Path Planning Method for Cooperative Operation of Driverless Sanitation Vehicles

Abstract

The present invention discloses a path planning method for collaborative operation of unmanned sanitation vehicles, including: determining the operation areas of a number of unmanned sanitation vehicles; establishing a task road network based on the operation areas, using a clustering algorithm to group the task road network, classifying adjacent road sections into the same category to obtain a number of subtasks, assigning each subtask to an unmanned sanitation vehicle, and determining the optimal number of unmanned sanitation vehicles through multiple groupings; for each subtask after clustering, using the ant colony algorithm to determine the working path and non-working path of each unmanned sanitation vehicle; establishing a path planning model based on multiple factors; based on the path planning model, using an optimization algorithm for path search and optimization to find the optimal path under given constraints. The present invention can effectively arrange the paths of unmanned sanitation vehicles to improve transportation efficiency and reduce costs.

Description

Technical Field

[0001] The present invention belongs to the technical field of path planning for unmanned sanitation vehicles, and particularly relates to a path planning method for collaborative operation of unmanned sanitation vehicles. Background Art

[0002] With the vigorous development of autonomous driving technology, a large number of unmanned operation vehicles have emerged in the market, such as unmanned sweeping vehicles and unmanned sprinkler vehicles. These vehicles play an important role in improving efficiency, reducing costs, improving the working environment, and promoting environmental protection, bringing revolutionary changes to modern urban management and the transportation field. However, due to the complexity of the path problems of unmanned sanitation vehicles, solving the path planning problem of urban unmanned sanitation vehicles still poses challenges.

[0003] Vehicle path planning is an important task for modern transportation service providers and a research focus in both the industrial and academic fields. With the rapid progress of autonomous driving technology, the emergence of vehicles such as unmanned sweeping vehicles and unmanned sprinkler vehicles not only improves the operation efficiency but also makes positive contributions to urban maintenance and environmental protection. However, these unmanned operation vehicles need to be processed in different links during operation, such as charging, maintenance, garbage dumping, water adding, and operation. Each link requires traveling to specific operation sites, such as cleaning points, charging stations, parking points, dumping points, water supply replenishment points, maintenance points, and fueling points. Therefore, on the premise of ensuring that the vehicle completes the task, planning the operation path of the unmanned sanitation vehicle is of great significance for minimizing the operation cost and ensuring the efficient completion of the operation task.

[0004] However, the current path planning methods have obvious limitations in solving the path planning problem of unmanned sanitation vehicles with specific operation modes. Existing methods mainly rely on technologies such as operations research, deep learning, and reinforcement learning. Although significant results have been achieved in some cases, they still fail to fully solve the problems of the completion degree of specific link operations and overall path planning. In addition, existing methods do not fully consider the impact of multiple restrictive conditions such as operation time periods, seasons, environmental temperatures, weather conditions, and geographical locations on the operation cost. Therefore, in the operation of urban unmanned sanitation vehicles, how to conduct path planning while considering the dual balance of efficiency and cost has become an urgent problem to be deeply studied and solved. Summary of the Invention

[0005] To solve the above technical problems, the present invention proposes a path planning method for collaborative operation of unmanned sanitation vehicles.

[0006] To achieve the above object, the technical solution of the present invention is as follows:

[0007] In the first aspect, the present invention discloses a path planning method for collaborative operation of unmanned sanitation vehicles, including:

[0008] Step S1: Determine the operation areas of several unmanned sanitation vehicles;

[0009] Step S2: Based on the operation areas, establish a task road network, group the task road network using a clustering algorithm, classify adjacent road segments into the same category to obtain several subtasks, assign each subtask to an unmanned sanitation vehicle, and determine the optimal number of unmanned sanitation vehicles through multiple groupings;

[0010] Step S3: For each subtask after clustering, use the ant colony algorithm to determine the working path and non-working path of each unmanned sanitation vehicle;

[0011] The working path is the path followed by the unmanned sanitation vehicle when performing tasks;

[0012] The non-operation path is the path followed by the unmanned sanitation vehicle when going to the working path or the supply station;

[0013] Step S4: Based on one or more factors such as the cost of the unmanned sanitation vehicle, the geographical topology of the operation area, the spatio-temporal requirements of different operation tasks, the reasonable allocation of the operation time period, the influence of seasonal changes, the influence of environmental temperature, and the restriction of weather information on the operation cost of the unmanned sanitation vehicle, the current position of the unmanned sanitation vehicle, and the target position of the unmanned sanitation vehicle, establish a path planning model;

[0014] Step S5: Based on the path planning model, use an optimization algorithm for path search and optimization to find the optimal path under given constraints.

[0015] Based on the above technical solution, the following improvements can also be made:

[0016] As an optimal solution, Step S2 includes:

[0017] Step S2.1: Initialize the clustering number to an initial value;

[0018] Step S2.2: Use the clustering algorithm to divide the task roads in the task road network into clusters of the clustering number, and use the path planning algorithm to solve the path planning of each subtask to obtain the optimal path and path length of each cluster;

[0019] Step S2.3: Calculate the operation duration of all subtask roads and determine whether the termination condition is met;

[0020] If not, if the sum of the path lengths of the subtasks does not reach the expected minimum value, increase the clustering number, and repeat Steps S2.2 - S2.3 until the sum of the path lengths of the subtasks reaches the expected minimum value or cannot be reduced further, and determine the current clustering number as the optimal number of unmanned sanitation vehicles;

[0021] If so, the current clustering number is the optimal number of unmanned sanitation vehicles.

[0022] As a preferred solution, step S3 includes:

[0023] Step S3.1: Initialize the pheromone trails on all road segments to a constant, and randomly place ants on the starting node of the search space;

[0024] Step S3.2: Each ant selects the next road segment to move according to the pheromone concentration and heuristic information;

[0025] Judge whether the remaining resources can reach the specified road segment;

[0026] If it can, the ant moves, and updates the resource surplus, working path and non-working path of the driverless sanitation vehicle;

[0027] Otherwise, after selecting the optimal supply station for replenishment, the ant moves, and updates the resource surplus, working path and non-working path of the driverless sanitation vehicle;

[0028] Step S3.3: Repeat step S3.2 until all road segments are traversed;

[0029] Step S3.4: Calculate the duration of task completion;

[0030] Step S3.5: Judge whether the duration meets the operation duration limit;

[0031] If it meets, record the path information of all ants and update the pheromone;

[0032] If it does not meet, set the path cost to infinity, record the path information of all ants, and update the pheromone;

[0033] Step S3.6: Used to judge whether the maximum iteration number is reached or the convergence condition is met;

[0034] If so, return the path with the optimal cost;

[0035] Otherwise, repeat steps S3.2 - S3.6.

[0036] As a preferred solution, in step S4, the path planning model is as follows:

[0037]

[0038] Where:

[0039] F·Q represents the total cost required to use Q driverless sanitation vehicles, and F represents the fixed cost of each driverless sanitation vehicle;

[0040] represents the total energy consumption of the driverless sanitation vehicle on the working path and non-working path;

[0041] is the operation identifier of the k-th driverless sanitation vehicle, indicating that the k-th driverless sanitation vehicle is in the operation state at the t-step decision, indicating that the k-th driverless sanitation vehicle is in the non-operation state at the t-step decision, is the node and is the distance between and are the position points of the k-th driverless sanitation vehicle at the t-step decision and the (t + 1)-step decision respectively. w is the energy consumption cost per unit distance of the operation path, and r is the energy consumption cost per unit distance of the non-operation path;

[0042] represents the ratio of the length of the operation path to the length of the non-operation path of the driverless sanitation vehicle, and β is the adjustment coefficient.

[0043] As a preferred solution, the value of w is different in different operation periods, different temperatures, different weather information, different operation cities, and different operation areas.

[0044] As a preferred solution, in step S5, the limiting conditions include one or more of the following: resource constraints during the working process of the driverless sanitation vehicle, operation duration limit of the driverless sanitation vehicle, and energy constraint carried by the driverless sanitation vehicle.

[0045] In a second aspect, the present invention discloses a path planning device for collaborative operation of driverless sanitation vehicles, including:

[0046] An operation area determination module for determining the operation areas of a number of driverless sanitation vehicles;

[0047] A sanitation vehicle quantity determination module for establishing a task road network based on the operation areas, using a clustering algorithm to group the task road network, classifying adjacent road sections into the same category to obtain a number of subtasks, allocating each subtask to a driverless sanitation vehicle, and determining the optimal number of driverless sanitation vehicles through multiple groupings;

[0048] A path determination module for, for each clustered subtask, using an ant colony algorithm to determine the working path and non-working path of each driverless sanitation vehicle;

[0049] The working path is the path followed by the driverless sanitation vehicle when performing tasks;

[0050] The non-operation path is the path followed by the driverless sanitation vehicle when going to the working path or the supply station;

[0051] A path planning model establishment module, which is used to establish a path planning model based on one or more of the cost of the unmanned sanitation vehicle, the geographical topology of the operation area, the spatio-temporal requirements of different operation tasks, the reasonable allocation of the operation time period, the influence of seasonal changes, the influence of environmental temperature, the restriction of weather information on the operation cost of the unmanned sanitation vehicle, the current position of the unmanned sanitation vehicle, and the target position of the unmanned sanitation vehicle;

[0052] A path search module, which is used to perform path search and optimization based on the path planning model by using an optimization algorithm to find the optimal path under given constraints.

[0053] As an optimal solution, the sanitation vehicle quantity determination module includes:

[0054] A clustering initialization unit, which is used to initialize the number of clusters to an initial value;

[0055] A clustering unit, which is used to divide the task roads in the task road network into clusters with the number of clusters by using a clustering algorithm, and use a path planning algorithm to solve the path planning of each subtask to obtain the optimal path and path length of each cluster;

[0056] A clustering algorithm termination judgment unit, which is used to calculate the operation duration of all subtask roads and judge whether the termination condition is satisfied;

[0057] If not satisfied, if the sum of the path lengths of the subtasks does not reach the expected minimum value, increase the number of clusters, and repeat the methods in the clustering unit and the judgment unit until the sum of the path lengths of the subtasks reaches the expected minimum value or cannot be reduced any further, and determine the current number of clusters as the optimal number of unmanned sanitation vehicles;

[0058] If satisfied, the current number of clusters is the optimal number of unmanned sanitation vehicles.

[0059] As an optimal solution, the path determination module includes:

[0060] An ant colony initialization unit, which is used to initialize the pheromone trails on all road segments to a constant and randomly place ants on the starting node of the search space;

[0061] An ant colony movement unit, which is used for each ant to select the next road segment to move according to the pheromone concentration and heuristic information;

[0062] Judge whether the remaining resources can reach the specified road segment;

[0063] If it can, the ant moves, and the remaining resources, working path, and non-working path of the unmanned sanitation vehicle are updated;

[0064] Otherwise, after selecting the optimal supply station for replenishment, the ant moves, and the remaining resources, working path, and non-working path of the unmanned sanitation vehicle are updated;

[0065] A repeated execution unit for repeatedly executing the method in the ant colony movement unit until all road segments are traversed;

[0066] A duration calculation unit for calculating the duration of task completion;

[0067] A duration judgment unit for judging whether the duration meets the job duration limit;

[0068] If it meets the requirement, record the path information of all ants and update the pheromone;

[0069] If it does not meet the requirement, set the path cost to infinity, record the path information of all ants, and update the pheromone;

[0070] An ant colony algorithm termination judgment unit for judging whether the maximum iteration number is reached or the convergence condition is met;

[0071] If so, return the optimal cost path;

[0072] Otherwise, sequentially repeat the methods in the ant colony movement unit, the repeated execution unit, the duration calculation unit, the duration judgment unit, and the ant colony algorithm termination judgment unit.

[0073] As an optimal solution, in the path planning model establishment module, the path planning model is as follows:

[0074]

[0075] Where:

[0076] F·Q represents the total cost required to use Q unmanned sanitation vehicles, and F represents the fixed cost of each unmanned sanitation vehicle;

[0077] represents the total energy consumption of the unmanned sanitation vehicle on the working path and the non-working path;

[0078] is the operation identifier of the kth unmanned sanitation vehicle, represents that the kth unmanned sanitation vehicle is in the operation state at the t-step decision, represents that the kth unmanned sanitation vehicle is in the non-operation state at the t-step decision, is the node and the distance between, and are the position points of the kth unmanned sanitation vehicle at the t-step decision and the t+1-step decision respectively, w is the unit distance energy consumption cost of the operation path, and r is the unit distance energy consumption cost of the non-operation path;

[0079] It represents the ratio of the operation path length to the non - operation path length of the unmanned sanitation vehicle, and β is the adjustment coefficient.

[0080] As a preferred solution, the value of w is different in different operation time periods, different temperatures, different weather information, different operation cities, and different operation areas.

[0081] As a preferred solution, in the path search module, the limiting conditions include one or more of the following: resource constraints during the working process of the unmanned sanitation vehicle, operation duration limit of the unmanned sanitation vehicle, and energy constraint carried by the unmanned sanitation vehicle.

[0082] In addition, thirdly, the present invention discloses a storage medium storing one or more computer - readable programs. The one or more programs include instructions adapted to be loaded and executed by the memory to perform the path planning method for collaborative operation of any of the above - mentioned unmanned sanitation vehicles.

[0083] The present invention discloses a path planning method for collaborative operation of unmanned sanitation vehicles, which has the following beneficial effects:

[0084] First, the present invention designs a comprehensive path planning model to achieve optimal efficiency and minimum cost on the premise of considering multiple limiting conditions such as operation time period, season, environmental temperature, weather conditions, geographical location, etc., aiming at diverse types of unmanned sanitation vehicles to meet their respective specific operation areas and task requirements.

[0085] Second, the path planning model of the present invention covers the operation path and non - operation path of the unmanned sanitation vehicle to ensure that the unmanned sanitation vehicle can give full play to its benefits in practical applications.

[0086] Third, the present invention can effectively arrange the paths of unmanned sanitation vehicles to improve transportation efficiency and reduce costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0087] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the drawings required for the embodiments. It should be understood that the following drawings only show some embodiments of the present invention, so they should not be regarded as limiting the scope. For those of ordinary skill in the art, other relevant drawings can be obtained based on these drawings without creative efforts.

[0088] Figure 1 It is a flowchart of the path planning method for collaborative operation of unmanned sanitation vehicles provided by the embodiment of the present invention.

[0089] Figure 2 It is a flowchart of the ant colony algorithm provided by the embodiment of the present invention.

[0090] Figure 3Experimental graphs of the ten-fold values of the proportion of the total non-working path and the number of driverless vehicles used when three different algorithms (K-means_ant, task_greedy, and Naive_greedy) provided in the embodiments of the present invention complete their operations.

[0091] Figure 4 (a), (b), and (c) are graphs showing the proportion of the non-working path per 1 km traveled by each driverless vehicle under different actual sanitation operation scenarios (Scenario1, Scenario 2, Scenario 3) provided in the embodiments of the present invention. Detailed implementation manners

[0092] The preferred embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0093] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0094] In addition, the expression of "including" elements is an "open" expression, which only means that there are corresponding components or steps, and should not be construed as excluding additional components or steps.

[0095] In some embodiments of the path planning method for collaborative operation of driverless sanitation vehicles in order to achieve the purpose of the present invention,

[0096] The path planning method for collaborative operation of driverless sanitation vehicles includes:

[0097] Step S1: Determine the operation areas of a number of driverless sanitation vehicles;

[0098] Step S2: Based on the operation areas, establish a task road network, use a clustering algorithm to group the task road network, classify adjacent road sections into the same category, obtain a number of subtasks, assign each subtask to a driverless vehicle, and determine the optimal number of driverless vehicles through multiple groupings;

[0099] Step S3: For each clustered subtask, use the ant colony algorithm to determine the working path and non-working path of each driverless vehicle;

[0100] The working path is the path followed by the driverless vehicle when performing tasks;

[0101] The non-operation path is the path followed by the driverless vehicle when going to the working path or the supply station;

[0102] Step S4: Based on one or more of the cost of the unmanned sanitation vehicle, the geographical topology of the operation area, the spatio-temporal requirements of different operation tasks, the reasonable allocation of the operation time period, the influence of seasonal changes, the influence of environmental temperature, and the restriction of weather information on the operation cost of the unmanned sanitation vehicle, the current position of the unmanned sanitation vehicle, and the target position of the unmanned sanitation vehicle, establish a path planning model;

[0103] Step S5: Based on the path planning model, use an optimization algorithm to perform path search and optimization, and find the optimal path under the given constraints.

[0104] The following elaborates on each step in detail.

[0105] Furthermore, Step S2 uses the K-means algorithm for clustering. The K-means clustering algorithm is used to group the task road network, and similar road segments are grouped into the same category. This can reduce the complexity and computational amount of path planning, making path planning more efficient. Each subtask is assigned to an unmanned operation vehicle, and the optimal number of unmanned sanitation vehicles is finally determined through multiple groupings.

[0106] K-means is a commonly used clustering algorithm for dividing samples in a dataset into K different groups or clusters. Its basic principle is to find the center points of the clusters through iterative optimization, minimizing the sum of the distances from the sample points to the center points of their respective clusters. The following are the basic steps of the K-means algorithm:

[0107] Basic steps of the K-means algorithm:

[0108] 1. Initialization: Randomly select K samples as the initial clustering centers (centroids).

[0109] 2. Assign sample points: Assign each sample point to the cluster where the nearest clustering center is located.

[0110] 3. Update the clustering centers: Calculate the mean of all sample points in each cluster as the new clustering center.

[0111] 4. Repeat Steps 2 and 3 until the clustering centers no longer change or reach a predetermined number of iterations.

[0112] 5. Convergence: When the clustering centers no longer change, the algorithm converges and the final clustering result is obtained.

[0113] The advantages of K-means include simplicity and high efficiency, and it is suitable for large datasets. The K-means algorithm is used to cluster the task roads in the road network matrix so as to decompose the problem into multiple sub-tasks, with each sub-task corresponding to a cluster. The purpose of doing this is to use path planning algorithms to solve more complex path planning problems. In the given code, by continuously increasing the number of clusters (i.e., the number of unmanned sanitation vehicles), and then using path planning algorithms (such as the ant colony algorithm) to solve the path planning of each sub-problem (i.e., each cluster) until the number of clusters that meets the constraint conditions is found.

[0114] Specifically, step S2 includes:

[0115] Step S2.1: Initialize the number of clusters to an initial value, such as 2.

[0116] Step S2.2: Use the clustering algorithm to divide the task roads in the task road network into clusters of the number of clusters, and use the path planning algorithm to solve the path planning of each sub-task, obtaining the optimal path and path length of each cluster.

[0117] Step S2.3: Calculate the operation duration of all sub-task roads and determine whether the termination condition is met.

[0118] If not, if the sum of the path lengths of the sub-tasks does not reach the expected minimum value, increase the number of clusters and repeat steps S2.2 - S2.3 until the sum of the path lengths of the sub-tasks reaches the expected minimum value or cannot be reduced any further, and then determine the current number of clusters as the optimal number of unmanned sanitation vehicles.

[0119] If so, the current number of clusters is the optimal number of unmanned sanitation vehicles.

[0120] Through the above steps, the number of clusters can be continuously optimized, and finally the optimal number of unmanned sanitation vehicles can be determined to minimize the overall path length.

[0121] Furthermore, for each sub-task after clustering, the ant colony algorithm is used to determine the working path and non-working path of each unmanned sanitation vehicle. The present invention adopts a determination method based on the ant colony algorithm, ensuring the comparability and accuracy of the path planning scheme. The ant colony optimization (ACO) algorithm is inspired by the foraging behavior of ants and is a powerful meta-heuristic algorithm widely used to solve optimization problems, especially path planning problems.

[0122] Path planning is a fundamental problem in various fields, including robotics, transportation, and logistics. Its goal is to find the optimal path from a starting point to a target point while considering constraints such as obstacles, terrain conditions, and resource limitations. Traditional path planning methods often face challenges when dealing with complex environments and high-dimensional spaces. To overcome these challenges, researchers have turned to nature-inspired algorithms, among which the Ant Colony Optimization (ACO) algorithm has emerged as a promising approach.

[0123] The ACO algorithm is inspired by the foraging behavior of ants. Ants communicate with each other through pheromone trails to find the shortest path between their nests and food sources. The algorithm mimics this behavior by simulating a group of virtual ants moving in the search space and releasing pheromones on the explored paths. The key principles of the ACO algorithm include: positive feedback mechanism, heuristic factor, and pheromone evaporation.

[0124] Specifically, as Figure 2 shown, step S3 includes:

[0125] Step S3.1: Initialize the pheromone trails on all road segments to a constant and randomly place m ants on the starting node of the search space;

[0126] Step S3.2: Each ant selects the next road segment to move based on the pheromone concentration and heuristic information;

[0127] Judge whether the remaining resources can reach the specified road segment;

[0128] If it can, the ant moves, updating the resource balance, working path, and non-working path of the unmanned sanitation vehicle;

[0129] Otherwise, after selecting the optimal supply station for resupply, the ant moves, updating the resource balance, working path, and non-working path of the unmanned sanitation vehicle;

[0130] Step S3.3: Repeat step S3.2 until all road segments are traversed;

[0131] Step S3.4: Calculate the duration to complete the task;

[0132] Step S3.5: Judge whether the duration meets the operation duration limit;

[0133] If it meets, record the path information of all ants and update the pheromone;

[0134] If it does not meet, set the path cost to infinity, record the path information of all ants, and update the pheromone;

[0135] Step S3.6: Used to judge whether the maximum iteration number is reached or the convergence condition is met;

[0136] If so, return the cost-optimal path;

[0137] Otherwise, repeat steps S3.2 - S3.6.

[0138] Furthermore, step S4 mainly involves the modeling problem.

[0139] The present invention comprehensively considers the following factors: the cost of the unmanned sanitation vehicle, the geographical topology of the operation area, the spatio-temporal requirements of different operation tasks, the reasonable allocation of operation time periods, the influence of seasonal changes, the influence of environmental temperature, the restriction of weather information on the operation cost of the unmanned sanitation vehicle, and the current position and the upcoming target position of the unmanned sanitation vehicle, etc.

[0140] On the premise of ensuring the continuous working state of the vehicle to the greatest extent, the working path will minimize the driving distance and operation cost and ensure the efficient completion of the task; the non-working path will optimize the selection and order of supply points to minimize the driving distance and cost of the unmanned sanitation vehicle.

[0141] Given the following requirements: 1. When any unmanned sanitation vehicle is working, the resources on the vehicle can support its consumption at the operation location; 2. The operation duration limit; 3. The energy carried by any unmanned sanitation vehicle can support its operation and refueling; 4. Road full coverage to ensure the completion of the operation task; 5. The goal is to minimize the total cost, usually including the fixed cost of the vehicle and the path cost. In the path planning model, the goal generates a series of

[0142] The path planning model is as follows:

[0143]

[0144] Each decision of the algorithm will generate the next target point of the k-th unmanned sanitation vehicle Through T k -1 decisions, a complete path with optimal efficiency and minimum cost can be generated for the k-th unmanned sanitation vehicle, that is

[0145] Where:

[0146] F·Q represents the total cost required to use Q unmanned sanitation vehicles, and F represents the fixed cost of each unmanned sanitation vehicle; this item calculates the cost-optimal number of unmanned sanitation vehicles by controlling the fixed cost of the unmanned sanitation vehicle;

[0147] represents the total energy consumption of the unmanned sanitation vehicle on the working path and the non-working path;

[0148] is the kth unmanned sanitation vehicle operation identifier, It means that the kth unmanned sanitation vehicle is in the operating state after the t-step decision. It means that the kth unmanned sanitation vehicle is in the non-operating state after the t-step decision. For Node and The distance between and are the positions of the kth unmanned sanitation vehicle at the t-step decision and the t+1-step decision, w is the energy consumption cost per unit distance of the operation path (w has different values in different operation periods, different temperatures, different weather information, different operation cities and different operation areas), and r is the energy consumption cost per unit distance of the non-operation path;

[0149] It represents the ratio of the unmanned sanitation vehicle’s operating path length to the non-operating path length, and β is the adjustment coefficient. Ensures maximum efficiency of unmanned sanitation vehicles.

[0150] is the kth unmanned sanitation vehicle on the operation path and This item is calculated by comprehensively considering the total resource consumption of the unmanned sanitation vehicle in the operating and non-operating paths in different operating periods, different temperatures, different weather conditions, different operating cities, and different operating areas, so as to ensure that the total resource consumption of the overall path is minimized.

[0151] Further, in step S5, the restriction conditions include one or more of the following: resource constraints of the unmanned sanitation vehicle's working process, operating time limits of the unmanned sanitation vehicle, and energy constraints carried by the unmanned sanitation vehicle.

[0152] Specifically, the constraints can be expressed as:

[0153]

[0154] In order to meet the actual operation conditions of unmanned sanitation vehicles, including:

[0155] C k is the current resource value of the kth unmanned sanitation vehicle (such as water, etc.);

[0156] c is the unit distance attribute consumption;

[0157] For the kth unmanned sanitation vehicle, resource constraints are imposed during its working process to ensure that the resources it carries are sufficient to support the consumption of completing the scheduled work location.

[0158] τ represents the upper limit of the operating time of the unmanned sanitation vehicle;

[0159] speed k Represents the operating speed of the k-th driverless sanitation vehicle. This constraint ensures that when multiple driverless sanitation vehicles work simultaneously, the longest operating duration does not exceed τ.

[0160] Constrains the operating duration limit of the k-th driverless sanitation vehicle.

[0161] E k Is the energy value (electric energy or fuel energy) of the k-th driverless sanitation vehicle;

[0162] Ensures that the energy carried by the k-th driverless sanitation vehicle can support its operation and replenishment.

[0163] The energy value (electric energy or fuel energy) of the driverless sanitation vehicle k

[0164] If the operation area is abstracted as an operation graph, it is represented by G = {V, ε}.

[0165] Among them, V represents the set of road nodes in the operation area; ε represents the set of edges in the operation area. Then the constraint Can ensure full coverage of the road.

[0166] In summary, the advantages of the present invention are as follows:

[0167] First, comprehensive consideration of factors: The present invention comprehensively considers factors such as the cost of driverless sanitation vehicles, the geographical topology of the operation area, the spatio-temporal requirements of different operation tasks, the reasonable allocation of operation time periods, the influence of seasonal changes, the influence of environmental temperature, the restriction of weather information on the operation cost of driverless sanitation vehicles, and the current position and upcoming target position of driverless sanitation vehicles, making the path planning more comprehensive and accurate.

[0168] Second, application of optimization algorithm: By using an optimization algorithm for path search and optimization, the optimal path can be found under given constraints, improving the efficiency and accuracy of path planning.

[0169] Third, task road network grouping: Using the K-means clustering algorithm to group the task road network, grouping similar road sections into the same category, and then assigning each sub-task to a driverless operation vehicle. Finally, the optimal number of driverless sanitation vehicles is determined through multiple groupings, which can reduce the complexity and computational amount of path planning and make the path planning more efficient.

[0170] Fourth, the application of the ant colony algorithm: For each sub-task after clustering, the ant colony algorithm is used to determine the working path and non-working path of each unmanned sanitation vehicle. This heuristic algorithm inspired by the foraging behavior of ants in nature can effectively solve the path planning problem and improve the efficiency and accuracy of path planning. By introducing heuristic factors and pheromone concentration, it can better guide the unmanned sanitation vehicle to select the optimal path and improve the efficiency and performance of path planning.

[0171] 5. It has generality and scalability, is applicable to various complex road conditions and task requirements, and has high application value and market prospects.

[0172] To better illustrate the embodiments of the present invention, the following experiments were conducted in this embodiment:

[0173] The sample set used in the embodiments of the present invention includes a total of 3 real-scenario operation tasks. The specific experimental process is as follows:

[0174] I. Performance indicators

[0175] The embodiments of the present invention use: the proportion of the overall non-working path, the number of vehicles, and the proportion of the non-working path per 1 km traveled by each unmanned vehicle.

[0176] II. Comparative experiments

[0177] K-means_greedy: Task allocation is performed using the K-means algorithm, and the greedy algorithm is used to find the optimal path;

[0178] Naive_greedy: A greedy algorithm without task allocation;

[0179] The present invention uses K-means_greedy, Naive_greedy, and K-means_ant provided by the embodiments of the present invention to conduct a comparative evaluation of the overall performance.

[0180] Figure 3 It shows the ten-fold value of the proportion of the overall non-working path and the number of unmanned vehicles used when three different algorithms (K-means_ant, task_greedy, and Naive_greedy) complete the operation. These two indicators are represented by dark gray and light gray bar charts respectively.

[0181] It can be seen from the chart that the K-means_ant algorithm proposed by the present invention performs excellently in the proportion of the overall non-working path, and its value is relatively low, indicating that the algorithm performs well in path optimization. At the same time, the number of unmanned vehicles used by this algorithm is the least, indicating that it also has good efficiency in terms of resource utilization (i.e., the number of unmanned vehicles).

[0182] In contrast, the Naive_greedy algorithm has a higher value in the proportion of non-working paths and uses more unmanned vehicles than the other two algorithms, which may imply that the performance of this algorithm in path optimization and resource utilization is inferior to that of other algorithms.

[0183] Figure 4 It respectively shows the proportion of non-working paths per 1 km traveled by each unmanned vehicle under three different real sanitation operation scenarios (Scenario 1, Scenario 2, Scenario 3). This proportion is represented by a bar chart. At the same time, the scatter plot in the figure shows the proportion of non-working paths of a single unmanned vehicle corresponding to different algorithms. The algorithms compared include K-means_ant, task_greedy, and Naive_greedy.

[0184] From Figure 4 It can be seen that regardless of the scenario, the proportion of non-working paths of the K-means_ant algorithm is relatively low, which means that this algorithm performs well in optimizing the driving path of unmanned vehicles. At the same time, the scatter plot also shows that this algorithm also has a good performance in the proportion of non-working paths of a single vehicle.

[0185] The experimental results show that the K-means_ant algorithm shows good performance in reducing the non-working paths of unmanned vehicles, that is, improving the driving efficiency.

[0186] In summary, the method proposed in the present invention performs best in three indicators: the overall proportion of non-working paths, the number of vehicles, and the proportion of non-working paths per 1 km traveled by each unmanned vehicle.

[0187] The embodiment of the present invention also discloses a path planning device for collaborative operation of unmanned sanitation vehicles, including:

[0188] An operation area determination module, configured to determine the operation areas of a number of unmanned sanitation vehicles;

[0189] A sanitation vehicle number determination module, configured to establish a task road network based on the operation area, use a clustering algorithm to group the task road network, classify adjacent road sections into the same category to obtain a number of subtasks, assign each subtask to an unmanned sanitation vehicle, and determine the optimal number of unmanned sanitation vehicles through multiple groupings;

[0190] A path determination module, configured to use the ant colony algorithm to determine the working path and non-working path of each unmanned sanitation vehicle for each clustered subtask;

[0191] The working path is the path followed by the unmanned sanitation vehicle when performing tasks;

[0192] The non-operation path is the path followed by the unmanned sanitation vehicle when going to the working path or the supply station;

[0193] The path planning model establishment module is used to establish a path planning model based on one or more of the cost of the unmanned sanitation vehicle, the geographical topology of the operation area, the spatio-temporal requirements of different operation tasks, the reasonable allocation of the operation time period, the influence of seasonal changes, the influence of environmental temperature, the restriction of weather information on the operation cost of the unmanned sanitation vehicle, the current position of the unmanned sanitation vehicle, and the target position of the unmanned sanitation vehicle;

[0194] The path search module is used to perform path search and optimization based on the path planning model using an optimization algorithm to find the optimal path under given constraints.

[0195] Furthermore, the sanitation vehicle quantity determination module includes:

[0196] The clustering initialization unit is used to initialize the number of clusters to an initial value;

[0197] The clustering unit is used to divide the task roads in the task road network into clusters of the number of clusters using a clustering algorithm and use a path planning algorithm to solve the path planning of each subtask to obtain the optimal path and path length of each cluster;

[0198] The clustering algorithm termination judgment unit is used to calculate the operation duration of all subtask roads and judge whether the termination condition is satisfied;

[0199] If not, if the sum of the path lengths of the subtasks does not reach the expected minimum value, increase the number of clusters and repeat the methods in the clustering unit and the judgment unit until the sum of the path lengths of the subtasks reaches the expected minimum value or cannot be reduced further, and determine the current number of clusters as the optimal number of unmanned sanitation vehicles;

[0200] If satisfied, the current number of clusters is the optimal number of unmanned sanitation vehicles.

[0201] Furthermore, the path determination module includes:

[0202] The ant colony initialization unit is used to initialize the pheromone trails on all road segments to a constant and randomly place ants on the starting node of the search space;

[0203] The ant colony movement unit is used for each ant to select the next road segment to move according to the pheromone concentration and heuristic information;

[0204] Judge whether the remaining resources can reach the specified road segment;

[0205] If it can, the ant moves, updating the resource balance, working path and non-working path of the unmanned sanitation vehicle;

[0206] Otherwise, after replenishing at the optimal replenishment station, the ants move, and the remaining resources, working path, and non-working path of the unmanned sanitation vehicle are updated;

[0207] The repeating execution unit is used to repeatedly execute the method in the ant colony movement unit until all road segments are traversed;

[0208] The duration calculation unit is used to calculate the duration of task completion;

[0209] The duration judgment unit is used to judge whether the duration meets the operation duration limit;

[0210] If it meets the requirement, record the path information of all ants and update the pheromone;

[0211] If it does not meet the requirement, set the path cost to infinity, record the path information of all ants, and update the pheromone;

[0212] The ant colony algorithm termination judgment unit is used to judge whether the maximum number of iterations is reached or the convergence condition is met;

[0213] If so, return the path with the optimal cost;

[0214] Otherwise, repeatedly execute the methods in the ant colony movement unit, repeating execution unit, duration calculation unit, duration judgment unit, and ant colony algorithm termination judgment unit in sequence.

[0215] Furthermore, in the path planning model establishment module, the path planning model is as follows:

[0216]

[0217] Among them:

[0218] F·Q represents the total cost required for using Q unmanned sanitation vehicles, and F represents the fixed cost of each unmanned sanitation vehicle;

[0219] represents the total energy consumption of the unmanned sanitation vehicle on the working path and non-working path;

[0220] is the operation identifier of the kth unmanned sanitation vehicle, represents that the kth unmanned sanitation vehicle is in the operation state at the t-step decision, represents that the kth unmanned sanitation vehicle is in the non-operation state at the t-step decision, is the node and the distance between, and are the position points of the kth unmanned sanitation vehicle at the t-step decision and t + 1-step decision respectively, w is the unit distance energy consumption cost of the working path, and r is the unit distance energy consumption cost of the non-working path;

[0221] It represents the ratio of the operating path length to the non-operating path length of the unmanned sanitation vehicle, and β is the adjustment coefficient.

[0222] Furthermore, the value of w is different in different operating periods, different temperatures, different weather information, different operating cities, and different operating regions.

[0223] Furthermore, in the path search module, the limiting conditions include one or more of the following: resource constraints during the working process of the unmanned sanitation vehicle, operating duration limit of the unmanned sanitation vehicle, and energy constraint carried by the unmanned sanitation vehicle.

[0224] In this embodiment, the specific content of the path planning device for the collaborative operation of unmanned sanitation vehicles is similar to the content of the path planning method for the collaborative operation of unmanned sanitation vehicles disclosed in the above embodiment, and will not be elaborated here.

[0225] In addition, an embodiment of the present invention discloses a storage medium storing one or more computer-readable programs, and the one or more programs include instructions suitable for being loaded and executed by a memory to perform the path planning method for the collaborative operation of unmanned sanitation vehicles disclosed in any of the above embodiments.

[0226] The present invention discloses a path planning method for the collaborative operation of unmanned sanitation vehicles, which has the following beneficial effects:

[0227] First, in view of diverse types of unmanned sanitation vehicles to meet their respective specific operating areas and task requirements, and considering multiple limiting conditions such as operating time periods, seasons, environmental temperatures, weather conditions, and geographical locations, the present invention designs a comprehensive path planning model to achieve optimal efficiency and minimum cost.

[0228] Second, the path planning model of the present invention covers the operating paths and non-operating paths of unmanned sanitation vehicles to ensure that unmanned sanitation vehicles can fully exert their benefits in practical applications.

[0229] Third, the present invention can effectively arrange the paths of unmanned sanitation vehicles to improve transportation efficiency and reduce costs.

[0230] It should be understood that the various technologies described herein can be implemented in combination with hardware or software, or a combination thereof. Thus, the method and device of the present invention, or certain aspects or parts of the method and device of the present invention, can take the form of program code (i.e., instructions) embedded in a tangible medium, such as a floppy disk, CD-ROM, hard disk drive, or any other machine-readable storage medium, where when the program is loaded into a machine such as a computer and executed by the machine, the machine becomes a device for practicing the present invention.

[0231] The basic principles, main features and advantages of the present invention have been shown and described above. Those skilled in the art should understand that the present invention is not limited by the above embodiments. What is described in the above embodiments and the specification only illustrates the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention will have various changes and improvements, and these changes and improvements all fall within the scope of the present invention claimed. The scope of the present invention claimed is defined by the appended claims and their equivalents.

Claims

1. A path planning method for collaborative operation of unmanned sanitation vehicles, characterized in that, Including: Step S1: Determine the operation areas of a number of unmanned sanitation vehicles; Step S2: Based on the operation areas, establish a task road network, group the task road network using a clustering algorithm, classify adjacent road segments into the same category to obtain a number of subtasks, assign each subtask to an unmanned sanitation vehicle, and determine the optimal number of unmanned sanitation vehicles through multiple groupings; Step S3: For each subtask after clustering, use the ant colony algorithm to determine the working path and non-working path of each unmanned sanitation vehicle; The working path is the path followed by the unmanned sanitation vehicle when performing tasks; The non-working path is the path followed by the unmanned sanitation vehicle when going to the working path or the supply station; Step S4: Based on one or more of the cost of the unmanned sanitation vehicle, the geographical topology of the operation area, the spatio-temporal requirements of different operation tasks, the reasonable allocation of the operation time period, the influence of seasonal changes, the influence of environmental temperature, the restriction of weather information on the operation cost of the unmanned sanitation vehicle, the current position of the unmanned sanitation vehicle, and the target position of the unmanned sanitation vehicle, establish a path planning model; The path planning model includes: the total cost required by the unmanned sanitation vehicle, the total energy consumption of the unmanned sanitation vehicle on the working path and the non-working path, and the ratio of the length of the working path to the length of the non-working path of the unmanned sanitation vehicle; Step S5: Based on the path planning model, use an optimization algorithm for path search and optimization to find the optimal path under given constraints.

2. The path planning method according to claim 1, wherein The said Step S2 includes: Step S2.1: Initialize the number of clusters to an initial value; Step S2.2: Use the clustering algorithm to divide the task roads in the task road network into clusters of the number of clusters, and use the path planning algorithm to solve the path planning of each subtask to obtain the optimal path and path length of each cluster; Step S2.3: Calculate the operation duration of all subtask roads and determine whether the termination condition is met; If not, if the sum of the path lengths of the subtasks does not reach the expected minimum value, increase the number of clusters and repeat Steps S2.2 - S2.3 until the sum of the path lengths of the subtasks reaches the expected minimum value or cannot be reduced further, then determine the current number of clusters as the optimal number of unmanned sanitation vehicles; If met, the current number of clusters is the optimal number of unmanned sanitation vehicles.

3. The path planning method according to claim 1, wherein The said Step S3 includes: Step S3.1: Initialize the pheromone trails on all road segments to a constant, and randomly place ants at the starting node of the search space; Step S3.2: Each ant selects the next road segment to move according to the pheromone concentration and heuristic information; Judge whether the remaining resources can reach the specified road segment; If yes, the ant moves, and update the resource margin, working path and non-working path of the unmanned sanitation vehicle; Otherwise, after selecting the optimal supply station for replenishment, the ant moves, and update the resource margin, working path and non-working path of the unmanned sanitation vehicle; Step S3.3: Repeat Step S3.2 until all road segments are traversed; Step S3.4: Calculate the duration of task completion; Step S3.5: Judge whether the duration meets the operation duration limit; If met, record the path information of all ants and update the pheromone; If not satisfied, set the path cost to infinity, record the path information of all ants, and update the pheromone; Step S3.6: Used to determine whether the maximum number of iterations is reached or the convergence condition is satisfied; If so, return the path with the optimal cost; Otherwise, repeat Steps S3.2 - S3.

6.

4. The path planning method according to claim 1, characterized in that In the said Step S4, the path planning model is as follows: Where: F·Q represents the total cost required to use Q driverless sanitation vehicles, and F represents the fixed cost of each driverless sanitation vehicle; Indicates the total energy consumption of the unmanned sanitation vehicle on the working path and the non-working path; is the operation identifier for the k-th unmanned sanitation vehicle, indicating that the k-th unmanned sanitation vehicle is in the operation state at the t-th step of decision-making, indicating that the k-th unmanned sanitation vehicle is in the non-operation state at the t-th step of decision-making, is the node and the distance between, and are the position points of the k-th unmanned sanitation vehicle at the t-th step of decision-making and the (t + 1)-th step of decision-making respectively. w is the energy consumption cost per unit distance of the operation path, and r is the energy consumption cost per unit distance of the non-working path; represents the ratio of the operating path length of the unmanned sanitation vehicle to the non-operating path length, and β is the adjustment coefficient.

5. The path planning method according to claim 4, wherein The value of w is different in different operation periods, different temperatures, different weather information, different operation cities, and different operation areas.

6. The path planning method according to claim 1, characterized in that In the said Step S5, the restrictive conditions include one or more of the following: resource constraints during the working process of the driverless sanitation vehicle, operation duration limit of the driverless sanitation vehicle, and energy constraint carried by the driverless sanitation vehicle.

7. Path planning device for collaborative operation of unmanned sanitation vehicles, characterized in that, Including: An operation area determination module, used to determine the operation areas of several driverless sanitation vehicles; A sanitation vehicle quantity determination module, used to establish a task road network based on the operation area, group the task road network using a clustering algorithm, classify adjacent road segments into the same category to obtain several subtasks, assign each subtask to a driverless sanitation vehicle, and determine the optimal number of driverless sanitation vehicles through multiple groupings; A path determination module, used to, for each subtask after clustering, use the ant colony algorithm to determine the working path and non - working path of each driverless sanitation vehicle; The said working path is the path followed by the driverless sanitation vehicle when performing tasks; The said non - working path is the path followed by the driverless sanitation vehicle when going to the working path or the supply station; A path planning model establishment module, used to establish a path planning model based on one or more factors such as the cost of the driverless sanitation vehicle, the geographical topology of the operation area, the spatio - temporal requirements of different operation tasks, the reasonable allocation of operation time periods, the influence of seasonal changes, the influence of environmental temperature, the restriction of weather information on the operation cost of the driverless sanitation vehicle, the current position of the driverless sanitation vehicle, and the target position of the driverless sanitation vehicle; The said path planning model includes: the total cost required for the driverless sanitation vehicle, the total energy consumption of the driverless sanitation vehicle on the working path and the non - working path, and the ratio of the length of the working path to the length of the non - working path of the driverless sanitation vehicle; A path search module, used to, based on the path planning model, use an optimization algorithm for path search and optimization to find the optimal path under the given restrictive conditions.

8. The path planning device according to claim 7, wherein The said sanitation vehicle quantity determination module includes: A clustering initialization unit, used to initialize the number of clusters to an initial value; A clustering unit, used to use a clustering algorithm to divide the task roads in the task road network into clusters with the number of clusters, and use a path planning algorithm to solve the path planning of each subtask to obtain the optimal path and path length of each cluster; A clustering algorithm termination judgment unit, used to calculate the operation duration of all subtask roads and judge whether the termination condition is satisfied; If not satisfied, if the sum of the path lengths of the subtasks does not reach the expected minimum value, increase the number of clusters, and repeat the methods in the clustering unit and the judgment unit until the sum of the path lengths of the subtasks reaches the expected minimum value or cannot be reduced further, and determine the current number of clusters as the optimal number of driverless sanitation vehicles; If it is satisfied, the current number of clusters is the optimal number of driverless sanitation vehicles.

9. The path planning device according to claim 7, wherein The path determination module includes: An ant colony initialization unit, which is used to initialize the pheromone trails on all road segments to a constant and randomly place ants on the starting node of the search space; An ant colony movement unit, which is used for each ant to select the next road segment to move according to the pheromone concentration and heuristic information; Judge whether the remaining resources can reach the specified road segment; If it can, the ant moves, and the remaining resources, working path, and non-working path of the driverless sanitation vehicle are updated; Otherwise, after selecting the optimal supply station for replenishment, the ant moves, and the remaining resources, working path, and non-working path of the driverless sanitation vehicle are updated; A repeated execution unit, which is used to repeatedly execute the method in the ant colony movement unit until all road segments are traversed; A duration calculation unit, which is used to calculate the duration of task completion; A duration judgment unit, which is used to judge whether the duration meets the operation duration limit; If it is satisfied, record the path information of all ants and update the pheromone; If it is not satisfied, set the path cost to infinity, record the path information of all ants, and update the pheromone; An ant colony algorithm termination judgment unit, which is used to judge whether the maximum number of iterations is reached or the convergence condition is satisfied; If so, return the path with the optimal cost; Otherwise, repeatedly execute the methods in the ant colony movement unit, the repeated execution unit, the duration calculation unit, the duration judgment unit, and the ant colony algorithm termination judgment unit in sequence.

10. The path planning device according to claim 7, wherein In the path planning model establishment module, the path planning model is as follows: Where: F·Q represents the total cost required to use Q driverless sanitation vehicles, and F represents the fixed cost of each driverless sanitation vehicle; It represents the total energy consumption of the unmanned sanitation vehicle on the working path and the non-working path; is the operation identifier of the k-th unmanned sanitation vehicle, indicating that the k-th unmanned sanitation vehicle is in the operation state at the t-th step of decision-making, indicating that the k-th unmanned sanitation vehicle is in the non-operation state at the t-th step of decision-making, is the node and the distance between, and are the position points of the k-th unmanned sanitation vehicle at the t-th step of decision-making and the (t + 1)-th step of decision-making respectively. w is the unit distance energy consumption cost of the operation path, and r is the unit distance energy consumption cost of the non-working path; It represents the ratio of the operation path length of the unmanned sanitation vehicle to the non-working path length, and β is the adjustment coefficient.

11. The path planning device according to claim 10, characterized in that, The value of w is different in different operation periods, different temperatures, different weather information, different operation cities, and different operation areas.

12. The path planning device according to claim 7, wherein, In the path search module, the limiting conditions include one or more of the following: resource constraints during the operation process of the driverless sanitation vehicle, operation duration limit of the driverless sanitation vehicle, and energy constraint carried by the driverless sanitation vehicle.

13. Storage medium, characterized in that, The storage medium stores one or more computer-readable programs, and the one or more programs include instructions, and the instructions are suitable for being loaded and executed by the memory to execute the path planning method for collaborative operation of driverless sanitation vehicles according to any one of claims 1-6 above.

Images