Fire-fighting robot enhanced path planning method and system based on som network

Abstract

This invention discloses an enhanced path planning method and system for firefighting robots based on a Surface Object Model (SOM) network. The method includes: constructing an initial SOM network; selecting a winning neuron from the initial SOM network according to a competitive strategy mechanism; adjusting the weights of the winning neuron and its neighboring neurons in the initial SOM network according to a weight adjustment mechanism considering the direction vector; repeating this process until the neuron weights no longer change, obtaining the initial path for each firefighting robot; performing a first post-processing operation on the initial path of each firefighting robot to obtain a first optimized path; and performing a second post-processing operation on the first optimized path to obtain the final path. This method is of great significance for achieving efficient obstacle avoidance and task collaborative allocation in complex environments.

Description

Technical Field

[0001] This invention relates to the field of path planning technology, and in particular to an enhanced path planning method and system for firefighting robots based on SOM networks. Background Technology

[0002] With the rapid development of technology, firefighting robots have shown great application potential in various fields such as high-rise buildings, petrochemical industrial zones, underground spaces, warehouses, and logistics centers. However, under complex tasks, a single robot is often limited by its processing power, perception range, and response speed, making it difficult to meet the requirements of high real-time performance and high precision in safety rescue.

[0003] Therefore, multi-robot collaborative technology has attracted widespread attention due to its ability to effectively address the challenges of complex tasks. Task allocation and path planning, as core technologies of multi-robot collaboration, directly affect the overall performance and task execution efficiency of multiple firefighting robots. Specifically, task allocation determines which robots are responsible for which tasks, while path planning is responsible for planning the optimal or suboptimal path from the starting point to the destination for the robots, ensuring that the robots can complete the tasks safely and efficiently.

[0004] Therefore, the collaborative optimization of task allocation and path planning for firefighting robots has become an emerging research focus and is particularly important in complex firefighting and rescue mission scenarios.

[0005] Currently, mainstream solutions for task allocation and path planning focus on two strategies:

[0006] One approach is to solve the task allocation result as a prerequisite for path planning independently. Although this approach is logically clear, the coupling between algorithms is low, and the overall solution architecture tends to be complex.

[0007] Secondly, we attempted to couple task allocation with path planning to achieve more efficient collaboration. However, this strategy also faces challenges such as complex collaborative algorithm design and increased difficulty in overall system optimization.

[0008] Crucially, traditional task allocation algorithms are significantly limited in effectiveness and adaptability when dealing with small-scale environments with numerous obstacles, making it difficult to meet the demands for high-precision and high-flexibility task execution.

[0009] Therefore, exploring a practical algorithm that can efficiently coordinate task allocation and path planning, and effectively avoid obstacles in the complex environment of indoor fires, has become one of the urgent technical challenges to be solved.

[0010] In existing research, the SOM neural network is an unsupervised neural network that can automatically cluster data into groups based on the similarity of the input data. In task allocation, this means that the SOM network can automatically group tasks according to some similarity (distance, difficulty, resource requirements, etc.), and at the same time, it can automatically generate a path from the starting point to the target point based on the position of neurons at different times, thus completing path planning while allocating tasks.

[0011] Given the inherent greedy selection characteristic of neurons in SOM neural networks, current SOM-based algorithms often struggle to guarantee the generation of an optimal path that completely avoids obstacles in path planning. This limitation restricts the effectiveness of traditional SOM algorithms in complex and dynamic environments.

[0012] To address existing problems, a common solution is to construct a virtual repulsive field around obstacles to guide the robot to autonomously avoid them during task execution, ensuring operational safety. However, this method faces significant limitations in complex and ever-changing environments. The core issue lies in the fixed and finite number of neurons in the SOM neural network, making it difficult to fully adapt to the dual demands of high-precision obstacle avoidance and efficient task allocation in complex environments. In complex scenarios, the limited number of neurons restricts the flexibility and accuracy of obstacle avoidance strategies and may also lead to conflicts between obstacle avoidance requirements and task allocation objectives, becoming a key factor restricting the overall performance of the system. Summary of the Invention

[0013] To address the shortcomings of existing technologies, this invention provides an enhanced path planning method and system for fire-fighting robots based on the Path Organization Model (SOM) network. It explores an improved SOM algorithm capable of dynamically adjusting neuron configuration and enhancing environmental adaptability, which is of great significance for achieving efficient obstacle avoidance and collaborative task allocation in complex environments. When traditional SOM is applied to task allocation and path planning, the fixed number of neurons is the main reason why this algorithm struggles to cope with the dual challenges of obstacle avoidance and task allocation. Therefore, to resolve the contradiction between task allocation and obstacle avoidance in the SOM neural network algorithm, the key lies in appropriately selecting a neuron growth strategy and dynamically adapting it to changes in the indoor fire environment.

[0014] On the one hand, an enhanced path planning method for firefighting robots based on SOM networks is provided;

[0015] An enhanced path planning method for firefighting robots based on SOM networks includes:

[0016] (1) Construct an initial SOM network; select the winning neuron from the initial SOM network according to the competition strategy mechanism;

[0017] (2) Adjust the weights of the winning neurons and their neighboring neurons in the initial SOM network according to the weight adjustment mechanism that takes into account the direction vector;

[0018] (3) Repeat steps (1) to (2) until the neuron weights no longer change, and obtain the initial path of each fire robot;

[0019] (4) Perform the first post-processing operation on the initial path of each fire-fighting robot to obtain the first optimized path;

[0020] (5) Perform a second post-processing operation on the first optimized path to obtain the final path.

[0021] On the other hand, an enhanced path planning system for firefighting robots based on SOM networks is provided;

[0022] An enhanced path planning system for firefighting robots based on SOM networks includes:

[0023] The building module is configured to: build an initial SOM network; and select the winning neuron from the initial SOM network according to a competition strategy mechanism.

[0024] The weight adjustment module is configured to adjust the weights of the winning neurons and their neighboring neurons in the initialized SOM network according to a weight adjustment mechanism that considers the direction vector.

[0025] The repeating module is configured to repeatedly execute the building module and the weight adjustment module until the neuron weights no longer change, thus obtaining the initial path for each fire-fighting robot.

[0026] The primary processing module is configured to perform the first post-processing operation on the initial path of each fire-fighting robot to obtain the first optimized path.

[0027] The secondary processing module is configured to perform a second post-processing operation on the first optimized path to obtain the final path.

[0028] Furthermore, an electronic device is also provided, including:

[0029] Memory, used for non-transitory storage of computer-readable instructions; and

[0030] Processor, for executing the computer-readable instructions,

[0031] When the computer-readable instructions are executed by the processor, they perform the method described in the first aspect above.

[0032] In another aspect, a storage medium is also provided for non-transitory storage of computer-readable instructions, wherein when the non-transitory computer-readable instructions are executed by a computer, the instructions of the method described in the first aspect are executed.

[0033] In another aspect, a computer program product is also provided, including a computer program that, when run on one or more processors, is used to implement the method described in the first aspect above.

[0034] The above technical solution has the following advantages or beneficial effects:

[0035] This invention innovatively proposes a dual-neuron growth strategy to optimize obstacle avoidance and task allocation for multiple firefighting and rescue robots in complex fire scenes. Simultaneously, it introduces a novel simplified polygonal obstacle avoidance algorithm to ensure collision-free paths. Ultimately, this algorithm enables efficient task allocation and path planning with small turning radii for multiple firefighting and rescue robots in complex environments, ensuring that all robots can complete obstacle avoidance using the minimum path, thereby improving firefighting and rescue efficiency.

[0036] This invention effectively resolves the contradiction between obstacle avoidance and task allocation encountered when using the original SOM (Solution-Oriented Model) for task allocation and path planning by replicating the parent neuron to generate the child neuron and inhibiting the parent neuron's activity. The overall approach is clear and the implementation is simple and convenient.

[0037] This invention achieves a strategy that completely avoids obstacles by adding a new neuron between two adjacent neurons, thereby ensuring that the path generated after path planning will not cross any obstacles, thus guaranteeing the reliability and safety of the path planning process.

[0038] This invention proposes a novel post-processing algorithm that simplifies polylines and obstacle avoidance, reduces path spikes, and lowers curvature. When applied to SOM task allocation and path planning, it generates smooth paths, ensuring that all robots can complete obstacle avoidance with the lowest possible path, thereby improving firefighting and rescue efficiency. Attached Figure Description

[0039] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an improper limitation of the invention.

[0040] Figure 1 This is a schematic diagram of method S102;

[0041] Figure 2 Here is the flowchart for method S102;

[0042] Figure 3 This is a schematic diagram of method S104;

[0043] Figure 4 Here is the flowchart for method S104;

[0044] Figure 5 This is a schematic diagram of method S105, case one;

[0045] Figure 6 This is a schematic diagram of method two for case S105;

[0046] Figure 7 This is a simulation diagram of the final path;

[0047] Among them, 1 is the first position point, 2 is the second position point, and 3 is the third position point. Detailed Implementation

[0048] It should be noted that the following detailed descriptions are exemplary and intended to provide further illustration of the invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0049] Example 1

[0050] This embodiment provides an enhanced path planning method for firefighting robots based on SOM networks;

[0051] An enhanced path planning method for firefighting robots based on SOM networks includes:

[0052] S101: Construct the initial SOM network; select the winning neuron from the initial SOM network according to the competition strategy mechanism;

[0053] S102: Adjust the weights of the winning neurons and their neighboring neurons in the initial SOM network according to the weight adjustment mechanism that considers the direction vector;

[0054] S103: Repeat S101 to S102 until the neuron weights no longer change, and obtain the initial path of each fire-fighting robot;

[0055] S104: Perform the first post-processing operation on the initial path of each fire-fighting robot to obtain the first optimized path;

[0056] S105: Perform a second post-processing operation on the first optimized path to obtain the final path.

[0057] Further, S101: Constructing the initial SOM network includes:

[0058] Based on the number of firefighting robots k and the number of target points m, an initial SOM network is constructed, with k groups of neurons, each group containing m neurons; the weights of each group of neurons are set to the initial position information of the corresponding robot.

[0059] Among them, the neuron weights are represented by R. ij Let i = 1, ..., k, j = 1, ..., m, R ij R represents the weight of the j-th neuron in the i-th group of neurons. ij It also represents the coordinate position of the j-th neuron in the i-th group of neurons.

[0060] It should be understood that the SOM network is divided into an input layer and an output layer. The number of neurons in the input layer is determined by the dimension of the input vector, and one neuron corresponds to one feature. In the output layer, one node represents a clustered class.

[0061] Furthermore, in S101: according to the competition strategy mechanism, the SOM network is initialized, and the winning neuron is selected. The specific competition strategy adopts a greedy strategy.

[0062] A greedy strategy is an algorithm that makes the best or optimal choice in each step of the current state, hoping to achieve the best or optimal result globally.

[0063] The greedy strategy selects the closest neuron as the winner based on the distance between the neuron and the target point, thereby achieving efficient self-organization of the network.

[0064] Furthermore, such as Figure 1 and Figure 2 As shown, S102: According to the weight adjustment mechanism considering the direction vector, the weights of the winning neurons and their neighboring neurons in the initialized SOM network are adjusted, including:

[0065] Step (1-1): Assume neuron R 11 It has become the winning neuron through a greedy strategy;

[0066] Step (1-2): Using a traditional neuron weight adjustment strategy, neuron R... 11 The weights are updated to the position of the first location point 1, and it is checked whether the updated neuron is inside the obstacle.

[0067] Steps (1-3): When neuron R is detected 11When the fire robot is located within an obstacle, the line connecting the fire robot and the first position point 1 is rotated clockwise and counterclockwise, with the fire robot as the origin. The line is rotated twice clockwise to obtain the second position point 2, and twice counterclockwise to obtain the third position point 3. The line is then used as a direction vector.

[0068] The formula for the rotation angle is as follows:

[0069] θ=tβ

[0070] Where θ represents the angle that the vector needs to rotate, t represents the number of times the vector needs to rotate, clockwise is t = t1 (t1 < 0), counterclockwise is t = t2 (t2 > 0), and β represents the size of the angle rotated each time, which can be 15°, 30° and 45°. In this invention, β = 15° is used.

[0071] Furthermore, when the second position point 2 is obtained, t1 = -2, and when the third position point 3 is obtained, t2 = 2;

[0072] Step (1-4): Assume the second position point 2 is outside the obstacle, while the third position point 3 is still inside the obstacle. Therefore, in order for the third position point 3 to be outside the obstacle, t2 ≥ 3 is required. At this time, |t1| < |t2|. Therefore, the second position point 2 obtained by rotating clockwise is selected as the final weight point. At this time, neuron R... 11 Updated to position 2;

[0073] Steps (1-5): The second position point 2 is obtained by rotating the direction vector. Therefore, the parent neuron R is copied at the second position point 2. 11 Obtain sub-neurons And set the parent neuron R 11 No longer participating in the competition process;

[0074] Steps (1-6): Sub-neurons The process involves participating in the competition and repeatedly executing steps (1-1), (1-2), (1-3), (1-4), and (1-5) until the weights of the neurons in the group no longer change, thus obtaining an initial path from the starting point to the ending point.

[0075] Furthermore, the traditional neuron weight adjustment strategy is formulated as follows:

[0076]

[0077] Where β is the learning rate, η is a small positive constant, and D min R is the minimum distance between any two neurons. km (t+1) is the updated weight of the neuron, R km(t) represents the weights of the neuron before the update, T. i (t) is the i-th target point, D ikm The i-th target point and R km The weighted distance of (t), the neighborhood function f(d) j G) is defined as follows:

[0078]

[0079] Where, d j G is the distance between the j-th neuron and the winning neuron, where j = 1, 2, ..., m, G is the neighborhood function gain, and r is the neighborhood radius.

[0080] Furthermore, such as Figure 3 and Figure 4 As shown, S104: For the initial path of each fire-fighting robot, perform the first post-processing operation to obtain the first optimized path, including:

[0081] According to the neuron arrangement order of each robot's initial path, starting from the starting neuron, check each neuron one by one to the last neuron, and check in turn whether there are obstacles between each pair of adjacent neurons;

[0082] If there is an obstacle between the paths connecting adjacent neurons A and B, then a new neuron R is added to the corresponding neuron group. add Its location is between neurons A and B, and the new neuron R add The weights are initialized to the arithmetic mean of the weights of neurons A and B;

[0083] Examine the new neuron R at this time add Whether it is inside an obstacle; if not inside an obstacle, retain the neuron R. add Continue checking each neuron from the starting neuron to the last neuron;

[0084] If the new neuron R add If the object is within an obstacle, draw the perpendicular bisector of path AB. The newly added neuron R will then... add Located on the foot of the perpendicular, with AB as the dividing line, and neuron R add Starting from the initial position, the device moves along the first and second directions away from the foot of the perpendicular. Each movement is a distance *dist*. After each movement of *dist*, it is determined whether the new position is within the obstacle. If the new position is not within the obstacle, then the new position is the neuron R. add Candidate location points are obtained, where candidate location point C is obtained from the first direction and candidate location point D is obtained from the second direction. The calculation is performed from neuron R... addThe number of moves required to reach the two candidate positions C and D from the initial position is calculated. The candidate position C with fewer moves is selected as the final position of the new neuron. The path ACB, which passes through points A, C, and B in sequence, is used as the first optimization path.

[0085] Understandably, this adjustment process aims to guide new neurons "around" obstacles, thereby ensuring the continuity and unobstructedness of the path, and specifically includes the following steps:

[0086] Step (2-1): Check whether a certain path AB passes through an obstacle;

[0087] Step (2-2): After determining that path AB passes through an obstacle, add neuron R at the midpoint of the path. add And test R add Is it inside an obstacle? Confirm R add Once the object falls into the obstacle, the subsequent operation will be performed.

[0088] Step (2-3): Move it a fixed distance dist to both sides along a direction perpendicular to path AB, and record the number of times it moves away from the obstacle, s1 and s2, until it reaches points C and D;

[0089] Step (2-4): Compare the number of times s1 and s2 are moved away from the obstacle, and select point C with fewer moves as the newly added neuron after the update;

[0090] Step (2-5): Connect ACB to form a new path.

[0091] Furthermore, such as Figure 5 As shown, S105: Perform a second post-processing operation on the first optimized path to obtain the final path. Case 1 includes:

[0092] Step (3-1): First, mark the starting point of the path and several designated location points. Assume that the path passes through points C, D, E, and F in sequence. Assume that points C and D are marked points, and points E and F are unmarked points. Marked points are manually set path points that must be passed, while unmarked points are candidate path points in the path.

[0093] Step (3-2): Starting from point C, pass through intermediate path point D and connect CE. Determine whether path point D is a marked point. If path point D is a marked point, it means that the path must pass through point D and cannot be eliminated. Therefore, CE is not connected, and CD and DE are retained as the path.

[0094] Step (3-3): Starting from point D, pass through the intermediate path point E and connect DF. Since the intermediate path point E is not marked, determine whether the line segment DF passes through an obstacle. If the line segment DF passes through an obstacle, it means that DF cannot be used as a simplified path and point E cannot be eliminated.

[0095] Step (3-4): Obtain the final path CDEF.

[0096] Furthermore, if the line DF connecting the preceding point D and the following point F of an unmarked point E in the path passes through an obstacle, then the unmarked point E cannot be eliminated; if the line DF connecting the preceding point D and the following point F of an unmarked point E in the path does not pass through an obstacle, then the unmarked point E can be eliminated.

[0097] Furthermore, such as Figure 6 As shown, S105: Perform a second post-processing operation on the first optimized path to obtain the final path. Case 2 includes:

[0098] Step (4-1): First, mark the starting point of the path and the points that must be passed through to reach the target, indicating that these points are indispensable in the path. Mark point C, and points D, E and F are unmarked.

[0099] Step (4-2): Starting from point C, cross the intermediate path point D and connect CE. Determine whether CE passes through an obstacle and whether the intermediate path point D is a marked point. In this case, both are negative, indicating that the path does not need to pass through point D. Remove point D and connect CE. Delete path CDE and keep path CE.

[0100] Step (4-3): Still starting from point C, cross the intermediate path point E and connect CF. Since CF passes through an obstacle, point E cannot be eliminated. Therefore, CF is not connected, and CE and EF are retained as the path.

[0101] Step (4-4): Obtain the final path CEF.

[0102] It should be understood that the second post-processing operation employs a deflection post-processing algorithm, designed to simplify the polyline path and avoid obstacles, reduce jagged edges, and lower the curvature after kinematic planning, thereby generating a smooth path. The final path is as follows: Figure 7 As shown, R represents the robot, and T represents the target point.

[0103] By introducing a dual-neuron growth strategy and integrating it into the competitive learning mechanism and path post-processing stage of SOM (Solution-Oriented Machine) firefighting, this strategy effectively resolves the conflict between task allocation and obstacle avoidance in complex indoor fire environments, significantly optimizing the task allocation and path planning capabilities of multiple firefighting robots during firefighting and rescue operations. Furthermore, this disclosure innovatively introduces a simplified polygonal obstacle avoidance algorithm. This algorithm intelligently selects two strategies based on different scenarios, not only significantly reducing redundant turns in the path and lowering the overall bend degree of the path, but also ensuring a collision-free path. Ultimately, this algorithm enables efficient task allocation and low-bend path planning for multiple robots in complex environments, optimizing path length while ensuring all robots can safely avoid obstacles, improving firefighting and rescue efficiency, and successfully completing firefighting and rescue tasks.

[0104] Example 2

[0105] This embodiment provides an enhanced path planning system for firefighting robots based on SOM networks;

[0106] An enhanced path planning system for firefighting robots based on SOM networks includes:

[0107] The building module is configured to: build an initial SOM network; and select the winning neuron from the initial SOM network according to a competition strategy mechanism.

[0108] The weight adjustment module is configured to adjust the weights of the winning neurons and their neighboring neurons in the initialized SOM network according to a weight adjustment mechanism that considers the direction vector.

[0109] The repeating module is configured to repeatedly execute the building module and the weight adjustment module until the neuron weights no longer change, thus obtaining the initial path for each fire-fighting robot.

[0110] The primary processing module is configured to perform the first post-processing operation on the initial path of each fire-fighting robot to obtain the first optimized path.

[0111] The secondary processing module is configured to perform a second post-processing operation on the first optimized path to obtain the final path.

[0112] It should be noted that the aforementioned construction module, weight adjustment module, repetition module, primary processing module, and secondary processing module correspond to steps S101 to S105 in Embodiment 1. The examples and application scenarios implemented by these modules and their corresponding steps are the same, but they are not limited to the content disclosed in Embodiment 1. It should also be noted that these modules, as part of the system, can be executed in a computer system, such as a set of computer-executable instructions.

[0113] The descriptions of each embodiment in the above embodiments have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0114] The proposed system can be implemented in other ways. For example, the system embodiments described above are merely illustrative, and the division of modules described above is only a logical functional division. In actual implementation, there may be other division methods. For example, multiple modules may be combined or integrated into another system, or some features may be ignored or not executed.

[0115] Example 3

[0116] This embodiment also provides an electronic device, including: one or more processors, one or more memories, and one or more computer programs; wherein, the processor is connected to the memory, and the one or more computer programs are stored in the memory. When the electronic device is running, the processor executes the one or more computer programs stored in the memory to cause the electronic device to perform the method described in Embodiment 1.

[0117] It should be understood that in this embodiment, the processor can be a central processing unit (CPU), or it can be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor, etc.

[0118] Memory may include read-only memory and random access memory, and provides instructions and data to the processor. A portion of memory may also include non-volatile random access memory. For example, memory may also store information about the device type.

[0119] In the implementation process, each step of the above method can be completed by the integrated logic circuits in the processor hardware or by software instructions.

[0120] The method in Embodiment 1 can be directly implemented by a hardware processor, or implemented by a combination of hardware and software modules within the processor. The software modules can reside in readily available storage media in the art, such as random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, or registers. This storage medium is located in memory; the processor reads information from the memory and, in conjunction with its hardware, completes the steps of the above method. To avoid repetition, a detailed description is not provided here.

[0121] Those skilled in the art will recognize that the units and algorithm steps described in connection with the various examples of this embodiment 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 implementation should not be considered beyond the scope of this invention.

[0122] Example 4

[0123] This embodiment also provides a computer-readable storage medium for storing computer instructions, which, when executed by a processor, complete the method described in Embodiment 1.

[0124] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. An enhanced path planning method for firefighting robots based on SOM networks, characterized by: include: (1) Construct the initial SOM network; According to the competitive strategy mechanism, the winning neuron is selected in the initial SOM network; (2) Adjust the weights of the winning neurons and their neighboring neurons in the initial SOM network according to the weight adjustment mechanism that considers the direction vector; (3) Repeat (1)~(2) until the neuron weights no longer change, and obtain the initial path of each fire robot; (4) Perform the first post-processing operation on the initial path of each fire-fighting robot to obtain the first optimized path, including: According to the neuron arrangement order of each robot's initial path, starting from the starting neuron, check each neuron one by one to the last neuron, and check in turn whether there are obstacles between each pair of adjacent neurons; If there is an obstacle in the path connecting adjacent neurons A and B, then a new neuron is added to the corresponding neuron group. Its location is between neurons A and B, the new neuron The weights are initialized to the arithmetic mean of the weights of neurons A and B; Examine the new neurons at this time Whether it is inside an obstacle; if not inside an obstacle, retain the neuron. Continue checking each neuron from the starting neuron to the last neuron; If new neurons If the object is within an obstacle, draw the perpendicular bisector of path AB. The newly added neurons will then... Located on the foot of the perpendicular, with AB as the dividing line, and the neuron Starting from the initial position, move along the perpendicular line in the first and second directions away from the foot of the perpendicular. The distance moved each time is... Each move Afterwards, it is determined whether the moved position is within the obstacle. If the moved position is not within the obstacle, it indicates that the moved position is a neuron. The candidate locations are obtained, where candidate location C is obtained from the first direction and candidate location D is obtained from the second direction. The calculation is performed from the neuron... The number of moves required to reach the two candidate positions C and D from the initial position is used to select the candidate position C with fewer moves as the final position of the new neuron. The path ACB that passes through points A, C and B in sequence is used as the first optimization path. (5) Perform a second post-processing operation on the first optimized path to obtain the final path. The specific steps are as follows: Step (3-1): First, mark the starting point of the path and several designated location points. Assume that the path passes through points C, D, E, and F in sequence. Assume that points C and D are marked points, and points E and F are unmarked points. Marked points are manually set path points that must be passed, while unmarked points are candidate path points in the path. Step (3-2): Starting from point C, pass through the intermediate path point D and connect CE. Determine whether path point D is a marked point. If path point D is a marked point, it means that the path must pass through point D and cannot be eliminated. Therefore, CE is not connected, and CD and DE are retained as the path. Step (3-3): Starting from point D, pass through the intermediate path point E and connect DF. Since the intermediate path point E is not marked, determine whether the line segment DF passes through an obstacle. If the line segment DF passes through an obstacle, it means that DF cannot be used as a simplified path and point E cannot be eliminated. Step (3-4): Obtain the final path CDEF.

2. The enhanced path planning method for firefighting robots based on SOM networks as described in claim 1, characterized in that, Constructing the initial SOM network includes: Based on the number of firefighting robots and number of target points Construct and initialize the SOM network, setting the number of neurons to [value missing]. Groups, each group of neurons includes There are 10 neurons; the weights of each group of neurons are set to the initial position information of the corresponding robot.

3. The enhanced path planning method for firefighting robots based on SOM networks as described in claim 1, characterized in that, According to the competition strategy mechanism, the SOM network is initialized and the winning neuron is selected. The specific competition strategy adopts a greedy strategy, which selects the neuron with the closest distance to the target point as the winning neuron.

4. The enhanced path planning method for firefighting robots based on SOM networks as described in claim 1, characterized in that, The weights of the winning neurons and their neighboring neurons in the initial SOM network are adjusted according to the weight adjustment mechanism that considers the direction vector, including: Step (1-1): Hypothesize the neuron It has become the winning neuron through a greedy strategy; Steps (1-2): Using traditional neuron weight adjustment strategies, the neuron... The weights are updated to the position of the first location point, and it is checked whether the updated neuron is inside the obstacle. Steps (1-3): When neurons are detected When the robot is located within an obstacle, with the fire-fighting robot as the origin, the line connecting the fire-fighting robot and the first position point is rotated clockwise and counterclockwise respectively. The line is rotated twice clockwise to obtain the second position point, and twice counterclockwise to obtain the third position point; the line is used as the direction vector. Steps (1-4): Assume the second position point is already outside the obstacle, while the third position point is still inside the obstacle. Therefore, in order for the third position point to be outside the obstacle, such that... ,at this time Therefore, the second position point obtained by rotating clockwise is chosen as the final weight point, at which point the neuron... Update to the second position point; Steps (1-5): The second location point is obtained by rotating the direction vector; therefore, the parent neuron is copied at the second location point. Obtain sub-neurons And set the parent neuron No longer participating in the competition process; Steps (1-6): Sub-neurons The process involves participating in the competition and repeatedly executing steps (1-1), (1-2), (1-3), (1-4), and (1-5) until the weights of the neurons in the group no longer change, thus obtaining an initial path from the starting point to the ending point.

5. The enhanced path planning method for firefighting robots based on SOM networks as described in claim 1, characterized in that, The first optimized path is then subjected to a second post-processing operation to obtain the final path. The specific steps may also include: Step (4-1): First, mark the starting point of the path and the points that must be passed through to reach the target, indicating that these points are indispensable in the path. Mark point C, and points D, E and F are unmarked points. Step (4-2): Starting from point C, cross the intermediate path point D and connect CE. Determine whether CE passes through an obstacle and whether the intermediate path point D is a marked point. In this case, both are negative, indicating that the path does not need to pass through point D. Remove point D and connect CE. Delete path CDE and keep path CE. Step (4-3): Still starting from point C, cross the intermediate path point E and connect CF. Since CF passes through an obstacle, point E cannot be eliminated. Therefore, CF is not connected, and CE and EF are retained as the path. Step (4-4): Obtain the final path CEF.

6. An enhanced path planning system for firefighting robots based on SOM networks, characterized in that: include: The building module is configured to: build an initial SOM network; and select the winning neuron from the initial SOM network according to a competition strategy mechanism. The weight adjustment module is configured to adjust the weights of the winning neurons and their neighboring neurons in the initialized SOM network according to a weight adjustment mechanism that considers the direction vector. The repeating module is configured to repeatedly execute the building module and the weight adjustment module until the neuron weights no longer change, thus obtaining the initial path for each fire-fighting robot. The primary processing module is configured to perform a first post-processing operation on the initial path of each firefighting robot to obtain a first optimized path, including: According to the neuron arrangement order of each robot's initial path, starting from the starting neuron, check each neuron one by one to the last neuron, and check in turn whether there are obstacles between each pair of adjacent neurons; If there is an obstacle in the path connecting adjacent neurons A and B, then a new neuron is added to the corresponding neuron group. Its location is between neurons A and B, the new neuron The weights are initialized to the arithmetic mean of the weights of neurons A and B; Examine the new neurons at this time Whether it is inside an obstacle; if not inside an obstacle, retain the neuron. Continue checking each neuron from the starting neuron to the last neuron; If new neurons If the object is within an obstacle, draw the perpendicular bisector of path AB. The newly added neurons will then... Located on the foot of the perpendicular, with AB as the dividing line, and the neuron Starting from the initial position, move along the perpendicular line in the first and second directions away from the foot of the perpendicular. The distance moved each time is... Each move Afterwards, it is determined whether the moved position is within the obstacle. If the moved position is not within the obstacle, it indicates that the moved position is a neuron. The candidate locations are obtained, where candidate location C is obtained from the first direction and candidate location D is obtained from the second direction. The calculation is performed from the neuron... The number of moves required to reach the two candidate positions C and D from the initial position is used to select the candidate position C with fewer moves as the final position of the new neuron. The path ACB that passes through points A, C and B in sequence is used as the first optimization path. The secondary processing module is configured to perform a second post-processing operation on the first optimized path to obtain the final path. The specific steps are as follows: Step (3-1): First, mark the starting point of the path and several designated location points. Assume that the path passes through points C, D, E, and F in sequence. Assume that points C and D are marked points, and points E and F are unmarked points. Marked points are manually set path points that must be passed, while unmarked points are candidate path points in the path. Step (3-2): Starting from point C, pass through the intermediate path point D and connect CE. Determine whether path point D is a marked point. If path point D is a marked point, it means that the path must pass through point D and cannot be eliminated. Therefore, CE is not connected, and CD and DE are retained as the path. Step (3-3): Starting from point D, pass through the intermediate path point E and connect DF. Since the intermediate path point E is not marked, determine whether the line segment DF passes through an obstacle. If the line segment DF passes through an obstacle, it means that DF cannot be used as a simplified path and point E cannot be eliminated. Step (3-4): Obtain the final path CDEF.

7. An electronic device, characterized in that it comprises: Memory is used to store computer-readable instructions in a non-transitory manner. as well as Processor, for executing the computer-readable instructions, When the computer-readable instructions are executed by the processor, they perform the method described in any one of claims 1-5.

8. A storage medium characterized by being non-transitory. The system stores computer-readable instructions, wherein, when the non-transitory computer-readable instructions are executed by a computer, the instructions of the method according to any one of claims 1-5 are executed.

Images