Sanitation vehicle intelligent scheduling method and system, and storage medium

Abstract

The invention relates to a sanitation vehicle intelligent scheduling method and system and a storage medium, and the method comprises the steps: a data collection step: collecting environment data and vehicle information in real time, the environment data including garbage accumulation amount, traffic condition and weather forecast, and the vehicle information including vehicle position and state; a data processing step: cleaning and formatting the collected environment data and vehicle information, and constructing a feature vector and an input matrix; a scheduling optimization step: processing the data generated in the data processing step by using a quantum heuristic evolutionary algorithm, and finding out an optimal scheduling scheme; and an execution monitoring step: executing a cleaning task according to the optimal scheduling scheme generated in the optimization scheduling step, monitoring the cleaning progress in real time, collecting the cleaning progress and state information, and feeding back the information to the data acquisition step to form closed-loop control. According to the invention, the accuracy and reliability of scheduling decisions are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of vehicle dispatching, and in particular to an intelligent dispatching method, system and storage medium for sanitation vehicles. Background Art

[0002] Existing sanitation vehicle scheduling methods mainly rely on manual experience or simple GPS tracking systems to plan cleaning routes and assign tasks. For example, street cleaning vehicles are dispatched according to a fixed schedule. However, in this way, the scheduling of vehicles mainly depends on past experience and current cleaning needs, but fails to fully consider the impact of factors such as real-time traffic conditions, weather changes, and emergencies, resulting in low resource utilization and unsatisfactory cleaning results.

[0003] In order to improve this situation, some sanitation vehicle dispatching systems use optimization algorithm-based methods, such as genetic algorithms and ant colony algorithms, to seek the optimal cleaning path. For example, when a large amount of garbage suddenly accumulates in a certain area, the dispatching center will start an emergency response program. The program first uses GPS positioning technology to obtain the real-time location of all vehicles and draws a real-time map in combination with the geographic information system (GIS). After that, the dispatcher will choose an appropriate optimization algorithm (such as genetic algorithm) to replan the cleaning route. Although this method has improved the dispatching efficiency to a certain extent, there are still several shortcomings: the control strategy based on traditional optimization algorithms often converges slowly, especially when dealing with large-scale problems, and fails to make full use of real-time data, resulting in low accuracy of the final dispatching plan. In addition, these methods have limited capabilities in multi-objective optimization, and it is difficult to find an ideal balance between multiple objectives such as cost control, work efficiency improvement, and environmental protection.

[0004] Therefore, it is necessary to develop a new intelligent scheduling method, system and storage medium for sanitation vehicles. Summary of the invention

[0005] The object of the present invention is to provide a method, system and storage medium for intelligent dispatching of sanitation vehicles, which can improve the accuracy and reliability of dispatching decisions.

[0006] In a first aspect, a method for intelligently dispatching sanitation vehicles according to the present invention comprises the following steps:

[0007] Data collection step: real-time collection of environmental data and vehicle information, wherein the environmental data includes garbage accumulation, traffic conditions and weather forecasts, and the vehicle information includes vehicle location and status;

[0008] Data processing steps: clean and format the collected environmental data and vehicle information, and construct feature vectors and input matrices;

[0009] Optimize the scheduling step: Use the quantum-inspired evolutionary algorithm to process the data generated by the data processing step to find the optimal scheduling solution;

[0010] Execution monitoring step: Execute the cleaning task according to the optimal scheduling plan generated by the optimization scheduling step, monitor the cleaning progress in real time, collect the cleaning progress and status information, and feed the information back to the data collection step to form a closed-loop control to ensure the continuous optimization of the scheduling system. At the same time, this dynamic scheduling mechanism based on real-time data improves the accuracy and reliability of scheduling decisions.

[0011] Optionally, the data processing step is specifically:

[0012] Step 21: Clean and format the collected environmental data and vehicle information;

[0013] Step 22: Construct feature vector:

[0014] Calculate the amount of garbage accumulated in each area; build a vehicle availability matrix based on the current location and status of the vehicle, including fuel level and whether it is performing a task; calculate the travel time of each road section based on real-time traffic data; predict and preset weather factors that will affect the cleaning operation in the future based on weather forecast data;

[0015] The formula for constructing the feature vector is as follows:

[0016]

[0017] Among them, V ij is a matrix, i represents the vehicle, j represents the area, if vehicle i can clear area j, then V ij =1, otherwise 0;

[0018] Step 23: Construct the input matrix. The construction formula of the input matrix is ​​as follows:

[0019]

[0020] Among them, M ij is the input matrix, which is an m×n matrix, where m is the number of vehicles and n is the number of regions; if vehicle i can go to region j, then M ij =1, otherwise 0.

[0021] Optionally, a quantum-inspired evolutionary algorithm is used to process the data generated by the data processing step to find the optimal scheduling solution, specifically:

[0022] Step 31: Initialize the population: Generate a set of random scheduling schemes as the initial population, each scheduling scheme represents an initial individual;

[0023] Step 32: Fitness evaluation: Evaluate the fitness value of each individual in the initial population according to the fitness function;

[0024] Step 33: Selection operation: According to the fitness value calculated in step 32, select the individuals with better performance as the parent individuals of the next generation population;

[0025] Step 34: Crossover operation: Based on the parent individuals selected in step 33, a crossover operation is performed to generate new individuals;

[0026] Step 35: Mutation operation: Randomly mutate the new individual generated in step 34, that is, change some attributes of the individual with a preset probability to generate a mutated individual;

[0027] Step 36: Update probability amplitude: Use quantum revolving door technology to update the probability amplitude of the individual mutated in step 35.

[0028] Step 37: Iterative optimization: Repeat steps 33 to 36 until a predetermined stop condition is reached or the optimal scheduling solution is found. Based on the formed input matrix, the present invention creates an initial population through QEA, and efficiently finds the optimal scheduling solution through a series of operations such as updating probability amplitude, selection, crossover and mutation, fitness evaluation, etc. QEA simulates the state superposition and entanglement characteristics of quantum bits, enhances search capabilities and adaptability, and is particularly outstanding in multi-objective optimization, and can better balance multiple goals such as cost, efficiency and environmental protection.

[0029] Optionally, the step 31 is specifically as follows:

[0030] Based on the input matrix constructed in step 23, an initial population of qubit encoding is created, where each individual x represents a possible scheduling scheme, and finally a series of individuals {x1, x2, …, x n}. Provide a basic individual set for the subsequent optimization process.

[0031] Optionally, the step 32 is specifically:

[0032] Based on the initial population formed in step 31, the fitness value of each individual x in the initial population is evaluated according to the fitness function;

[0033] The fitness function is defined as:

[0034] f(x)=w1·cost(x)+w2·efficiency(x)+w3·eco(x)

[0035] Among them, f(x) is the fitness value of individual x; w1 is the cost weight coefficient, w2 is the efficiency weight coefficient, w3 is the environmental protection weight coefficient, w1+w2+w3=1, w1, w2, w3 correspond to the importance of cost, efficiency and environmental protection respectively; cost(x) represents cost; efficiency(x) represents efficiency; eco(x) represents environmental protection index. Based on the fitness function, the fitness value of each individual in the initialized population is calculated to evaluate the adaptability of the individual in the current environment and provide a basis for the subsequent selection, crossover and mutation operations.

[0036] Optionally, the step 34 is specifically as follows:

[0037] Select two parent individuals, randomly choose a crossover point, and exchange some of the qubits of the two individuals;

[0038] The crossover operation formula is:

[0039]

[0040] Among them, P1 represents parent individual 1, P2 represents parent individual 2, c is the crossover point, and C(P1, P2, c) is the two new individuals after the crossover. By combining the dominant genes of different individuals, new individuals are generated, increasing the diversity of the population.

[0041] Optionally, the step 35 is specifically as follows:

[0042] Construct the mutation operation formula, which is as follows:

[0043]

[0044] Among them, x is an individual; x ′ is the individual after mutation; a is the position where mutation occurs; A(x,a) is the result of mutation operation. If mutation occurs at position a, then A(x,a) is equal to x. ′ , that is, x in individual x a Replace with x a ′ , where x a represents the state of the qubit at position a in individual x before mutation, x a ′ is the state of the qubit at position a in individual x after mutation. Some properties of the individual are changed with a certain probability, new variants are introduced, and the diversity of the population is increased.

[0045] Optionally, the step 36 is specifically as follows:

[0046] The function of a quantum rotating gate is simulated by updating the angle of the qubit, as follows:

[0047]

[0048] Among them, θ b,x is the angle of individual x at qubit b; is the angle of the quantum bit b of individual x at time t+1; is the angle of qubit b of individual x at time t; α is the learning rate; is the angle update calculated based on the fitness function, specifically the fitness function f(x) for the qubit θ b,x The partial derivative of , which represents the quantum bit θ b,x The effect of changes in the fitness function f(x). Adjust the angle of the qubit according to the fitness value of the individual to optimize the performance of the individual, and adjust the angle based on the mutated individual to make it more competitive in future performance.

[0049] In the second aspect, a sanitation vehicle intelligent dispatching system described in the present invention includes a memory and a controller, wherein the memory stores a computer-readable program, and when the computer-readable program is called by the controller, it can execute the steps of the sanitation vehicle intelligent dispatching method described in the present invention.

[0050] In a third aspect, a storage medium described in the present invention stores a computer-readable program therein, and when the computer-readable program is called by a controller, it can execute the steps of the intelligent scheduling method for sanitation vehicles described in the present invention.

[0051] Beneficial effects of the present invention: The present invention constructs an intelligent scheduling system and method for sanitation vehicles by introducing a quantum-inspired evolutionary algorithm (QEA), which effectively solves the problems existing in traditional scheduling methods. Specifically, the present invention collects key data such as garbage accumulation and traffic conditions in real time by deploying Internet of Things devices, and uses QEA to analyze the data to quickly generate the optimal cleaning path. QEA improves the search efficiency by simulating the state superposition and entanglement characteristics of quantum bits, enabling the system to find a scheduling solution close to the optimal solution in a short time. As a result, the present invention not only improves the accuracy and response speed of scheduling decisions, but also finds a better balance between multiple objectives such as cost control, cleaning efficiency improvement, and environmental protection, thereby significantly reducing the cleaning cost and improving the overall efficiency and environmental protection level of the cleaning work. In addition, the dynamic scheduling capability of the present invention based on real-time data enables the system to make rapid adjustments in the face of emergencies, ensuring the efficient completion of cleaning tasks. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 is a flow chart of the intelligent dispatching method for sanitation vehicles described in an embodiment of the present application;

[0053] Figure 2 is a flow chart of the data collection steps in the embodiment of the present application;

[0054] Figure 3 is a flow chart of the data processing steps in the embodiment of the present application;

[0055] Figure 4 is a flowchart of the optimization scheduling steps in an embodiment of the present application;

[0056] Figure 5 It is a principle block diagram of the intelligent dispatching system for sanitation vehicles described in the embodiments of the present application;

[0057] Figure 6 It is a functional module diagram of a computer-readable program in an embodiment of the present application. DETAILED DESCRIPTION

[0058] The following will describe the embodiments of the present invention with reference to the accompanying drawings and preferred embodiments. Those skilled in the art can 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 understood that the preferred embodiments are only for illustrating the present invention, not for limiting the scope of protection of the present invention.

[0059] In an embodiment of the present application, a method for intelligent scheduling of sanitation vehicles is a method for intelligent scheduling of sanitation vehicles based on a quantum-inspired evolutionary algorithm. The quantum-inspired evolutionary algorithm explores the optimal solution in the solution space by simulating the state superposition and entanglement characteristics of quantum bits. In the present invention, the scheduling problem of sanitation vehicles is converted into an optimization problem, the goal of which is to minimize the total cleaning cost while maximizing the cleaning efficiency and environmental protection index.

[0060] like Figure 1 As shown, in an embodiment of the present application, a method for intelligent dispatching of sanitation vehicles includes the following specific steps:

[0061] 1. Data collection steps

[0062] Collect environmental data and vehicle information in real time; environmental data includes but is not limited to garbage accumulation, traffic conditions and weather forecasts; vehicle information includes vehicle location and status; and transmit the collected data to the data center for aggregation.

[0063] Specifically, IoT devices are used to collect information about garbage accumulation, traffic conditions, vehicle location and status in each area, and weather forecasts for the next period of time are obtained through the weather forecast API. Finally, the collected environmental data and vehicle information are aggregated to the data center.

[0064] like Figure 2 As shown, the data collection steps specifically include:

[0065] 1.1: Deploy IoT devices

[0066] Sensors are installed in garbage areas to monitor the amount of garbage accumulation in real time.

[0067] Install cameras on the road and obtain traffic flow information through the cameras.

[0068] Install a GPS device on the sanitation vehicle and obtain the real-time location of the vehicle through the GPS device.

[0069] A vehicle status acquisition module is provided on the sanitation vehicle, through which the vehicle status is acquired, including the fuel level and whether a task is being performed.

[0070] Get weather forecast data from weather stations or third-party APIs.

[0071] 1.2: Aggregate Data

[0072] The collected data is transmitted to the data center for aggregation.

[0073] 2. Data processing steps

[0074] Based on the data collected in the data collection step, feature extraction and formatting are performed, and feature vectors and input matrices are constructed to provide necessary input for the optimization scheduling step. In other words, the data processing step works based on the original data provided by the data collection step to ensure that the quality and format of the data are suitable for further processing. Its output is the input of the optimization scheduling step, which needs to optimize the scheduling plan based on the processed data.

[0075] In a possible embodiment, Figure 3 As shown, the data processing steps specifically include:

[0076] Step 21: Data cleaning and formatting, that is, cleaning and formatting the data in the data collection step.

[0077] Clean the data and remove invalid or duplicate records.

[0078] Format data fields to ensure consistency and accuracy.

[0079] Step 22: Construct feature vector:

[0080] Calculate the amount of garbage accumulated in each area.

[0081] A vehicle availability matrix is ​​constructed based on the vehicle's current location and status, where the status includes fuel level and whether the vehicle is performing a mission.

[0082] Calculate the travel time of each road section based on real-time traffic data.

[0083] Based on the weather forecast data, the weather factors that will affect the cleaning operation in the future are predicted and preset.

[0084] Assuming that there are currently five cleaning areas and three sanitation vehicles, the constructed feature vector is a two-dimensional array, where rows represent areas and columns represent vehicles. An element value of 1 indicates that the vehicle can go to the area, and a value of 0 indicates that the vehicle cannot go to the area.

[0085] The formula for constructing the feature vector is as follows:

[0086]

[0087] Among them, V ij is a matrix, i represents the vehicle, j represents the area, if vehicle i can clear area j, then V ij =1, otherwise 0.

[0088] Step 23: Construct an input matrix where each row represents a vehicle and each column represents an area. Each element represents the probability of the vehicle going to the area.

[0089] The input matrix formula is as follows:

[0090]

[0091] Among them, M ij is the input matrix, which is an m×n matrix, where m is the number of vehicles and n is the number of regions; if vehicle i can go to region j, then M ij =1, otherwise 0.

[0092] 3. Optimize scheduling steps

[0093] The data generated by the data processing step is processed using a quantum-inspired evolutionary algorithm to find the optimal scheduling solution. The total cleaning cost is minimized while maximizing the cleaning efficiency and environmental protection index. That is, the optimization scheduling step depends on the input matrix and eigenvector provided by the data processing step to generate the optimal scheduling solution.

[0094] like Figure 4 As shown in the figure, the optimization scheduling steps include population initialization, fitness evaluation, selection operation, crossover operation, mutation operation, probability amplitude update and iterative optimization. Each step is progressive in sequence, where:

[0095] Step 31: Initialize the population: used to generate a set of random scheduling schemes as the initial population, each scheduling scheme represents an initial individual; provide a basic individual set for the subsequent optimization process.

[0096] Step 32: Fitness evaluation: Evaluate the fitness value of each individual in the initial population according to the fitness function; used to evaluate the adaptability of the individual in the current environment and provide a basis for subsequent selection, crossover and mutation operations.

[0097] Step 33: Selection operation: According to the fitness value calculated in step 32, select individuals with better performance as the parent individuals of the next generation population, so as to retain excellent genes and provide high-quality "parents" for the crossover operation.

[0098] Step 34: Crossover operation: Based on the parent individuals selected in step 33, a crossover operation is performed to generate new individuals. Specifically, new individuals are generated by combining the dominant genes of different individuals to increase the diversity of the population.

[0099] Step 35: Mutation operation: Perform random mutation on the new individuals generated in step 34, that is, change certain attributes of the individuals with a preset probability, generate mutated individuals, and increase the diversity of the population.

[0100] Step 36: Update probability amplitude: Use quantum revolving door technology to update the probability amplitude of the individual after the mutation in step 35. This can optimize the performance of the individual and make it more competitive in future performance.

[0101] Step 37: Iterative optimization: Repeat steps 33 to 36 until the predetermined stop condition is reached or the optimal scheduling solution is found. That is, repeat the selection operation, crossover operation, mutation operation and probability amplitude update operation to gradually improve the overall quality of the population until the predetermined stop condition is reached or the optimal scheduling solution is found.

[0102] Through this series of steps, the optimization scheduling step can continuously improve the scheduling plan and ultimately achieve efficient resource allocation and task scheduling. The following is a detailed introduction to each step:

[0103] Step 31: Initialize the population

[0104] Based on the input matrix formed in step 23, an initial population of qubit encoding is created, each individual x represents a possible scheduling scheme, and finally a series of individuals {x1, x2, …, x n},x1,x2,…,x n Represent different individuals.

[0105] Qubit encoding rules: Each qubit represents a decision variable, such as a vehicle's choice of next destination.

[0106] Assume that five areas need to be cleaned and there are three vehicles available for dispatch. The individual x1=[1,0,0,1,0] indicates that vehicle 1 goes to area 1 and area 4.

[0107] Therefore, this step forms an initial population containing multiple individuals, that is, generates a set of scheduling schemes as the initial population.

[0108] Step 32: Fitness Evaluation

[0109] Based on the initial population formed in step 31, the fitness value of each individual x in the initial population is evaluated according to the fitness function.

[0110] The fitness function is defined as:

[0111] f(x)=w1·cost(x)+w2·efficiency(x)+w3·eco(x)

[0112] Among them, f(x) is the fitness value of individual x, w1 is the cost weight coefficient, w2 is the efficiency weight coefficient, w3 is the environmental protection weight coefficient, w1+w2+w3=1, w1, w2, w3 correspond to the importance of cost, efficiency and environmental protection respectively. cost(x) represents the cost, taking into account factors such as vehicle operating cost and fuel consumption. efficiency(x) represents efficiency, which evaluates the speed at which the cleaning task is completed. eco(x) represents the environmental protection index, taking into account factors such as reducing carbon emissions and noise pollution.

[0113] This step obtains the fitness value of each individual, which is used for selection in the next round of iteration.

[0114] Step 33: Select Action

[0115] Based on the fitness value of each individual calculated in step 32, individuals with better performance are selected according to the fitness value to enter the next generation.

[0116] Assuming that individual one has a higher fitness and individual two has a lower fitness, individual one is more likely to be selected to enter the next generation.

[0117] Step 34: Crossover Operation

[0118] Based on the parent individuals selected in step 33, a crossover operation is performed. New individuals are generated by combining excellent genes of different individuals. Specifically, two parent individuals are selected, i.e., as parent individuals, a crossover point is randomly selected, and some qubits of the two individuals are exchanged.

[0119] The crossover operation formula is:

[0120]

[0121] Among them, P1 represents parent individual one, P2 represents parent individual two, c is the crossover point, and C(P1,P2,c) are the two new individuals after the crossover.

[0122] Suppose two individuals [1,0,1,0] and [0,1,1,1] are selected for crossover, and the crossover point c is the second qubit, then the individuals after crossover are [1,1,1,0] and [0,0,1,1].

[0123] Step 35: Mutation Operation

[0124] Based on the individuals formed in step 34, the quantum bit state of a certain individual is changed with a preset probability to introduce diversity.

[0125] The mutation operation formula is:

[0126]

[0127] Among them, x is an individual, x ′ is the individual after mutation, and a is the position where the mutation occurs. A(x,a) is the result of the mutation operation. If the mutation occurs at position a, then A(x,a) is equal to x ′ , that is, x in individual x a Replace with x a ′ , where x a represents the state of the qubit at position a in individual x before mutation, x a ′ is the state of the quantum bit at position a in individual x after mutation.

[0128] Suppose there is an individual x = [1,0,0], which means the first vehicle goes to area 1. If the mutation occurs at position three (i.e. the third position), that is, the value of position three changes from 0 to 1, then: A(x,a) = [1,0,1].

[0129] Step 36: Update the probability amplitude

[0130] Based on the mutated new individuals, the probability amplitude of the individuals is updated through the quantum rotating gate to reflect the impact of the new information.

[0131] The quantum rotating gate is a basic operation in quantum computers that is used to change the state of a qubit. Here, the function of the quantum rotating gate is simulated by updating the angle of the qubit. The formula is as follows:

[0132]

[0133] Among them, θ b,x is the angle of individual x at qubit b, is the angle of the quantum bit b of individual x at time t+1; is the angle of qubit b of individual x at time t, α is the learning rate, is the angle update calculated based on the fitness function, specifically the fitness function f(x) for the qubit θ b,x The partial derivative of , which represents the quantum bit θ b,x The impact of the change on the fitness function f(x). f(x) is the f(x) in step 3.2.

[0134] The purpose of updating the probability amplitude in this step is to adjust the angle of the qubit according to the feedback of the fitness function to optimize the performance of the individual. By adjusting the angle of the qubit, the individual can be guided to evolve in a better direction.

[0135] Step 37: Iterative Optimization

[0136] Steps 33 to 36 are repeatedly executed until a predetermined stop condition is reached or an optimal scheduling solution is found.

[0137] 4. Execute monitoring steps

[0138] Based on the optimal scheduling plan generated by the optimization scheduling step, the cleaning task is executed, and the cleaning progress is monitored in real time. The execution results are collected and the collected cleaning progress and status information are fed back to the data collection step to form a closed-loop control for the input data of the next round of scheduling.

[0139] In this step, the cleaning task is executed according to the optimized path, and the cleaning progress is monitored in real time. At the same time, feedback on the execution results is collected and used as input data for the next round of scheduling. That is, the vehicle goes to the designated area for cleaning according to the optimized path, and uploads the cleaning progress and status information in real time, which will be used as input data for the next round of scheduling.

[0140] This method improves the search capability and adaptability of the algorithm, especially the response speed when dealing with emergencies, by introducing a quantum-inspired evolutionary algorithm. In addition, this method performs better in multi-objective optimization and can better balance multiple objectives such as cost, efficiency and environmental protection. At the same time, the present invention performs the cleaning task according to the optimal solution generated by the optimized scheduling, monitors the cleaning progress in real time, collects the cleaning progress and status information and feeds it back to the data collection step, forms a closed-loop control, and is used as input data for the next round of scheduling. This dynamic scheduling mechanism based on real-time data improves the accuracy and reliability of scheduling decisions.

[0141] This method effectively solves the problem existing in the prior art, that is, the traditional sanitation vehicle dispatching method is too dependent on manual experience or simple GPS tracking system, resulting in low resource utilization efficiency and poor cleaning effect. In particular, when faced with factors such as real-time traffic conditions, weather changes and emergencies, the existing dispatching methods are unable to respond quickly and accurately.

[0142] like Figure 5As shown, in an embodiment of the present application, a sanitation vehicle intelligent scheduling system includes a memory and a controller. A computer-readable program is stored in the memory. When the computer-readable program is called by the controller, it can execute the steps of the sanitation vehicle intelligent scheduling method described in the embodiment of the present application.

[0143] By introducing quantum-inspired evolutionary algorithms, this system designs an intelligent dispatching system for sanitation vehicles that can adjust the cleaning path in real time and excel in multi-objective optimization. The goal of this system is to minimize the total cleaning cost while maximizing the cleaning efficiency and environmental protection index, thereby finding the best balance between cost control, cleaning efficiency improvement, and environmental protection.

[0144] like Figure 6 As shown, in a possible embodiment, the computer-readable program is divided according to functional modules, including a data acquisition module, a data processing module, an optimization scheduling module and an execution monitoring module. Among them:

[0145] Data acquisition module: used to collect environmental data and vehicle information in real time. The environmental data includes garbage accumulation, traffic conditions and weather forecasts, and the vehicle information includes vehicle location and status.

[0146] Data processing module: used to clean and format the collected environmental data and vehicle information, and to construct feature vectors and input matrices.

[0147] Optimization scheduling module: Use quantum-inspired evolutionary algorithms to process the data generated by the data processing module to find the optimal scheduling plan, including the optimal path and cleaning order of the vehicle.

[0148] Execution monitoring module: Executes the cleaning task according to the optimal scheduling plan generated by the optimization scheduling module, monitors the cleaning progress in real time, collects the cleaning progress and status information, and feeds the information back to the data acquisition module to form a closed-loop control.

[0149] Through the collaborative work of the above four modules, the system can achieve effective management and scheduling of sanitation vehicles. The data acquisition module ensures that the system obtains accurate real-time data; the data processing module ensures the availability and consistency of data; the optimization scheduling module uses advanced algorithms to generate the optimal scheduling plan; the execution monitoring module is responsible for supervising the execution of the plan and collecting feedback to ensure the efficient operation of the entire system. This closed-loop control mechanism not only improves the efficiency of cleaning work, but also reduces the impact on the environment, achieving cost savings and resource optimization.

[0150] In an embodiment of the present application, a storage medium stores a computer-readable program therein, and when the computer-readable program is called by a controller, the steps of the intelligent scheduling method for sanitation vehicles as described in the present invention can be executed.

[0151] In an embodiment of the present application, the storage medium may be a tangible storage medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. The storage medium may be a machine-readable signal medium or a machine-readable storage medium. The storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, or equipment, or any suitable combination of the foregoing. A more specific example of a storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0152] The above-mentioned embodiments are preferred implementation modes of the present invention, but the implementation modes of the present invention are not limited to the above-mentioned embodiments. Any other changes, modifications, substitutions, combinations and simplifications made without departing from the spirit and principle of the present invention shall be equivalent replacement modes and shall be included in the protection scope of the present invention.

Claims

1. A method for intelligent dispatching of sanitation vehicles, characterized in that: The following steps are involved: Data collection step: real-time collection of environmental data and vehicle information, wherein the environmental data includes garbage accumulation, traffic conditions and weather forecasts, and the vehicle information includes vehicle location and status; Data processing steps: clean and format the collected environmental data and vehicle information, and construct feature vectors and input matrices; Optimize the scheduling step: Use the quantum-inspired evolutionary algorithm to process the data generated by the data processing step to find the optimal scheduling solution; Execution monitoring step: Execute the cleaning task according to the optimal scheduling plan generated by the optimization scheduling step, monitor the cleaning progress in real time, collect the cleaning progress and status information, and feed the information back to the data collection step to form a closed-loop control.

2. The intelligent dispatching method for sanitation vehicles according to claim 1, characterized in that: The data processing steps are specifically: Step 21: Clean and format the collected environmental data and vehicle information; Step 22: Construct feature vector: Calculate the amount of garbage accumulated in each area; build a vehicle availability matrix based on the current location and status of the vehicle, including fuel level and whether it is performing a task; calculate the travel time of each road section based on real-time traffic data; predict and preset weather factors that will affect the cleaning operation in the future based on weather forecast data; The formula for constructing the feature vector is as follows: Among them, V ij is a matrix, i represents the vehicle, j represents the area, if vehicle i can clear area j, then V ij =1, otherwise 0; Step 23: Construct the input matrix. The construction formula of the input matrix is ​​as follows: Among them, M ij is the input matrix, which is an m×n matrix, where m is the number of vehicles and n is the number of regions; if vehicle i can go to region j, then M ij =1, otherwise 0.

3. The intelligent dispatching method for sanitation vehicles according to claim 2, characterized in that: The data generated by the data processing step is processed using a quantum-inspired evolutionary algorithm to find the optimal scheduling solution, specifically: Step 31: Initialize the population: Generate a set of random scheduling schemes as the initial population, each scheduling scheme represents an initial individual; Step 32: Fitness evaluation: Evaluate the fitness value of each individual in the initial population according to the fitness function; Step 33: Selection operation: According to the fitness value calculated in step 32, select the individuals with better performance as the parent individuals of the next generation population; Step 34: Crossover operation: Based on the parent individuals selected in step 33, a crossover operation is performed to generate new individuals; Step 35: Mutation operation: Randomly mutate the new individual generated in step 34, that is, change some attributes of the individual with a preset probability to generate a mutated individual; Step 36: Update the probability amplitude: Use the quantum rotating gate technology to update the probability amplitude of the individual mutated in step 35; Step 37: Iterative optimization: Repeat steps 33 to 36 until a predetermined stop condition is reached or the optimal scheduling solution is found.

4. The intelligent dispatching method for sanitation vehicles according to claim 3 is characterized in that: The step 31 is specifically as follows: Based on the input matrix constructed in step 23, an initial population of qubit encoding is created, where each individual x represents a possible scheduling scheme, and finally a series of individuals {x1, x2, …, x n }.

5. The intelligent dispatching method for sanitation vehicles according to claim 4, characterized in that: The step 32 is specifically as follows: Based on the initial population formed in step 31, the fitness value of each individual x in the initial population is evaluated according to the fitness function; The fitness function is defined as: f(x)=w1·cost(x)+w2·efficiency(x)+w3·eco(x) Among them, f(x) is the fitness value of individual x; w1 is the cost weight coefficient, w2 is the efficiency weight coefficient, w3 is the environmental protection weight coefficient, w1+w2+w3=1, w1, w2, w3 correspond to the importance of cost, efficiency and environmental protection respectively; cost(x) represents cost; efficiency(x) represents efficiency; eco(x) represents environmental protection index.

6. The intelligent dispatching method for sanitation vehicles according to claim 5, characterized in that: The step 34 is specifically as follows: Select two parent individuals, randomly choose a crossover point, and exchange some of the qubits of the two individuals; The crossover operation formula is: Among them, P1 represents parent individual one, P2 represents parent individual two, c is the crossover point, and C(P1,P2,c) are the two new individuals after the crossover.

7. The intelligent dispatching method for sanitation vehicles according to claim 6, characterized in that: The step 35 is specifically as follows: Construct the mutation operation formula, which is as follows: Among them, x is an individual; x ′ is the individual after mutation; a is the position where mutation occurs; A(x,a) is the result of mutation operation. If mutation occurs at position a, then A(x,a) is equal to x. ′ , and the x in individual x a Replace with x a ′ Get x ′ , where x a represents the state of the qubit at position a in individual x before mutation, x a ′ is the state of the quantum bit at position a in individual x after mutation.

8. The intelligent dispatching method for sanitation vehicles according to claim 7, characterized in that: The step 36 is specifically as follows: The function of a quantum rotating gate is simulated by updating the angle of the qubit, as follows: Among them, θ b,x is the angle of individual x at qubit b; is the angle of the quantum bit b of individual x at time t+1; is the angle of qubit b of individual x at time t; α is the learning rate; is the angle update calculated based on the fitness function, specifically the fitness function f(x) for the qubit θ b,x The partial derivative of , which represents the quantum bit θ b,x The impact of changes in the fitness function f(x).

9. An intelligent dispatching system for sanitation vehicles, characterized in that: It comprises a memory and a controller, wherein the memory stores a computer-readable program, and when the computer-readable program is called by the controller, it can execute the steps of the intelligent scheduling method for sanitation vehicles as described in any one of claims 1 to 8.

10. A storage medium, characterized in that: A computer-readable program is stored therein, and when the computer-readable program is called by the controller, the steps of the intelligent scheduling method for sanitation vehicles as described in any one of claims 1 to 8 can be executed.

Images