Loader path planning method based on adaptive ant colony algorithm
Through the adaptive ant colony algorithm and Transformer point selection strategy model, the loader path is dynamically optimized, which solves the problem of insufficient real-time and adaptability of traditional methods in complex environments, and realizes efficient and safe loader operation in scrap steel scenarios.
Patent Information
- Application Number
- CN202411340356.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-25
- Publication Date
- 2025-01-03
AI Technical Summary
Traditional path planning methods are difficult to achieve real-time, efficient and safe loader path planning in complex and changeable scrap steel treatment scenarios, especially when facing dynamic obstacles and environmental changes, resulting in low loader operation efficiency and high safety risks.
Using the path planning method based on the adaptive ant colony algorithm, the pheromone concentration and function parameters are initialized, combined with the Transformer point selection strategy model, dynamically update the pheromone concentration and path quality, iterative optimization finds the best path, ensuring that the loader is in scrap steel scene efficient homework.
It effectively improves the operating efficiency of the loader in scrap steel scenarios, avoids obstacles, ensures the safety of the loader, other equipment and personnel, realizes refined safety management and production process control, and improves production efficiency and operation control efficiency.
Smart Images
Figure CN119245676A_ABST
Abstract
Description
Technical Field
[0001] The invention belongs to the technical field of path planning and relates to a loader path planning method based on an adaptive ant colony algorithm. Background Art
[0002] In the scrap steel processing industry, loaders are core operating equipment, and their operating efficiency and safety are directly related to the smoothness and cost-effectiveness of the entire scrap steel processing process. Due to its particularity, the scrap steel processing scene presents the characteristics of complex and changeable areas, numerous obstacles and irregular distribution. These scrap steel materials not only have different shapes and weights, but also the stacking areas often change frequently due to factors such as site restrictions and material flow, which brings great challenges to the operation of loaders.
[0003] In the scrap steel processing scenario, path planning is the key to achieving efficient and safe operation of the loader. Traditional path planning methods include visualization method, artificial potential field method, simulated annealing method, etc. These methods can provide feasible driving paths for the loader to a certain extent. However, these methods often seem to be unable to cope with the complex and changeable scrap steel site with numerous obstacles.
[0004] Traditional path planning methods often fail to achieve the desired results in such a complex and ever-changing environment. They are either too computationally complex, resulting in poor real-time performance and unable to meet actual operational needs; or they are not adaptable enough to environmental changes and are unable to effectively deal with dynamic obstacles that frequently appear in scrap steel sites, such as moving vehicles, personnel, and the ever-changing shape of scrap steel piles. The existence of these problems not only affects the efficiency of the loader's stacking, but also increases the risk of collision during the operation, posing a serious threat to operational safety. Summary of the invention
[0005] In view of this, an object of the present invention is to provide a loader path planning method based on an adaptive ant colony algorithm.
[0006] In order to achieve the above object, the present invention provides the following technical solutions:
[0007] A loader path planning method based on an adaptive ant colony algorithm comprises the following steps:
[0008] S1: Initialize pheromone concentration and function parameters;
[0009] S2: Determine whether the number of points reached by the current ant exceeds the existing number of points. If the number of points reached by the ant does not exceed the existing number of points, use the Transformer-based point selection strategy model to determine the next path to be selected; if the ant reaches or exceeds the existing number of points, execute step S3;
[0010] S3: Calculate the quality of the points;
[0011] S4: Update the pheromone concentration on the path according to the quality of the ant's path, increase the pheromone on the high-quality path; record the best path currently found;
[0012] S5: Iterate steps S2-S4 until all ants are processed and the best path found is output as the final optimization result.
[0013] Furthermore, in step S1, the total number of ants is defined as K, k is the ant number, N represents the number of planned points, n represents the sequence number of the current point, and the pheromone represented by each path is expressed as:
[0014]
[0015] Where τ represents pheromone, 1-ρ represents the pheromone residual coefficient, Δτ ij (t) represents the accumulated amount of pheromone, and records the pheromone left by all ants from point i to point j to update the pheromone concentration. The pheromone concentration increased by the k-th ant releasing pheromone on the path connecting point i to point j:
[0016]
[0017] Where Q is the pheromone constant, L k is the total length of the path traversed by ant k.
[0018] Furthermore, the construction process of the Transformer-based point selection strategy model is as follows:
[0019] 1) Data preprocessing: The point selection problem is abstracted into a series of discrete points (tokens), each of which has a corresponding feature vector;
[0020] 2) Position coding: Position coding is introduced to maintain the order information of each point; the position coding is given by sine and cosine functions of different frequencies. For a sequence of length I, the coding of the kth position is defined by the following formula:
[0021]
[0022] Among them, pos is the position of the point, d model is the dimension of the model;
[0023] 3) Self-attention mechanism: Use the self-attention mechanism to capture the relationship between points. The calculation of self-attention is expressed as:
[0024]
[0025] Among them, Q, K, and V are query, key, and value matrices respectively. k is the dimension of the key;
[0026] 4) Multi-head attention mechanism: Introduce a multi-head attention mechanism to consider the relationship between points from multiple perspectives at the same time;
[0027] 5) Value function: Define a value function V(point) to evaluate the suitability of each point as a selection point; the value function is based on the characteristics of the point, its relationship with other points, and problem-specific constraints;
[0028] 6) Loss function: Use cross entropy loss to measure the difference between the predicted probability distribution and the actual distribution:
[0029]
[0030] Where C is the number of categories, y is the onehot encoding of the true label, and p is the probability distribution predicted by the model;
[0031] 7) Optimization algorithm: Use the optimization algorithm to update the model parameters and minimize the loss function through the back propagation algorithm;
[0032] 8) Model evaluation: Evaluate the performance of the model on the test set, using the accuracy and recall rate indicators to measure the accuracy of the model selection points;
[0033] 9) Iterative optimization: Through iterative training and evaluation, the model structure and parameters are continuously optimized until satisfactory performance is achieved.
[0034] On the other hand, the present invention provides an electronic device comprising a memory and a processor; the memory is used to store a computer program; the processor is used to implement the loader path planning method based on the adaptive ant colony algorithm as described in any of the above items when executing the computer program.
[0035] On the other hand, the present invention provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the loader path planning method based on the adaptive ant colony algorithm as described in any one of the above items is implemented.
[0036] The beneficial effects of the present invention are as follows: the present invention can effectively allow the loader to select the optimal path when performing the stacking operation, avoid obstacles in the path that may affect the stacking operation, greatly improve the efficiency of the loader in stacking in the scrap steel scene, and also ensure the equipment safety of the loader itself, thereby ensuring the safety of other equipment and personnel in the scrap steel yard. The present invention realizes the refined safety management of engineering vehicles and production process control in the scrap steel yard, further improves production efficiency, maintains the safety of production order, and at the same time realizes the improvement of the loader's operating efficiency. The present invention transparently monitors the loader to improve the efficiency of ground transportation control.
[0037] Other advantages, objectives and features of the present invention will be described in the following description to some extent, and to some extent, will be obvious to those skilled in the art based on the following examination and study, or can be taught from the practice of the present invention. The objectives and other advantages of the present invention can be realized and obtained through the following description. BRIEF DESCRIPTION OF THE DRAWINGS
[0038] In order to make the purpose, technical solutions and advantages of the present invention more clear, the present invention will be described in detail below in conjunction with the accompanying drawings, wherein:
[0039] Figure 1 This is a flow chart of the loader path planning method based on adaptive ant colony algorithm. DETAILED DESCRIPTION
[0040] The following describes the embodiments of the present invention by specific examples, and those skilled in the art can easily understand other advantages and effects of the present invention from the contents disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that the illustrations provided in the following embodiments only illustrate the basic concept of the present invention in a schematic manner, and the following embodiments and features in the embodiments can be combined with each other without conflict.
[0041] It should be noted that the illustrations provided in the following embodiments are only schematic illustrations of the basic concept of the present invention, and thus the drawings only show components related to the present invention rather than being drawn according to the number, shape and size of components in actual implementation. In actual implementation, the type, quantity and proportion of each component may be changed arbitrarily, and the component layout may also be more complicated.
[0042] In the following description, numerous details are discussed to provide a more thorough explanation of the embodiments of the present invention. However, it is obvious to those skilled in the art that the embodiments of the present invention can be implemented without these specific details. In other embodiments, well-known structures and devices are shown in the form of block diagrams rather than in detail to avoid making the embodiments of the present invention difficult to understand.
[0043] like Figure 1 As shown, the present invention provides a loader path planning method based on an adaptive ant colony algorithm, by installing a path planning algorithm on an industrial computer of the loader, using a position sensor on the loader to collect position data, and performing path planning for the loader.
[0044] First, define the total number of ants as K, where k is the ant number, N represents the number of planned points, and n represents the sequence number of the current point.
[0045]
[0046] Formula (1) shows the pheromone represented by each path, where τ represents the pheromone, 1-ρ represents the pheromone residual coefficient, and Δτ ij (t) represents the accumulated amount of pheromone, and records the pheromone left by all ants from point i to point j to update the pheromone concentration. The pheromone concentration increases when the k-th ant releases pheromone on the path connecting point i to point j.
[0047]
[0048] Where Q is the pheromone constant, L k is the total length of the path traversed by ant k.
[0049] The first step of the path planning algorithm is initialization, including pheromone concentration and function parameters. Pheromone concentration refers to the chemical substances released by ants on the path to guide other ants to choose the path. Function parameters include various settings that affect the behavior of the algorithm, such as the path selection method and the rules of the calculation process.
[0050] The second step is to determine whether the number of points reached by the current ant exceeds the number of existing points. This step is to control the movement of the ants to ensure that they do not go back or go to points that have already been explored. If the number of points reached by the ant does not exceed the number of existing points, the algorithm will use the Transformer point selection strategy to determine the next path to choose.
[0051] When the ant reaches or exceeds the point where there are already points, it enters the third step, which is to calculate the quality of the points. The quality of the points is usually determined by some heuristic methods, such as evaluating the quality of the path based on factors such as path length and pheromone concentration of the path. This step is to prepare for updating the pheromone concentration and recording the best path.
[0052] The fourth step is to update the pheromone concentration and record the best path. According to the quality of the ant's path, the pheromone concentration on the path is updated, the pheromone on the high-quality path is increased, and the pheromone on the low-quality path is reduced. At the same time, the best path currently found is recorded for subsequent comparison and output.
[0053] If the number of ants k currently processed is less than the total number of ants K, then return to the first step and repeat the second, third, and fourth steps until the number of ants k processed is greater than or equal to the total number of ants K. At this time, the algorithm outputs the best path that has been found as the final optimization result.
[0054] The design and implementation of the entire algorithm aims to simulate the behavior of ant colonies in finding the optimal path. Through the dynamic adjustment of pheromone concentration and the quality evaluation of the path, the entire process can gradually converge to a better solution. This ant colony algorithm has great application potential in solving combinatorial optimization problems and can effectively find better solutions to complex problems.
[0055] The following is a detailed introduction to the Transformer point selection strategy model building process:
[0056] 1) Data preprocessing: The point selection problem is abstracted into a series of discrete points (tokens). Each point has a corresponding feature vector. These features may include information such as the coordinates of the point and the distance to other points.
[0057] 2) Position coding: Position coding is introduced to maintain the order information of each point. The position coding is given by sine and cosine functions of different frequencies. For a sequence of length I, the coding of the kth position is defined by the following formula:
[0058]
[0059] Among them, pos is the position of the point, d model is the dimension of the model
[0060] 3) Self-attention mechanism: Use the self-attention mechanism to capture the relationship between points. The calculation of self-attention can be expressed as:
[0061]
[0062] Among them, Q, K, and V are query, key, and value matrices respectively.k is the dimension of the key.
[0063] 4) Multi-head attention: Through the multi-head attention mechanism, the model can consider the relationship between points from multiple perspectives at the same time, which helps to find the global optimal solution.
[0064] 5) Value function: Define a value function V(point) to evaluate the suitability of each point as a selection point. This value function can be based on the characteristics of the point, its relationship with other points, and problem-specific constraints.
[0065] 6) Loss function: Design a loss function to train the Transformer model so that it can predict the most likely point. The loss function can be cross entropy loss, which is used to measure the difference between the predicted probability distribution and the actual distribution:
[0066]
[0067] Where C is the number of categories, y is the onehot encoding of the true label, and p is the probability distribution predicted by the model
[0068] 7) Optimization algorithm: Use an optimization algorithm (such as Adam) to update the model parameters and minimize the loss function through the back propagation algorithm.
[0069] 8) Model evaluation: Evaluate the performance of the model on the test set. Indicators such as accuracy and recall can be used to measure the accuracy of the model selection points.
[0070] 9) Iterative optimization: Through iterative training and evaluation, the model structure and parameters are continuously optimized until satisfactory performance is achieved.
[0071] The loader adopts autonomous global planning, which is more reliable in long-term and high-intensity operation than manual operation. The loader's orientation and coordinates are obtained through position sensors to obtain the loader's status. It has higher reliability and accuracy.
[0072] In the above embodiments, the description's reference to "this embodiment" indicates that a particular feature, structure, or characteristic described in conjunction with the embodiment is included in at least some embodiments, but not necessarily all embodiments. Multiple occurrences of "this embodiment" do not necessarily all refer to the same embodiment.
[0073] In the above-described embodiments, although the invention has been described in conjunction with specific embodiments of the invention, many substitutions, modifications, and variations of these embodiments will be apparent to those of ordinary skill in the art based on the foregoing description. For example, other storage structures (e.g., dynamic RAM (DRAM)) may use the embodiments discussed. Embodiments of the invention are intended to encompass all such substitutions, modifications, and variations that fall within the broad scope of the appended claims.
[0074] This embodiment further provides a computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, any one of the methods in this embodiment is implemented.
[0075] This embodiment also provides an electronic terminal, including: a processor and a memory;
[0076] The memory is used to store a computer program, and the processor is used to execute the computer program stored in the memory, so that the terminal executes any one of the methods in this embodiment.
[0077] The computer-readable storage medium in this embodiment can be understood by ordinary technicians in this field: all or part of the steps of implementing the above-mentioned method embodiments can be completed by hardware related to the computer program. The aforementioned computer program can be stored in a computer-readable storage medium. When the program is executed, the steps of the above-mentioned method embodiments are executed; and the aforementioned storage medium includes: ROM, RAM, magnetic disk or optical disk and other media that can store program codes.
[0078] The electronic terminal provided in this embodiment includes a processor, a memory, a transceiver and a communication interface. The memory and the communication interface are connected to the processor and the transceiver and complete communication with each other. The memory is used to store computer programs, the communication interface is used to communicate, and the processor and the transceiver are used to run computer programs so that the electronic terminal executes each step of the above method.
[0079] In this embodiment, the memory may include a random access memory (RAM), and may also include a non-volatile memory (non-volatile memory), such as at least one disk memory.
[0080] The above-mentioned processor can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.
[0081] The present invention can be used in many general or special computing system environments or configurations, such as personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronic devices, network PCs, minicomputers, mainframe computers, distributed computing environments including any of the above systems or devices, and the like.
[0082] The present invention may be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. The present invention may also be practiced in distributed computing environments where tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules may be located in local and remote computer storage media, including storage devices.
[0083] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solution of the present invention can be modified or replaced by equivalents without departing from the purpose and scope of the technical solution, which should be included in the scope of the claims of the present invention.
Claims
1. A loader path planning method based on adaptive ant colony algorithm, characterized in that: The following steps are involved: S1: Initialize pheromone concentration and function parameters; S2: Determine whether the number of points reached by the current ant exceeds the existing number of points. If the number of points reached by the ant does not exceed the existing number of points, use the Transformer-based point selection strategy model to determine the next path to be selected; if the ant reaches or exceeds the existing number of points, execute step S3; S3: Calculate the quality of the points; S4: Update the pheromone concentration on the path according to the quality of the ant's path, increase the pheromone on the high-quality path; record the best path currently found; S5: Iterate steps S2-S4 until all ants are processed and the best path found is output as the final optimization result.
2. The loader path planning method based on the adaptive ant colony algorithm according to claim 1 is characterized in that: In step S1, the total number of ants is defined as K, k is the ant number, N represents the number of planned points, n represents the sequence number of the current point, and the pheromone represented by each path is expressed as: Where τ represents pheromone, 1-ρ represents the pheromone residual coefficient, Δτ ij (t) represents the accumulated amount of pheromone, and records the pheromone left by all ants from point i to point j to update the pheromone concentration. The pheromone concentration increased by the k-th ant releasing pheromone on the path connecting point i to point j: Where Q is the pheromone constant, L k is the total length of the path traversed by ant k.
3. The loader path planning method based on the adaptive ant colony algorithm according to claim 1 is characterized in that: The construction process of the Transformer-based point selection strategy model is as follows: 1) Data preprocessing: The point selection problem is abstracted into a series of discrete points (tokens), each of which has a corresponding feature vector; 2) Position coding: Position coding is introduced to maintain the order information of each point; the position coding is given by sine and cosine functions of different frequencies. For a sequence of length I, the coding of the kth position is defined by the following formula: Among them, pos is the position of the point, d model is the dimension of the model; 3) Self-attention mechanism: Use the self-attention mechanism to capture the relationship between points. The calculation of self-attention is expressed as: Among them, Q, K, and V are query, key, and value matrices respectively. k is the dimension of the key; 4) Multi-head attention mechanism: Introduce a multi-head attention mechanism to consider the relationship between points from multiple perspectives at the same time; 5) Value function: Define a value function V(point) to evaluate the suitability of each point as a selection point; the value function is based on the characteristics of the point, its relationship with other points, and problem-specific constraints; 6) Loss function: Use cross entropy loss to measure the difference between the predicted probability distribution and the actual distribution: Where C is the number of categories, y is the onehot encoding of the true label, and p is the probability distribution predicted by the model; 7) Optimization algorithm: Use the optimization algorithm to update the model parameters and minimize the loss function through the back propagation algorithm; 8) Model evaluation: Evaluate the performance of the model on the test set, using the accuracy and recall rate indicators to measure the accuracy of the model selection points; 9) Iterative optimization: Through iterative training and evaluation, the model structure and parameters are continuously optimized until satisfactory performance is achieved.
4. An electronic device, characterized in that: including memory and processor; The memory is used to store computer programs; The processor is used to implement the loader path planning method based on the adaptive ant colony algorithm as described in any one of claims 1 to 3 when executing the computer program.
5. A computer-readable storage medium, characterized in that: The storage medium stores a computer program, and when the computer program is executed by the processor, the loader path planning method based on the adaptive ant colony algorithm as described in any one of claims 1 to 3 is implemented.