Path planning method and device based on path optimization technology
By employing parameter initialization, collision detection, and path optimization techniques, a low-cost and efficient path planning method is generated, addressing the shortcomings of existing algorithms in terms of resource consumption and stability, and achieving efficient path planning in complex scenarios.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-12-30
- Publication Date
- 2026-04-14
AI Technical Summary
Existing path planning algorithms have shortcomings in terms of computational speed, resource consumption, stability, and reproducibility. They are particularly difficult to run efficiently on resource-constrained devices and are prone to getting trapped in local optima.
By using initialization parameter settings, collision detection, redundancy detection, and path optimization techniques, a reasonable and low-cost path planning method is generated, including ray detection, fast random trees, potential field method, hash table detection, and path optimization algorithms. Combined with Gaussian filtering, nearest distance, and reinforcement learning, the path planning is dynamically adjusted.
Without significantly increasing complexity, it generates more reasonable and lower-cost paths, improves computational speed and stability, avoids local optima, and enhances the reproducibility and adaptability of complex scenarios.
Smart Images
Figure CN121855501A_ABST
Abstract
Description
Technical Field
[0001] This disclosure relates to path planning technology, and in particular to a path planning method and apparatus based on path optimization technology. Background Technology
[0002] Currently, previous research on path planning problems has focused on optimizing algorithms such as RRT, A, Dijkstra's algorithm, PRM, genetic algorithms, and ant colony algorithms to address issues like high path costs, slow computation speed, susceptibility to local optima, and poor reproducibility. To reduce path costs, some studies have improved heuristic functions to more accurately guide the search direction, thereby finding lower-cost paths. Regarding slow computation speed, some optimization strategies employ parallel computing techniques, leveraging the advantages of multi-core processors to handle multiple search tasks simultaneously, significantly improving computational efficiency. To address the predicament of getting trapped in local optima, some research has introduced random perturbation mechanisms, adding random factors during the algorithm's search process to help it escape local optima and continue exploring better solutions. To improve reproducibility, researchers have established unified standard testing environments and parameter specifications to ensure consistency of algorithm results under identical conditions.
[0003] Despite the optimizations mentioned above, these algorithms still face some insurmountable problems. For example, while some optimization methods reduce path costs, they increase algorithm complexity, leading to a significant increase in computational resource consumption; parallel computing, while accelerating speed, places high demands on hardware, limiting its application on resource-constrained devices; random perturbation mechanisms affect the stability of the algorithms to some extent, resulting in a degree of randomness in each run; and while unified standards improve reproducibility, their applicability is challenged in complex and ever-changing real-world scenarios.
[0004] Therefore, one or more methods are needed to solve the above problems.
[0005] It should be noted that the information disclosed in the background section above is only used to enhance the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention
[0006] The purpose of this disclosure is to provide a path planning method and apparatus based on path optimization technology, thereby overcoming, at least to some extent, one or more problems caused by the limitations and defects of related technologies.
[0007] According to one aspect of this disclosure, a path planning method based on path optimization technology is provided, comprising: Initial parameters are generated by initializing the parameters of the path planning, and the path is extended from the preset starting point according to the initial parameters. Based on the collision detection algorithm, an extension path is generated by finding a path that extends towards the endpoint. According to the duplicate point detection method, redundant points in the extended path are detected, and the detected redundant points are removed according to the redundant point removal algorithm to generate an initial path. The initial path is optimized using path optimization techniques to generate an output path, thus completing the path planning.
[0008] In one exemplary embodiment of this disclosure, extending the path from a preset starting point includes: Based on the ray detection method, a preset starting point is set as the center point, and rays are emitted at preset angle intervals to extend the ray. When an obstacle is detected, the extension of the ray stops, and the path is extended again by emitting rays at preset angle intervals by setting the stopping point as the center point. Based on the fast random tree algorithm, sample points are generated by randomly sampling the space, and the path is extended by connecting the preset starting point with the sample points. Based on the potential field method, the target point is regarded as the source of attraction and the obstacle is regarded as the source of repulsion. By calculating the resultant force between the target point and the obstacle, the path extends in that direction when the calculated resultant force is greater than the preset resultant force value, and the path extends away from that direction when the calculated resultant force is not greater than the preset resultant force value.
[0009] In one exemplary embodiment of this disclosure, the collision detection includes: geometric collision detection based on geometric algorithms, deep learning collision detection based on deep learning, ultrasonic collision detection based on ultrasound, and infrared collision detection based on infrared sensors.
[0010] In one exemplary embodiment of this disclosure, finding a path extending towards the endpoint includes: Based on the collision detection algorithm, when an obstacle or boundary is detected in the extension direction of the path, the search strategy for the ray is adjusted to move away from the obstacle or boundary. Based on the collision detection algorithm, when the endpoint is detected in the extended environment of the path, the search strategy for the ray is adjusted to improve accuracy, and an extended path is generated.
[0011] In one exemplary embodiment of this disclosure, detecting redundant points in the extended path includes: Based on the hash table detection algorithm, the current node in the path extension direction is mapped by the hash function to generate the current hash value; When the current hash value is detected to exist in the hash table, and the current node overlaps with a node in the extended path, the current node is set as a redundant point. When it is detected that the current hash value does not exist in the hash table, the current node does not overlap with the nodes in the extended path, the extended path is extended, and an initial path is generated; Based on the spatial partitioning detection algorithm, the space of path planning is divided into fixed grid cells of a preset size, and the partitioned grid cells contain nodes in the extended path. Based on the coordinates of the current node, map the current node into the mesh cell; The current node is compared with the nodes in the grid cell. When the current node is found to overlap with the nodes in the grid cell, the current node is set as a redundant point. When it is detected that the current node does not overlap with the nodes in the grid cell, the extension path is extended to generate an initial path.
[0012] In one exemplary embodiment of this disclosure, the initial path is optimized according to a path optimization technique, including: According to the Gaussian filtering method, a smooth path is generated by smoothing the tortuosity of the initial path; Based on the nearest distance algorithm, the shortest path is generated by optimizing the distance between the smooth path and the obstacles; Based on the path output method of reinforcement learning, obstacles, boundaries and preset destinations in the environment are detected in real time, and the shortest path is adaptively adjusted and planned in the dynamic environment to generate an output path, thus completing the path planning.
[0013] In one exemplary embodiment of this disclosure, it further includes: When the cost of finding the endpoint in the current extension direction is higher than the preset cost, the extension direction of the path is replanned by adjusting the preset angle based on environmental feature information. When the initial path planning fails, the current node is backtracked to the previous node through a backtracking mechanism, and the extension direction of the path is replanned based on environmental feature information.
[0014] In one aspect of this disclosure, a path planning apparatus based on path optimization technology is provided, comprising: The initialization parameter setting module is used to generate initial parameters by initializing the parameters of the path planning, and to extend the path from the preset starting point according to the initial parameters. The path extension and collision detection module is used to generate an extension path by finding a path that extends towards the endpoint. The duplicate point detection module is used to detect redundant points in the extended path and remove the detected redundant points according to the redundant point removal algorithm to generate an initial path. The path optimization module is used to optimize the initial path, generate an output path, and complete the path planning.
[0015] In one aspect of this disclosure, an electronic device is provided, comprising: Processor; and A memory storing computer-readable instructions that, when executed by the processor, implement the method according to any one of the preceding claims.
[0016] In one aspect of this disclosure, a computer-readable storage medium is provided having a computer program stored thereon that, when executed by a processor, implements the method according to any one of the preceding claims.
[0017] Based on the embodiments of this disclosure, initial parameters are generated by initializing the parameters of path planning, and the path is extended from a preset starting point according to these initial parameters. An extended path is generated by finding a path extending towards the endpoint based on a collision detection algorithm. Redundant points in the extended path are detected and removed using a redundancy removal algorithm, generating the initial path. The initial path is then optimized using path optimization techniques to generate an output path, thus completing the path planning. Therefore, the embodiments of this disclosure provide an optimized path planning method that generates more reasonable and lower-cost paths without significantly increasing complexity, while improving computational speed and stability, avoiding local optima problems, and enhancing reproducibility and adaptability in complex scenarios.
[0018] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this disclosure.
[0019] The technical solutions of this disclosure will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description
[0020] The accompanying drawings, which form part of this specification, illustrate embodiments of this disclosure and, together with the description, serve to explain the principles of this disclosure.
[0021] This disclosure will become clearer with reference to the accompanying drawings and the following detailed description, wherein: Figure 1 A flowchart illustrating a path planning method based on path optimization technology, which is one embodiment of the method disclosed herein; Figure 2 The following is a flowchart illustrating the decision logic of a path planning method based on path optimization technology, which is an embodiment of the method disclosed herein. Figure 3A comparison diagram of path optimization for a path planning method based on path optimization technology, which is an embodiment of the method disclosed herein; Figure 4 Two comparison figures illustrating a path optimization method based on path optimization technology, representing an embodiment of the method disclosed herein; Figure 5 This is a structural block diagram of a path planning device based on path optimization technology, which is an embodiment of the method disclosed herein. Figure 6 This is a system architecture diagram of a path planning device based on path optimization technology, which is an embodiment of the method disclosed herein. Figure 7 This is a block diagram of an electronic device according to an embodiment of the method of this disclosure. Detailed Implementation
[0022] Exemplary embodiments will now be described more fully with reference to the accompanying drawings. However, these exemplary embodiments can be implemented in many forms and should not be construed as limited to the embodiments set forth herein; rather, they are provided so that this disclosure will be thorough and complete, and will fully convey the concept of the exemplary embodiments to those skilled in the art. The same reference numerals in the drawings denote the same or similar parts, and therefore repeated descriptions of them will be omitted.
[0023] Furthermore, the described features, structures, or characteristics can be combined in any suitable manner in one or more embodiments. Numerous specific details are provided in the following description to give a thorough understanding of embodiments of this disclosure. However, those skilled in the art will recognize that the technical solutions of this disclosure can be practiced without one or more of the specific details described, or other methods, components, materials, apparatuses, steps, etc., can be employed. In other instances, well-known structures, methods, apparatuses, implementations, materials, or operations are not shown or described in detail to avoid obscuring various aspects of this disclosure.
[0024] The block diagrams shown in the accompanying drawings are merely functional entities and do not necessarily correspond to physically independent entities. That is, these functional entities can be implemented in software, or in one or more software-hardened modules, or in different network and / or processor devices and / or microcontroller devices.
[0025] In this disclosure, a path planning method based on path optimization technology is first provided; refer to Figure 1 As shown, this path planning method based on path optimization technology may include the following steps: Step S110: Initial parameters are generated by initializing the parameters of the path planning, and the path is extended from the preset starting point according to the initial parameters.
[0026] Step S120: Based on the collision detection algorithm, an extension path is generated by finding a path that extends towards the endpoint.
[0027] Step S130: According to the duplicate point detection method, redundant points in the extended path are detected, and the detected redundant points are removed according to the redundant point removal algorithm to generate an initial path.
[0028] Step S140: Optimize the initial path according to path optimization technology to generate an output path and complete the path planning.
[0029] Below, as Figure 2 As shown, a path planning method based on path optimization technology in an embodiment of this disclosure will be further described.
[0030] In step S110, initial parameters can be generated by initializing the parameters of the path planning, and the path can be extended from the preset starting point according to the initial parameters.
[0031] Existing path planning algorithms often optimize path costs by improving heuristic functions or increasing algorithm complexity. (For example, algorithm A requires maintaining a priority queue, and each iteration needs to evaluate and sort all neighboring nodes.) This can lead to a significant increase in computational resource consumption in complex environments. For resource-constrained devices (such as embedded systems or mobile devices), such computationally complex algorithms are difficult to run efficiently. Therefore, in some alternative embodiments of this example, path costs and computational complexity are reduced by simplifying the path search logic. Different methods can also be selected based on different path environments.
[0032] First, initialize the input parameters, including the start point, end point, map boundaries, obstacles, and other information. Simultaneously, initialize relevant parameters based on the selected path expansion method.
[0033] Method 1: Path Extension Based on Ray Detection. When using ray detection, first, initialize preset parameters such as angle intervals and initial step size. Then, using the starting point as the center, emit rays in multiple directions at preset angle intervals to extend the ray. For each ray, gradually increase the ray length (increasing the initial step size each time). When an obstacle or boundary is detected, stop the ray extension, and using the stopping point as the center, emit rays again in multiple directions at preset angle intervals until one ray reaches the endpoint. Record the length and direction of each ray for subsequent path optimization.
[0034] Method 2, based on path extension using a fast random tree. When using the fast random tree algorithm, first, the number of sample points, sampler, and other parameters are initialized. In each iteration, the fast random tree randomly selects a sample point; this sample point has random coordinates generated by the sampler. Then, the algorithm checks whether a path from an existing node in the fast random tree to this sample point is feasible. If so, the algorithm adds this random point to the fast random tree, causing the fast random tree to gradually expand towards the destination region until the expanded tree finds a path from the starting point to the destination. This method is particularly suitable for high-dimensional spaces and complex environments.
[0035] Method 3: Path extension based on potential field. When using the potential field method, first, initialize preset parameters such as the resultant force value, the magnitude of attraction, and the magnitude of repulsion. Treat the target point as the source of attraction and obstacles as sources of repulsion. Calculate the resultant force between the target point and the obstacles. When the calculated resultant force is greater than the preset resultant force value, the path extends in that direction; when the calculated resultant force is not greater than the preset resultant force value, the path extends away from that direction. This allows the path to gradually expand towards the endpoint region until a path from the starting point to the endpoint is connected. This method is more suitable for real-time navigation.
[0036] In step S120, an extension path can be generated by finding a path that extends towards the endpoint based on a collision detection algorithm.
[0037] When searching for a path extending towards the endpoint, it is also necessary to dynamically adjust the search direction and step size. In some optional embodiments of this example, the step size is dynamically adjusted using a collision detection algorithm based on the distance from the current point to the endpoint and the distribution of nearby obstacles.
[0038] Regarding collision detection algorithms, in addition to geometric collision detection (such as ray detection) based on geometric algorithms, ultrasonic collision detection and infrared collision detection based on ultrasonic or infrared sensors can be used as alternatives. These methods are suitable for close-range detection, but have limited detection range. Furthermore, when computing power allows, deep learning-based collision detection can be used. This involves using convolutional neural networks (CNNs) to analyze environmental images acquired by sensors in real time, predicting obstacle positions and collision risks.
[0039] Regarding search strategy adjustments, when an obstacle or boundary is detected in the path's extension direction, the search strategy for the ray is adjusted to move away from the obstacle or boundary. Furthermore, when an endpoint is detected in the path's extension environment, the search strategy for the ray is adjusted to improve accuracy, ultimately generating the extended path.
[0040] For example, if the distance from the current point to the destination is less than 5 units (adjusted according to the actual situation), the step size is appropriately reduced to improve accuracy. Alternatively, if obstacles are densely distributed near the current point, the step size is dynamically adjusted based on the distance to the obstacles to avoid collisions. Furthermore, if the path cost suddenly increases, other directions are prioritized for exploration to avoid getting trapped in local optima.
[0041] This makes the algorithm used in this example rely mainly on basic mathematical operations, without the need for a complex parallel computing framework, thus enabling it to run efficiently on a single-core processor or low-power device.
[0042] In step S130, the redundant points in the extended path can be detected by the duplicate point detection method, and the detected redundant points can be removed by the redundant point removal algorithm to generate the initial path.
[0043] In some of the alternative methods for generating extended paths in this example, redundant points will inevitably appear during the path generation process. This necessitates the introduction of new algorithms to remove redundant points from the path and improve its simplicity.
[0044] Method 1: Redundant point removal based on hash table detection algorithm. In redundant point detection, a hash table can be used to store the points to be generated. Whenever a new node is generated, a hash function is used to map the current node to a hash value, and it is checked whether the generated hash value already exists in the hash table. If the hash value exists in the hash table, the newly generated current node overlaps with nodes in the extended path, and the current node is set as a redundant point; if the hash value does not exist in the hash table, the newly generated current node does not overlap with nodes in the extended path, the extended path is extended, and an initial path is generated. This method is more suitable for scenarios with a large number of points and frequent redundant point checks.
[0045] Method 2 involves redundant point removal based on a spatial partitioning detection algorithm. The path planning space is divided into fixed grid cells of a preset size (set according to device performance and desired accuracy), each containing one or more points. Whenever a new node is generated, its coordinates are mapped to a grid cell. This new node is then compared with other points within the same grid cell, or with points in neighboring grid cells (neighboring cells are checked because grid boundaries may overlap, meaning some points may be located on the edges of adjacent cells). When the current node is detected to overlap with a node in the mapped grid cell, it is designated as a redundant point; when the current node is detected not to overlap with a node in the mapped grid cell, the path is extended to generate the initial path. This method reduces the number of point pairs that need to be checked through spatial partitioning, significantly improving efficiency, especially in cases of sparse points. Furthermore, the grid partitioning method is simple to implement, requires no complex data structures, and is suitable for situations with relatively uniform point distribution.
[0046] In step S140, the initial path can be optimized using path optimization techniques to generate an output path, thus completing the path planning.
[0047] In some alternative embodiments of this example, after the initial path is generated, multiple objectives such as path cost, computational complexity, stability, reproducibility, and scene adaptability need to be considered simultaneously. Through dynamic adjustment strategies and adaptive parameter mechanisms, comprehensive optimization of multiple objectives can be achieved.
[0048] First, a smooth path can be generated by smoothing the tortuosity of the initial path using Gaussian filtering. Alternatively, a path optimization method based on genetic algorithms can be used to gradually improve the initial path through selection, crossover, and mutation operations.
[0049] Then, based on the nearest distance algorithm (in addition to calculating Euclidean distance, if uncertainty and dynamic environment need to be considered, a distance judgment method based on fuzzy logic can also be used. Furthermore, if the computing power is sufficient, a path output method based on reinforcement learning (such as Q-learning) can be used instead), the shortest path is generated by optimizing the distance between the smooth path and the obstacles.
[0050] Furthermore, based on the path output method of reinforcement learning, by real-time detection of obstacles, boundaries, and preset endpoints in the environment, or by adopting a path optimization method based on dynamic programming, the shortest path is adjusted and optimized in stages to gradually reduce the path length and redundant points, generate the output path, and complete the path planning.
[0051] In a specific example, such as Figures 3-4As shown, an environment with multiple obstacles is set up with a boundary of 15×15. The starting point is set at position (1,1) in the environment, and the ending point is set at position (12,12) in the environment.
[0052] The path planning algorithm described above emits rays from the starting point in multiple directions to detect the feasibility of each path and dynamically adjusts the search direction and step size based on environmental characteristics. During path planning, assuming the distance from the current point to the destination is 4 units and there is an obstacle nearby, the system will dynamically adjust the step size to 1.5 units and prioritize exploring other directions to avoid collisions with obstacles and getting trapped in local optima.
[0053] After establishing the initial path from the start point to the end point, the initial path is smoothed by Gaussian filtering to transform it from a polyline into a smooth curve. Then, erosion processing removes redundant points from the path, and the distance between the path and obstacles is optimized using the nearest distance algorithm. Finally, the shortest path from the start point to the end point is output. Thus, this invention solves the problems of path redundancy, low computational efficiency, and insufficient dynamic obstacle avoidance capabilities in existing technologies by employing techniques such as dynamic collision detection, duplicate point detection, path optimization, and shortest path generation.
[0054] Alternatively, in any of the above embodiments of this disclosure, getting trapped in a local optimum can also be avoided in the following ways: Dynamic adjustment strategy: The search direction is intelligently adjusted based on environmental characteristics and path cost trends. For example, when a sudden increase in path cost is detected, causing the path cost to the destination to exceed the preset cost, the angle is adjusted to prioritize exploration in other directions, avoiding getting stuck in the optimal solution in a local area.
[0055] Backtracking mechanism: When path planning fails, the current node is automatically backtracked to the previous node, and the path is replanned based on the new environmental information to ensure that the globally optimal solution can be found.
[0056] It should be noted that although the steps of the method in this disclosure are described in a specific order in the accompanying drawings, this does not require or imply that the steps must be performed in that specific order, or that all the steps shown must be performed to achieve the desired result. Additional or alternative steps may be omitted, multiple steps may be combined into one step, and / or a step may be broken down into multiple steps.
[0057] Furthermore, in this example embodiment, a path planning device based on path optimization technology is also provided. (Refer to...) Figure 5As shown, the path planning device 300 based on path optimization technology may include: an initialization parameter setting module 310, a path extension and collision detection module 320, a duplicate point detection module 330, and a path optimization module 340. Wherein: The initialization parameter setting module 310 is used to generate initial parameters by initializing the parameters of the path planning, and extend the path from the preset starting point according to the initial parameters. The path extension and collision detection module 320 is used to generate an extension path by finding a path that extends towards the endpoint; The duplicate point detection module 330 is used to detect redundant points in the extended path and remove the detected redundant points according to the redundant point removal algorithm to generate an initial path. The path optimization module 340 is used to optimize the initial path, generate an output path, and complete the path planning.
[0058] In addition, such as Figure 6 As shown, when developing this path planning device 300 based on path optimization technology as a system, it is also necessary to set up input devices, output devices, distance judgment and path output modules, sensors, dynamic environment update modules, feedback control modules, etc.
[0059] The system includes an input device for receiving user input, an output device for displaying planning results, a distance judgment and path output module for determining whether the distance between the current path and the destination exceeds a preset threshold and outputting the final path, a sensor for receiving real-time environmental data from the map, a dynamic environment update module for updating obstacle information based on the received real-time environmental data, and a feedback control module for dynamically adjusting algorithm parameters based on feedback from the dynamic environment update module.
[0060] The path planning device based on path optimization technology in this disclosure corresponds to the path planning method based on path optimization technology described above, and the relevant content can be referred to each other, which will not be repeated here. The beneficial technical effects of the path planning device based on path optimization technology in this disclosure can be found in the corresponding beneficial technical effects in the above-described exemplary method section, which will not be repeated here.
[0061] It should be noted that although several modules or units of the path planning device 300 based on path optimization technology have been mentioned in the detailed description above, this division is not mandatory. In fact, according to embodiments of this disclosure, the features and functions of two or more modules or units described above can be embodied in one module or unit. Conversely, the features and functions of one module or unit described above can be further divided and embodied by multiple modules or units.
[0062] Below, for reference Figure 7 This describes an electronic device according to embodiments of the present disclosure. The electronic device may be either or both of a first device and a second device, or a standalone device independent of them, which may communicate with the first device and the second device to receive acquired input signals from them.
[0063] Figure 7 A block diagram of an electronic device according to an embodiment of the present disclosure is shown.
[0064] like Figure 7 As shown, the electronic device includes one or more processors and memory.
[0065] A processor can be a central processing unit (CPU) or other form of processing unit with data processing and / or instruction execution capabilities, and can control other components in an electronic device to perform desired functions.
[0066] The memory can store one or more computer program products, and the memory can include various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory. The volatile memory may include, for example, random access memory (RAM) and / or cache memory. The non-volatile memory may include, for example, read-only memory (ROM), hard disk, flash memory, etc. One or more computer program products can be stored on the computer-readable storage medium, and the processor can run the computer program products to implement the methods of the various embodiments of this disclosure described above and / or other desired functions.
[0067] In one example, the electronic device may also include input devices and output devices, which are interconnected via a bus system and / or other forms of connection mechanism (not shown).
[0068] In addition, the input device may also include, for example, a keyboard, a mouse, etc.
[0069] This output device can output various information to the outside, including determined distance information, direction information, etc. The output device may include, for example, a display, a speaker, a printer, and a communication network and its connected remote output devices, etc.
[0070] Of course, for the sake of simplicity, Figure 7 Only some of the components of the electronic device relevant to this disclosure are shown, omitting components such as buses, input / output interfaces, etc. In addition, the electronic device may include any other suitable components depending on the specific application.
[0071] In addition to the methods and apparatus described above, embodiments of this disclosure may also be computer program products comprising computer program instructions that, when executed by a processor, cause the processor to perform the steps in the methods of various embodiments of this disclosure described in the foregoing portion of this specification.
[0072] The computer program product can be written in any combination of one or more programming languages to perform the operations of the embodiments of this disclosure. The programming languages include object-oriented programming languages such as Java and C++, as well as conventional procedural programming languages such as C or similar languages. The program code can be executed entirely on a user's computing device, partially on a user's computing device, as a standalone software package, partially on a user's computing device and partially on a remote computing device, or entirely on a remote computing device or server.
[0073] Furthermore, embodiments of this disclosure may also be computer-readable storage media storing computer program instructions that, when executed by a processor, cause the processor to perform the steps in the methods of various embodiments of this disclosure described in the foregoing portion of this specification.
[0074] The computer-readable storage medium may be any combination of one or more readable media. A readable medium may be a readable signal medium or a readable storage medium. A readable storage medium may be, for example, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of readable storage media (a non-exhaustive list) include: an electrical connection having one or more wires, a portable disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof.
[0075] The basic principles of this disclosure have been described above with reference to specific embodiments. However, it should be noted that the advantages, benefits, and effects mentioned in this disclosure are merely examples and not limitations, and should not be considered as essential features of each embodiment of this disclosure. Furthermore, the specific details disclosed above are for illustrative and facilitative purposes only, and are not limitations. These details do not limit the scope of this disclosure to the necessity of employing the aforementioned specific details for implementation.
[0076] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For system embodiments, since they largely correspond to method embodiments, the description is relatively simple; relevant parts can be referred to the descriptions in the method embodiments.
[0077] The block diagrams of devices, apparatuses, devices, and systems disclosed herein are merely illustrative examples and are not intended to require or imply that they must be connected, arranged, or configured in the manner shown in the block diagrams. As those skilled in the art will recognize, these devices, apparatuses, devices, and systems can be connected, arranged, and configured in any manner. Words such as “comprising,” “including,” “having,” etc., are open-ended terms meaning “including but not limited to,” and are used interchangeably with them. The terms “or” and “and” as used herein refer to the terms “and / or,” and are used interchangeably with them unless the context clearly indicates otherwise. The term “such as” as used herein refers to the phrase “such as but not limited to,” and is used interchangeably with it.
[0078] The methods and apparatus of this disclosure may be implemented in many ways. For example, they may be implemented by software, hardware, firmware, or any combination of software, hardware, and firmware. The above-described order of steps for the methods is for illustrative purposes only, and the steps of the methods of this disclosure are not limited to the order specifically described above unless otherwise specifically stated. Furthermore, in some embodiments, this disclosure may also be implemented as a program recorded on a recording medium, the program including machine-readable instructions for implementing the methods according to this disclosure. Thus, this disclosure also covers recording media storing programs for performing the methods according to this disclosure.
[0079] It should also be noted that in the apparatus, devices, and methods of this disclosure, the components or steps can be disassembled and / or recombined. These disassemblies and / or recombinations should be considered as equivalent solutions to this disclosure.
[0080] The above description of the disclosed aspects is provided to enable any person skilled in the art to make or use this disclosure. Various modifications to these aspects will be readily apparent to those skilled in the art, and the general principles defined herein may be applied to other aspects without departing from the scope of this disclosure. Therefore, this disclosure is not intended to be limited to the aspects shown herein, but rather to be carried out within the widest scope consistent with the principles and novel features disclosed herein.
[0081] The above description has been given for purposes of illustration and description. Furthermore, this description is not intended to limit the embodiments of this disclosure to the forms disclosed herein. Although numerous exemplary aspects and embodiments have been discussed above, those skilled in the art will recognize certain variations, modifications, alterations, additions, and sub-combinations therein.
Claims
1. A path planning method based on path optimization technology, characterized in that, include: Initial parameters are generated by initializing the parameters of the path planning, and the path is extended from the preset starting point according to the initial parameters. Based on the collision detection algorithm, an extension path is generated by finding a path that extends towards the endpoint. According to the duplicate point detection method, redundant points in the extended path are detected, and the detected redundant points are removed according to the redundant point removal algorithm to generate an initial path. The initial path is optimized using path optimization techniques to generate an output path, thus completing the path planning.
2. The method according to claim 1, characterized in that, Extending the path from the preset starting point includes: Based on the ray detection method, a preset starting point is set as the center point, and rays are emitted at preset angle intervals to extend the ray. When an obstacle is detected, the extension of the ray stops, and the stopping point of the ray extension is set as the center point. Rays are then emitted again at preset angle intervals to extend the path. Based on the fast random tree algorithm, sample points are generated by randomly sampling the space, and the path is extended by connecting the preset starting point with the sample points. Based on the potential field method, the target point is regarded as the source of attraction and the obstacle is regarded as the source of repulsion. The resultant force between the target point and the obstacle is calculated. When the calculated resultant force is greater than the preset resultant force value, the path extends in that direction. When the calculated resultant force is not greater than the preset resultant force value, the path extends away from that direction.
3. The method according to claim 1, characterized in that, The collision detection includes: geometric collision detection based on geometric algorithms, deep learning collision detection based on deep learning, ultrasonic collision detection based on ultrasound, and infrared collision detection based on infrared sensors.
4. The method according to claim 3, characterized in that, By finding paths that extend towards the endpoint, including: Based on the collision detection algorithm, when an obstacle or boundary is detected in the extension direction of the path, the search strategy for the ray is adjusted to move away from the obstacle or boundary. Based on the collision detection algorithm, when the endpoint is detected in the extended environment of the path, the search strategy for the ray is adjusted to improve accuracy, and an extended path is generated.
5. The method according to claim 1, characterized in that, By detecting redundant points in the extended path, including: Based on the hash table detection algorithm, the current node in the path extension direction is mapped by the hash function to generate the current hash value; When the current hash value is detected to exist in the hash table, and the current node overlaps with a node in the extended path, the current node is set as a redundant point. When it is detected that the current hash value does not exist in the hash table, the current node does not overlap with the nodes in the extended path, the extended path is extended, and an initial path is generated; Based on the spatial partitioning detection algorithm, the space of path planning is divided into fixed grid cells of a preset size, and the partitioned grid cells contain nodes in the extended path. Based on the coordinates of the current node, map the current node into the mesh cell; The current node is compared with the nodes in the grid cell. When the current node is found to overlap with the nodes in the grid cell, the current node is set as a redundant point. When it is detected that the current node does not overlap with the nodes in the grid cell, the extension path is extended to generate an initial path.
6. The method according to claim 1, characterized in that, The initial path is optimized using path optimization techniques, including: According to the Gaussian filtering method, a smooth path is generated by smoothing the tortuosity of the initial path; Based on the nearest distance algorithm, the shortest path is generated by optimizing the distance between the smooth path and the obstacles; Based on the path output method of reinforcement learning, obstacles, boundaries and preset destinations in the environment are detected in real time, and the shortest path is adaptively adjusted and planned in the dynamic environment to generate an output path, thus completing the path planning.
7. The method according to any one of claims 1-6, characterized in that, Also includes: When the cost of finding the endpoint in the current extension direction is higher than the preset cost, the extension direction of the path is replanned by adjusting the preset angle based on environmental feature information. When the initial path planning fails, the current node is backtracked to the previous node through a backtracking mechanism, and the extension direction of the path is replanned based on environmental feature information.
8. A path planning device based on path optimization technology, characterized in that, include: The initialization parameter setting module is used to generate initial parameters by initializing the parameters of path planning, and to extend the path from a preset starting point according to the initial parameters. The path extension and collision detection module is used to generate an extension path by finding a path that extends towards the endpoint. The duplicate point detection module is used to detect redundant points in the extended path and remove the detected redundant points according to the redundancy point removal algorithm to generate an initial path. The path optimization module is used to optimize the initial path, generate an output path, and complete the path planning.
9. An electronic device, characterized in that, include: Memory, used to store computer program products; A processor for executing a computer program product stored in the memory, wherein when the computer program product is executed, it implements the method described in any one of claims 1-7.
10. A computer-readable storage medium having computer program instructions stored thereon, characterized in that, When the computer program instructions are executed by the processor, they implement the method described in any one of claims 1-7.