., they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs. Additionally, the functional units in the various embodiments of this invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.
[0074] If the method is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.
[0075] Finally, it should be noted that the above description is merely a preferred embodiment of the present invention. It should be pointed out that although preferred embodiments of the present invention have been described, those skilled in the art, once they understand the basic inventive concept of the present invention, can make several improvements and modifications without departing from the principles described herein. These improvements and modifications should also be considered within the scope of protection of the present invention. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of the embodiments of the present invention.
Claims
1. A robot path planning method based on adaptive fuzzy and hybrid strategies, characterized in that, include: The robot senses its surroundings and obtains environmental information. Based on the RRT-Connect algorithm framework, the improved RRT-Connect algorithm is obtained by integrating fuzzy control theory, hierarchical hybrid extension mechanism and dynamic path optimization idea; The improved RRT-Connect algorithm is used to realize robot path planning based on environmental information.
2. The robot path planning method based on adaptive fuzzy and hybrid strategies as described in claim 1, characterized in that, The environmental information includes: the Euclidean distance between the node in the current tree to be expanded that is closest to the other tree and the other tree. and the openness of the environment FSO .
3. The robot path planning method based on adaptive fuzzy and hybrid strategies as described in claim 2, characterized in that, Openness of the environment FSO The acquisition method is as follows: starting from the end node of the current tree, emit evenly to the surrounding space. N A detection ray is used to count the number of rays that do not collide with obstacles. ,by and N The ratio of the two values is used as the degree of openness of the environment. FSO ;in, N This is a preset integer value.
4. The robot path planning method based on adaptive fuzzy and hybrid strategies as described in claim 2, characterized in that, The improved RRT-Connect algorithm, based on the RRT-Connect algorithm framework, adds a fuzzy logic controller, a three-level hybrid extension module, and a tree management and optimization module; among them, The fuzzy logic controller is used to dynamically generate macroscopic policy parameters to guide the current expansion based on fuzzy control theory and environmental information; among these, the macroscopic policy parameters include: dynamic target bias rate. Dynamically expanding step size and the radius of the dynamic reconnection neighborhood ; The three-level hybrid extension module is used for a hierarchical hybrid extension mechanism to perform node generation operations at the micro level based on the macro-strategy parameters output by the fuzzy logic controller. The tree management and optimization module is used to connect newly generated nodes to a random tree and perform local optimization on the path.
5. The robot path planning method based on adaptive fuzzy and hybrid strategies as described in claim 4, characterized in that, Fuzzy logic controllers are specifically used for: take over and FSO And according to the preset membership function, and FSO These are respectively mapped to the membership degrees of different fuzzy sets; where, The fuzzy set is: {Near, Medium, Far}. FSO The fuzzy set is: {Congested, Normal, Open}; The system performs inference based on a pre-defined fuzzy rule base, and converts the fuzzy conclusions of each output variable obtained from the fuzzy inference back to a single numerical value using a defuzzification method. Finally, it outputs a set of macroscopic strategy parameters to guide the current expansion. The fuzzy rule base includes the following rules: Rule 1: IF ( is far) AND ( FSO is open) THEN ( is high) AND ( isbig) AND ( is (big); Rule 2: IF ( is near) AND ( FSO is crowded) THEN ( is low) AND ( is small) AND ( (is small); Rule 3: IF ( is medium) AND ( FSO is ordinary) THEN ( is medium) AND ( is medium) AND ( is medium).
6. The robot path planning method based on adaptive fuzzy and hybrid strategies as described in claim 4, characterized in that, The three-level hybrid extension module is specifically used for: First, the first-level strategy is executed, based on the output of the fuzzy logic controller. Generate a random number in the range [0, 1] with probability. ,like Less than Then the direction of expansion is determined to be toward the other tree; Otherwise, abandon the target-biased growth and directly trigger the secondary strategy; Select the node closest to the expansion direction from the current tree. Subsequently, the output of the fuzzy logic controller is along the expansion direction. Extend the process to generate a primary candidate node. ;right arrive Collision detection is performed on the connection path; if no collision is found, then... Considered as a valid new node in this iteration If a collision occurs, the first-level strategy is considered to have failed, and the second-level strategy is triggered. When the first-level strategy fails, at the end node of the current expansion. Calculate an escape force ; And along the escape force direction, the output of the fuzzy logic controller Extend the process to generate an escape candidate node. ;right Perform collision detection; if no collision occurs, then... Considered as a valid new node in this iteration ; If a collision still occurs, the secondary strategy is considered to have failed, and the tertiary strategy is triggered. When both the primary and secondary strategies fail, then... Centered on, with a radius of Select a node within the sphere as a backup candidate node. ; then on arrive Collision detection is performed on the connection path; if no collision is found, then... Considered as a valid new node in this iteration If a collision still occurs, the current iteration is considered to have failed, and no new node will be generated in this iteration; the iteration will wait for the next iteration.
7. The robot path planning method based on adaptive fuzzy and hybrid strategies as described in claim 6, characterized in that, escape force The calculation method is as follows: First, calculate the reinforcing repulsion force. : ; in, It is the repulsive force gain coefficient; yes Distance to the nearest obstacle; It is the range of the repulsive force; Then the rotational gravity was calculated. : ; in, It is the gravitational gain coefficient; It is the standard gravitational force pointing towards the target; It is a virtual gravitational force pointing in a random direction; Finally, the total escape force is synthesized. : .
8. The robot path planning method based on adaptive fuzzy and hybrid strategies as described in claim 4, characterized in that, The tree management and optimization module is specifically used for: Will The algorithm's concepts of reselecting parent nodes and reconnecting, and the output of the fuzzy logic controller. By combining these methods, a parent node reselection and reconnection mechanism is implemented within the RRT-Connect algorithm framework. This mechanism is followed by a process that occurs after a new node is successfully added. Path reconstruction is performed within a defined neighborhood to adaptively optimize the path.
