Time-dependent vehicle path planning method, device and equipment and storage medium

Abstract

The invention discloses a time-dependent vehicle path planning method, device and equipment and a storage medium, and relates to the technical field of intelligent logistics, the time-dependent vehicle path planning method comprises the following steps: obtaining vehicle information and time-varying road network information, and constructing constraint conditions and a target function according to the vehicle information and the time-varying road network information; obtaining a nectar source set according to the constraint condition, the time-varying road network information and the vehicle information, and constructing a hybrid artificial bee colony strategy according to the nectar source set; and obtaining a target value of a target function according to the constraint condition, the nectar source set and the hybrid artificial bee colony strategy, and performing path planning on the vehicle according to the target value. An objective function value is minimized by constructing an optimization model and adopting a hybrid artificial bee colony strategy, and optimal path planning is carried out on the vehicle according to the objective function value, so that the distribution efficiency is improved, and the cost is reduced.

Description

Technical Field

[0001] The present application relates to the field of smart logistics technology, and in particular to time-dependent vehicle path planning methods, devices, equipment and storage media. Background Art

[0002] With the popularization of the Internet and e-commerce, consumers' expectations for logistics services have become more diversified, and the demand has changed from the traditional single delivery or pickup demand to the distribution demand of the two. However, most of the research on the traditional vehicle routing problem (VRP) only involves delivery, and does not consider the pickup problem, which will cause many problems such as traffic congestion, low distribution efficiency, and high transportation costs. Therefore, some scholars have also studied this VRP extension problem, namely the vehicle routing problem with simultaneous delivery and pickup (VRPSDP). After years of development, VRPSDP and its extension problems have achieved many research results. However, the acceleration of urbanization has brought more and more complex traffic conditions, road traffic is often overloaded, and traffic congestion has become the norm. The vehicle routing problem that considers the time-varying characteristics of traffic congestion is called the time-dependent vehicle routing problem (TDVRP). Most of the existing VRPSDP studies assume that vehicles travel at a constant speed and do not consider the change in vehicle speed, making the VRPSDP model unable to characterize the time-varying characteristics of urban logistics distribution.

[0003] In order to be closer to reality, the Time-Dependent Vehicle Routing Problem with Simultaneous Delivery and Pickup (TDVRPSDP) was derived. However, there are few studies on TDVRPSDP, and the existing TDVRPSDP studies mainly use the "step time-varying speed" which deviates from the actual situation, and most studies ignore the impact of vehicle speed on energy consumption.

[0004] The above contents are only used to assist in understanding the technical solution of the present application and do not constitute an admission that the above contents are prior art. Summary of the invention

[0005] The main purpose of this application is to provide a time-dependent vehicle path planning method, device, equipment and storage medium, aiming to solve the technical problems that most existing vehicle path planning studies fail to fully consider the time-varying characteristics of urban traffic and the impact of vehicle speed changes on energy consumption, and cannot accurately simulate the low distribution efficiency and increased transportation costs caused by traffic congestion and speed changes in the real world.

[0006] To achieve the above objectives, the present application proposes a time-dependent vehicle path planning method, the method comprising:

[0007] Acquire vehicle information and time-varying road network information, and construct constraint conditions and objective functions according to the vehicle information and the time-varying road network information;

[0008] Obtaining a honey source set according to the constraint conditions, the time-varying road network information and the vehicle information, and constructing a hybrid artificial bee colony strategy according to the honey source set;

[0009] The target value of the objective function is obtained according to the constraint conditions, the nectar source set and the hybrid artificial bee colony strategy, and the path of the vehicle is planned according to the target value.

[0010] In one embodiment, the step of acquiring vehicle information and time-varying road network information, and constructing constraint conditions and objective functions according to the vehicle information and the time-varying road network information comprises:

[0011] Obtain vehicle information and time-varying road network information;

[0012] Constructing a time-dependent function and a comprehensive emission model according to the vehicle information and the time-varying road network information;

[0013] constructing a cost function based on the vehicle information, the integrated emission model, and the time-dependent function;

[0014] constructing a constraint condition based on the vehicle information and the time-dependent function;

[0015] An objective function is constructed based on the cost function.

[0016] In one embodiment, the step of obtaining a honey source set according to the constraint conditions, the time-varying road network information, and the vehicle information, and constructing a hybrid artificial bee colony strategy according to the honey source set includes:

[0017] Initialize the hybrid artificial bee colony strategy according to the vehicle information and preset parameters to obtain an initialized hybrid artificial bee colony strategy;

[0018] Obtaining a set of candidate customers within a preset distance of the current position of the vehicle according to the time-varying road network information;

[0019] Obtaining a visit record of the vehicle according to the vehicle information, and obtaining a non-visited customer set according to the visit record, the constraint condition, and the candidate customer set;

[0020] When the unvisited client set is not an empty set, obtaining a preset initialization mechanism setting value;

[0021] Determining a target initialization mechanism according to the preset initialization mechanism setting value;

[0022] Through the target initialization mechanism, the unvisited customer set is added to the honey source set of the initialized hybrid artificial bee colony strategy, and the hybrid artificial bee colony strategy is obtained according to the honey source set.

[0023] In one embodiment, the step of obtaining the target value of the objective function according to the constraint condition, the nectar source set and the hybrid artificial bee colony strategy includes:

[0024] Obtaining the number of exploitations, the preset number of exploitations, the number of iteration stagnations, the target number of iteration stagnations, the number of iterations, and the target number of iterations of the first difference nectar source;

[0025] According to the nectar source set, a fitness value is obtained, and according to the fitness value and the nectar source set, a first nectar source, a first difference nectar source and a second difference nectar source are obtained;

[0026] The employed bees based on the hybrid artificial bee colony strategy update the first nectar source according to the first differential nectar source, the roulette strategy and the preset mining strategy to obtain a second nectar source;

[0027] The follower bees based on the hybrid artificial bee colony strategy update the second nectar source according to the second difference nectar source, the difference elite individual strategy and the preset exploitation strategy to obtain a third nectar source, and update the number of exploitations;

[0028] When the updated number of mining times does not reach the preset number of mining times, returning to the step of updating the second nectar source according to the second difference nectar source, the difference elite individual strategy and the preset mining strategy by the follower bee based on the hybrid artificial bee colony strategy, obtaining a third nectar source, and updating the number of mining times;

[0029] When the number of mining times reaches the preset number of mining times, the number of iterative stagnation times is updated, and the third nectar source is updated by the scout bees of the hybrid artificial bee colony strategy to obtain a fourth nectar source;

[0030] When the updated number of iteration stagnation times is less than or equal to the target number of iteration stagnation times, returning to the step of updating the second nectar source according to the second difference nectar source, the difference elite individual strategy and the preset exploitation strategy by the follower bee based on the hybrid artificial bee colony strategy, obtaining a third nectar source, and updating the number of exploitation times, and updating the number of iterations;

[0031] When the updated iteration stagnation number is greater than the target iteration stagnation number, updating the fourth nectar source according to the first perturbation strategy or the second perturbation strategy to obtain a fifth nectar source;

[0032] When the updated number of iterations is greater than the target number of iterations, the fifth nectar source is used as the target value of the objective function.

[0033] In one embodiment, the hired bees based on the hybrid artificial bee colony strategy update the first nectar source according to the first differential nectar source, the roulette strategy and the preset mining strategy to obtain the second nectar source, comprising:

[0034] Obtaining the number of weight reset cycles, the number of target weight reset cycles, the preset annealing temperature, the removal operator score, the number of times the removal operator is used, the insertion operator score, and the number of times the insertion operator is used of the preset mining strategy;

[0035] The employed bees of the hybrid artificial bee colony strategy construct a first current solution according to the first differential nectar source set, and update the number of weight reset cycles;

[0036] Calculating a removal operator weight and an insertion operator weight according to the removal operator score, the number of times the removal operator is used, the insertion operator score, and the number of times the insertion operator is used;

[0037] According to the removal operator weight and the roulette strategy, a target removal operator is selected from the preset removal operators, and a destructive solution is obtained according to the target removal operator and the first current solution;

[0038] According to the insertion operator weight and the roulette strategy, a target insertion operator is selected from the preset insertion operators, and a repair solution is obtained according to the target insertion operator and the damage solution;

[0039] Obtaining a cost of the first current solution and a cost of the repair solution;

[0040] When the cost of the repair solution is greater than or equal to the cost of the first current solution, obtaining a preset inferior solution acceptance value, and obtaining an inferior solution acceptance probability according to the cost of the repair solution, the cost of the first current solution, and a preset annealing temperature;

[0041] When the inferior solution acceptance probability is greater than the preset inferior solution acceptance value, updating the first current solution according to the repair solution to obtain a second current solution;

[0042] When the cost of the repair solution is less than the cost of the first current solution, updating the first current solution according to the repair solution to obtain a second current solution;

[0043] When the cost of the second current solution is greater than or equal to the cost of the first honey source or the inferior solution acceptance probability is less than or equal to the preset inferior solution acceptance value, the removal operator score, the insertion operator score, the removal operator usage count, and the insertion operator usage count are updated according to the cost of the first honey source, the cost of the repair solution, the cost of the first current solution, and the cost of the second current solution;

[0044] When the updated weight reset cycle number is not equal to the target weight reset cycle number, the weight reset cycle number is updated, the second current solution is used as the first current solution, and the step of calculating the removal operator weight and the insertion operator weight according to the removal operator score, the number of times the removal operator is used, the insertion operator score, and the number of times the insertion operator is used is returned;

[0045] When the updated weight reset cycle number is equal to the target weight reset cycle number, the target removal operator and the target insertion operator are updated according to the updated removal operator score, the updated removal operator usage count, the updated insertion operator score, and the updated insertion operator usage count, and the target removal operator and the target insertion operator are replaced according to the updated result;

[0046] When the cost of the second current solution is less than the cost of the first honey source, the removal operator score, the insertion operator score, the number of times the removal operator is used, and the number of times the insertion operator is used are updated, and the first honey source is updated according to the second current solution to obtain a second honey source.

[0047] In one embodiment, the step of updating the removal operator score and the insertion operator score according to the cost of the first honey source, the cost of the repair solution, the cost of the first current solution, and the cost of the second current solution includes:

[0048] Obtaining a current score, a first score, a second score, and a third score of the preset removal operator and the preset insertion operator, wherein the third score is greater than the second score, and the second score is greater than the first score;

[0049] When the cost of the repair solution is equal to the cost of the first current solution, updating the removal operator score and the insertion operator score according to the first score and the current score;

[0050] When the cost of the second current solution is greater than or equal to the cost of the first honey source, updating the removal operator score and the insertion operator score according to the second score and the current score;

[0051] When the cost of the second current solution is less than the cost of the first honey source, the removal operator score and the insertion operator score are updated according to the third score and the current score.

[0052] In one embodiment, the step of updating the fourth nectar source according to the first disturbance strategy or the second disturbance strategy to obtain the fifth nectar source includes:

[0053] Acquire initial loading information of the vehicle in the fourth honey source, and acquire a first disturbance path corresponding to the initial loading information and a first customer position on the first disturbance path according to a first disturbance strategy;

[0054] Recording the first customer information on the first disturbance path, and clearing the first customer position;

[0055] placing the customers in the first customer information into the first disturbance path in sequence, recording the target loading information corresponding to the position of the first customer, and obtaining the first disturbance honey source according to the target loading information;

[0056] When the cost of the first disturbed nectar source is less than the cost of the fourth nectar source, updating the fourth nectar source according to the first disturbed nectar source to obtain a fifth nectar source;

[0057] When the cost of the first disturbance nectar source is greater than or equal to the cost of the fourth nectar source, obtaining the number of disturbances of the second disturbance strategy and the target number of disturbances of the second disturbance strategy;

[0058] Selecting a target vehicle from the fourth honey source, and obtaining a second disturbance path of the target vehicle;

[0059] Recording the second customer information on the second disturbance path, and clearing the second customer position on the second disturbance path;

[0060] Put the customers in the second customer information into the second disturbance path in sequence, record the second customer positions, obtain the second disturbance honey source, and update the disturbance times;

[0061] When the updated disturbance number is not equal to the target disturbance number and the cost of the second disturbance nectar source is greater than the cost of the fourth nectar source, returning to the step of selecting a target vehicle in the fourth nectar source and obtaining a second disturbance path of the target vehicle;

[0062] When the number of disturbances is equal to the target number of disturbances or the cost of the second disturbance nectar source is less than or equal to the cost of the fourth nectar source, the fourth nectar source is updated according to the second disturbance nectar source to obtain a fifth nectar source.

[0063] In addition, to achieve the above-mentioned purpose, the present application also proposes a time-dependent vehicle path planning device, the time-dependent vehicle path planning device comprising: a construction module, used to obtain vehicle information and time-varying road network information, and construct constraint conditions and objective functions according to the vehicle information and the time-varying road network information;

[0064] An obtaining module, used for obtaining a honey source set according to the constraint conditions, the time-varying road network information and the vehicle information, and constructing a hybrid artificial bee colony strategy according to the honey source set;

[0065] A solution module is used to obtain a target value of an objective function according to the constraint conditions, the nectar source set and the hybrid artificial bee colony strategy, and to plan a path for the vehicle according to the target value.

[0066] In addition, to achieve the above-mentioned purpose, the present application also proposes a time-dependent vehicle path planning device, which includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, and the computer program is configured to implement the steps of the time-dependent vehicle path planning method as described above.

[0067] In addition, to achieve the above-mentioned purpose, the present application also proposes a storage medium, which is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, the steps of the time-dependent vehicle path planning method described above are implemented.

[0068] In addition, to achieve the above-mentioned purpose, the present application also provides a computer program product, which includes a computer program, and when the computer program is executed by a processor, the steps of the time-dependent vehicle path planning method as described above are implemented.

[0069] One or more technical solutions proposed in this application have at least the following technical effects:

[0070] By acquiring real-time vehicle information and time-varying road network information, and building dynamic constraints and objective functions based on this information, the problem that traditional vehicle path planning methods cannot adapt to the time-varying nature of traffic conditions is solved. By using a hybrid artificial bee colony strategy, the solution space can be searched more intelligently to obtain a better objective function value, thereby efficiently planning the path for the vehicle. This not only improves distribution efficiency and reduces transportation costs, but also enhances the adaptability and accuracy of the planning scheme to real-time traffic changes, and realizes vehicle path planning in a complex and dynamic urban traffic environment. BRIEF DESCRIPTION OF THE DRAWINGS

[0071] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0072] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0073] Figure 1 A flow chart of the first embodiment of the time-dependent vehicle path planning method of the present application;

[0074] Figure 2 This is a TDVRPSPD problem description diagram in an embodiment of the time-dependent vehicle path planning method of this application;

[0075] Figure 3 A schematic flow chart of a hybrid artificial bee colony strategy provided in Example 1 of the time-dependent vehicle path planning method of the present application;

[0076] Figure 4 A schematic diagram of a process of adaptive large neighborhood search provided in Example 1 of the time-dependent vehicle path planning method of the present application;

[0077] Figure 5 A flow chart of the second embodiment of the time-dependent vehicle path planning method of the present application;

[0078] Figure 6 A schematic diagram of a time-dependent function provided for an embodiment of a time-dependent vehicle path planning method of the present application;

[0079] Figure 7 A schematic diagram of the effect provided by an embodiment of the time-dependent vehicle path planning method of the present application;

[0080] Figure 8A schematic diagram of an optimal delivery route provided for an embodiment of a time-dependent vehicle route planning method of the present application;

[0081] Fig. 9 This is a schematic diagram of the module structure of the time-dependent vehicle path planning device according to an embodiment of the present application;

[0082] Fig.10 Schematic diagram of the device structure of the hardware operating environment involved in the time-dependent vehicle path planning method in the embodiment of the present application.

[0083] The purpose, features and advantages of this application will be further described in conjunction with the embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0084] It should be understood that the specific embodiments described herein are only used to explain the technical solutions of the present application and are not used to limit the present application.

[0085] In order to better understand the technical solution of the present application, a detailed description will be given below in conjunction with the accompanying drawings and specific implementation methods.

[0086] The main solution of the embodiment of the present application is: obtaining vehicle information and time-varying road network information, and constructing constraint conditions and objective functions according to the vehicle information and the time-varying road network information;

[0087] Obtaining a honey source set according to the constraint conditions, the time-varying road network information and the vehicle information, and constructing a hybrid artificial bee colony strategy according to the honey source set;

[0088] The target value of the objective function is obtained according to the constraint conditions, the nectar source set and the hybrid artificial bee colony strategy, and the path of the vehicle is planned according to the target value.

[0089] In this embodiment, for ease of description, the following description is made by taking the identification of a time-dependent vehicle path planning device as the execution subject.

[0090] Due to the fact that most existing vehicle path planning studies in the prior art fail to fully consider the time-varying characteristics of urban traffic and the impact of vehicle speed changes on energy consumption, they are unable to accurately simulate the low delivery efficiency and increased transportation costs caused by traffic congestion and speed changes in the real world.

[0091] This application provides a solution that solves the problem that traditional vehicle path planning methods cannot adapt to the time-varying nature of traffic conditions by acquiring real-time vehicle information and time-varying road network information, and constructing dynamic constraints and objective functions based on this information. By hybridizing the artificial bee colony strategy, it is possible to search the solution space more intelligently and obtain a better objective function value, thereby performing efficient path planning for vehicles. This not only improves distribution efficiency and reduces transportation costs, but also enhances the adaptability and accuracy of the planning scheme to real-time traffic changes, and realizes vehicle path planning in a complex and dynamic urban traffic environment.

[0092] It should be noted that the execution subject of this embodiment can be a computing service device with data processing, network communication and program running functions, such as a tablet computer, a personal computer, a mobile phone, etc., or an electronic device capable of realizing the above functions, a time-dependent vehicle path planning device, etc. The following takes the time-dependent vehicle path planning device as an example to illustrate this embodiment and the following embodiments.

[0093] Based on this, the embodiment of the present application provides a time-dependent vehicle path planning method, referring to Figure 1 , Figure 1 This is a flow chart of the first embodiment of the time-dependent vehicle path planning method of the present application.

[0094] In this embodiment, the time-dependent vehicle path planning method includes steps S10 to S30:

[0095] Step S10, acquiring vehicle information and time-varying road network information, and constructing constraint conditions and objective functions according to the vehicle information and the time-varying road network information;

[0096] It should be noted that, refer to Figure 2 , Figure 2 This is the problem description diagram of the time-dependent vehicle path planning method TDVRPSPD for this application, as shown in Figure 2 As shown in Figure 1, the TDVRPSPD problem involves a distribution center, several vehicles, and a group of customers. Each customer has their own delivery and pickup requirements. The vehicle is responsible for delivering the goods from the distribution center to the customer and bringing the goods back to the distribution center after delivery. It is required to arrange the distribution path reasonably to minimize the overall cost. For example, the distribution center works in the time period of 7:00-12:00, and the customer point has a specific time window, such as 10:20-11:00, which means that the delivery vehicle must arrive at the customer point within this time period to complete the service. The TDVRPSPD problem assumes the following:

[0097] (1) The vehicle load cannot be greater than its maximum load;

[0098] (2) All vehicles depart from and return to a distribution center;

[0099] (3) Each vehicle can serve multiple customers, each customer is visited only once, and delivery and pickup can be completed at the same time;

[0100] (4) The customer's time window is a hard time window, which means that the vehicle must arrive within the time window. If the customer arrives early, he must wait, and if the customer arrives late, he will be rejected. The delivery vehicle can only visit the customer within the time window specified by the customer. If the customer arrives early, he can wait;

[0101] (5) The vehicle's speed will change over time, and the speed in each time period is known;

[0102] (6) When the vehicle is moving at a constant speed or a uniformly variable speed, the speed will not drop to 0 when the vehicle is moving at a uniformly decelerated speed;

[0103] (7) The vehicle takes fuel consumption and carbon emissions into consideration, but does not consider refueling.

[0104] It should be noted that vehicle information refers to a series of data and parameters directly related to the delivery vehicle, which are used to construct the subsequent objective function. In this embodiment, it includes but is not limited to the maximum load of the vehicle, the vehicle's own weight, the engine friction coefficient, the engine speed, the engine displacement, and the calorific value of the fuel.

[0105] In addition, it should be noted that time-varying road network information refers to various traffic-related data and parameters in the road network that change over time. In this embodiment, it includes but is not limited to vehicle and road network information such as vehicle speed, traffic flow, traffic accidents or construction information, road conditions, and real-time traffic congestion conditions in different time periods.

[0106] It should be noted that constraints are a series of restrictive regulations used to define the feasibility of solutions in optimization problems and mathematical modeling. In the vehicle path planning problem TDVRPSDP, constraints ensure that the generated path plan not only meets the requirements of actual operations, but also complies with specific business rules and regulatory standards. In this embodiment, the constraints of the TDVRPSDP optimization model include: vehicle number constraints, in-and-out balance constraints, path connection constraints, sub-loop elimination constraints, load balance constraints, time relationship constraints, and decision variable attribute constraints.

[0107] In addition, it should be noted that the objective function is a mathematical expression used to evaluate the quality of a solution in an optimization problem. The objective function defines the goal to be optimized, which can be to minimize or maximize a certain quantity. In this embodiment, the objective function is based on minimizing the sum of vehicle usage costs, fuel consumption, and carbon emission costs. In the process of minimizing the objective function, it is necessary to consider the time window, capacity, and delivery constraints of the delivery task.

[0108] It can be understood that the constraints and objective functions are constructed through vehicle information and time-varying road network information, and the objective function is minimized on the basis of the constraints, so as to find the optimal solution to the problem and then construct the vehicle's driving path.

[0109] Step S20, obtaining a honey source set according to the constraint conditions, the time-varying road network information and the vehicle information, and constructing a hybrid artificial bee colony strategy according to the honey source set;

[0110] It should be noted that the nectar source set represents a set of solutions considered at a certain moment. This concept comes from the Artificial Bee Colony (ABC) algorithm, which is an optimization algorithm that simulates the foraging behavior of bees in nature. The algorithm belongs to a type of swarm intelligence optimization algorithm. Bees show highly intelligent behavior in the process of searching for nectar. They can find nectar sources rich in nectar through information exchange. The artificial bee colony algorithm is inspired by this natural phenomenon and searches for the optimal solution to the problem by simulating the foraging mechanism of bees.

[0111] The honey source represents a feasible solution to the problem, and the quality of the honey source represents the fitness value of the solution. In this embodiment, the solution to the TDVRPSDP problem consists of the driving paths of all vehicles. This embodiment uses integer coding to represent the solution to the problem. For example, a honey source X = {0210376048590}, where the value 0 represents the distribution center and the non-zero number is the customer number served by the vehicle. The sequence between two values ​​0 represents the delivery path of a vehicle, and the delivery path of three vehicles constitutes a complete solution to the problem. It means that three vehicles serve 9 customers, and the route of vehicle 1 can be expressed as: {0210}; vehicle 2 is expressed as: {03760}; vehicle 3 is expressed as: {04890}. According to the coding rules, the dimension Dim of each solution is Dim = N+K+1, where N represents the number of customers and K represents the number of vehicles.

[0112] In addition, it should be noted that the hybrid artificial bee colony strategy is a method that combines the basic artificial bee colony algorithm (ABC) with other optimization techniques to improve the algorithm's search capability, accelerate convergence, and improve the quality of the solution.

[0113] It can be understood that the honey source set is used as the initial solution set of the algorithm, and then the hybrid artificial bee colony strategy is applied to conduct further optimization search. The hybrid artificial bee colony strategy combines the basic artificial bee colony algorithm and other optimization algorithms to improve the search efficiency and solution quality.

[0114] Step S30, obtaining a target value of an objective function according to the constraint conditions, the nectar source set and the hybrid artificial bee colony strategy, and performing path planning for the vehicle according to the target value.

[0115] It should be noted that the target value of the objective function represents the solution that makes the objective function value reach the optimal solution.

[0116] It can be understood that the nectar source set is iteratively searched by applying the hybrid artificial bee colony strategy based on the constraints to find a solution that optimizes the objective function value.

[0117] In a feasible implementation, step S20 may include steps S21 to S26:

[0118] Step S21, initializing the hybrid artificial bee colony strategy according to the vehicle information and preset parameters to obtain an initialized hybrid artificial bee colony strategy;

[0119] It should be noted that the preset parameters refer to the parameters for initializing the hybrid artificial bee colony strategy, which can be initialized with the vehicle information. In this embodiment, the initialization parameters include but are not limited to the distance matrix, the number of customers, the time window, and the vehicle load.

[0120] Step S22, obtaining a set of candidate customers within a preset distance of the current position of the vehicle according to the time-varying road network information;

[0121] It should be noted that the preset distance represents a set distance parameter used to determine the range that the vehicle can reach from the current position of the vehicle without considering changes in traffic conditions. In this embodiment, the preset distance represents the distance between the customer closest to the current position of the vehicle and the vehicle, and the candidate customer set is constructed by selecting customers within a certain range.

[0122] In addition, it should be noted that the candidate customer set refers to the set of all customers who may become the next service point of the vehicle within the preset distance of the vehicle. This set is determined based on the current location of the vehicle and the preset distance, but the specific selection of which customer to serve also needs to consider other factors, such as customer demand, service time window, road conditions, etc.

[0123] It is understandable that, based on the time-varying road network information and the current position of the vehicle, a set of all possible serviceable customers within a preset distance is obtained by calculation or query.

[0124] Step S23, obtaining a visit record of the vehicle according to the vehicle information, and obtaining a non-visited customer set according to the visit record, the constraint condition and the candidate customer set;

[0125] It should be noted that the visit record is a record of the customer points visited by the vehicle during the delivery task. The visit record usually contains information such as the customer's location, visit time, and the amount of goods delivered.

[0126] In addition, it should be noted that the unvisited customer set is a set of customers that have not been served, which is further screened from the candidate customer set and determined based on the vehicle's visit records and constraints. The unvisited customer set is the focus of the next step of path planning.

[0127] It can be understood that the customer set that meets the constraint and has not been visited is screened from the candidate customer set as the unvisited customer set.

[0128] Step S24, when the unvisited client set is not an empty set, obtaining a preset initialization mechanism setting value;

[0129] It should be noted that the preset initialization mechanism setting value represents a setting value for selecting an initialization mechanism. In this embodiment, a random number rand (0, 1) is used as the initialization mechanism setting value.

[0130] Step S25, determining a target initialization mechanism according to the preset initialization mechanism setting value;

[0131] It should be noted that the target initialization mechanism is an initialization strategy determined according to the preset initialization mechanism setting value, which defines how to generate the initial solution set of the algorithm, that is, the honey source set. In this embodiment, the target initialization mechanism includes two initialization mechanisms: the closest distance mechanism and the shortest travel time mechanism, and each mechanism sets an independent weight.

[0132] In addition, it is understandable that the selection of the next customer point is determined by the random number rand (0, 1), and a mechanism is randomly selected to construct the initial honey source. This makes the initial honey sources have certain differences, avoiding the defect of insufficient diversity of solutions generated by the traditional nearest neighbor method.

[0133] Step S26, adding the unvisited customer set to the honey source set of the initialized hybrid artificial bee colony strategy through the target initialization mechanism, and obtaining the hybrid artificial bee colony strategy according to the honey source set.

[0134] It can be understood that when rand(0,1)≤0.5, the shortest travel time mechanism is used to select the customer with the shortest travel time from the currently visited customer to join the honey source set. When rand(0,1)>0.5, the shortest distance mechanism is used to select the customer with the shortest distance from the currently visited customer to join the honey source set.

[0135] In addition, it can be understood that obtaining the hybrid artificial bee colony strategy according to the nectar source set means using the nectar source set as the basis of the hybrid artificial bee colony strategy.

[0136] In a feasible implementation, step S30 may include steps S31 to S39:

[0137] Step S31, obtaining the number of exploitations of the first difference honey source, the preset number of exploitations, the number of iteration stagnations, the target number of iteration stagnations, the number of iterations, and the target number of iterations;

[0138] It should be noted that the number of exploitation refers to the number of iterative updates of each nectar source, which is similar to the number of evaluations and improvements of each potential solution in natural selection. Correspondingly, the number of exploitation of the first difference nectar source represents the number of exploitation of the first difference nectar source, and the preset number of exploitation represents the number compared with the number of exploitation, representing the limit on the number of field searches.

[0139] In addition, it should be noted that the number of iteration stagnation refers to the number of iterations that are accumulated when the quality of the solution does not improve significantly in continuous iterations. Correspondingly, the target number of iteration stagnation is used to compare with the number of iteration stagnation to determine whether the algorithm falls into a local optimum.

[0140] It should be noted that the number of iterations refers to the total number of iterations of the algorithm from the beginning to now, and the target number of iterations refers to the upper limit of the number of iterations set by the algorithm, which is one of the stopping conditions of the algorithm.

[0141] Step S32, obtaining a fitness value according to the nectar source set, and obtaining a first nectar source, a first difference nectar source, and a second difference nectar source according to the fitness value and the nectar source set;

[0142] It should be noted that the fitness value is an indicator used to evaluate the quality of each honey source, that is, the potential solution, and is usually associated with the result of the objective function. In this embodiment, the fitness value is calculated as the inverse of the objective function value cost. Therefore, the better the individual, the smaller the objective function value, and the smaller the objective function value, the greater the individual's fitness value.

[0143] It can be understood that obtaining the first nectar source, the first difference nectar source and the second difference nectar source according to the fitness value and the nectar source set means that after calculating the fitness values ​​of the nectar sources in the nectar source set, they are sorted from high to low according to the fitness values, and the top 50% nectar sources in fitness value are taken as the first difference nectar sources; the nectar sources that are ranked in the top η% in fitness and have differences are taken as the second difference nectar sources, where η is the percentage coefficient of the number of follower bees; and the nectar source with the highest fitness value is taken as the first nectar source.

[0144] In addition, it should be noted that the first nectar source is the nectar source with the highest fitness value in the nectar source set before the iteration, which represents the optimal solution or the best path planning solution currently found.

[0145] It should be noted that the difference nectar source is a set of nectar sources ranked according to the fitness value after the nectar source set is sorted according to the fitness value. In this embodiment, the first difference nectar source is the top 50% nectar source in the fitness value ranking, and the second difference nectar source is the top 5% nectar source in the fitness value ranking.

[0146] Step S33, the employed bees based on the hybrid artificial bee colony strategy update the first nectar source according to the first differential nectar source, the roulette strategy and the preset mining strategy to obtain a second nectar source;

[0147] It should be noted that in the hybrid artificial bee colony strategy, the employed bees are responsible for searching within the neighborhood of the current optimal solution, that is, the nectar source, to find a better solution.

[0148] In addition, it should be noted that the roulette strategy is a selection strategy in evolutionary algorithms and optimization algorithms, which imitates the principle of roulette in gambling and is used to select individuals from the current population for crossover and mutation operations to generate the next generation population.

[0149] It can be understood that the process of the roulette strategy is: first calculate the fitness of each individual in the population, and calculate the sum of the fitness of all individuals in the population. Construct a "roulette wheel" in which the size of the area occupied by each individual is proportional to its fitness. The higher the fitness of the individual, the larger the area occupied on the roulette wheel. By generating a random number, determine the position of the pointer on the roulette wheel. The individual pointed by the pointer is selected for reproduction. Repeat the above process until a sufficient number of individuals are selected to generate the next generation.

[0150] It should be noted that the preset mining strategy represents adaptive large neighborhood search. Adaptive Large Neighborhood Search (ALNS) is a heuristic optimization algorithm that systematically explores large neighborhoods in the solution space to find the optimal or near-optimal solution to the problem. ALNS can effectively escape the local optimum and explore more solution spaces by dynamically selecting different removal and repair operators.

[0151] In addition, it should be noted that the second nectar source is the current optimal solution obtained by iterating on the basis of the first nectar source.

[0152] In this embodiment, updating the first honey source to obtain the second honey source means using a roulette wheel strategy to select a removal operator according to an adaptive weight update rule and an operator selection strategy to destroy the first honey source to obtain the second honey source.

[0153] Step S34, the follower bee based on the hybrid artificial bee colony strategy updates the second nectar source according to the second difference nectar source, the difference elite individual strategy and the preset exploitation strategy to obtain a third nectar source, and updates the exploitation times;

[0154] It should be noted that in the hybrid artificial bee colony strategy, the follower bees are responsible for searching within the neighborhood of the current optimal solution, that is, the nectar source, to find a better solution. Follower bees usually imitate the behavior of employed bees, but may adopt different search strategies.

[0155] In addition, it should be noted that the difference elite individual strategy is a selection strategy that emphasizes maintaining the diversity of solutions during the search process. By selecting elite individuals with differences, a wider solution space can be explored. Elite individuals are solutions with high fitness and large differences between each other.

[0156] It is understandable that the process of the differential elite individual strategy is to retain a certain number of "elite individuals", that is, those solutions with higher fitness or lower objective function values. By selecting elite individuals with differences, the algorithm can avoid the individuals in the population from converging too early, that is, avoiding premature convergence, which helps to obtain a wider solution space. In each iteration, the newly generated solutions are compared with the current elite individuals, and their fitness and differences are used to decide whether to be included in the elite group or replace part of the existing elite individuals.

[0157] In addition, it can be understood that the third nectar source refers to a new optimal solution obtained by iteratively updating the second nectar source.

[0158] Step S35, when the updated mining times do not reach the preset mining times, returning to the step of updating the second nectar source according to the second difference nectar source, the difference elite individual strategy and the preset mining strategy by the follower bee based on the hybrid artificial bee colony strategy, obtaining a third nectar source, and updating the mining times;

[0159] It can be understood that when the updated mining times do not reach the preset mining times, the third nectar source is taken as the new second nectar source to iteratively update the nectar source.

[0160] Step S36, when the number of mining times reaches the preset number of mining times, updating the number of iterative stagnation times, and updating the third nectar source by the scout bees of the hybrid artificial bee colony strategy to obtain a fourth nectar source;

[0161] It can be understood that when the number of mining times reaches the preset number of mining times, it means that the mining of the second difference nectar source is abandoned, and the scout bee will randomly select a nectar source from the remaining solutions for mining. The standard ABC algorithm will abandon the old solution and randomly generate a new solution in the scout bee stage. Compared with the iterated solution, the fitness of the new solution is low, and the chance of obtaining a neighborhood search later is smaller, which makes the algorithm more likely to stagnate. Therefore, this embodiment will not abandon the old solution in the scout bee stage, but randomly select a nectar source from the remaining solutions for exploration.

[0162] In addition, it can be understood that the fourth nectar source refers to a new optimal solution obtained by iteratively updating the third nectar source.

[0163] Step S37, when the updated number of iteration stagnation times is less than or equal to the target number of iteration stagnation times, returning to the step of updating the second nectar source according to the second difference nectar source, the difference elite individual strategy and the preset exploitation strategy by the follower bee based on the hybrid artificial bee colony strategy, obtaining a third nectar source, and updating the number of exploitation times, and updating the number of iterations;

[0164] Step S38, when the updated iteration stagnation number is greater than the target iteration stagnation number, updating the fourth nectar source according to the first perturbation strategy or the second perturbation strategy to obtain a fifth nectar source;

[0165] It can be understood that step S37 and step S38 respectively represent situations where the comparison results of the updated iteration stall number and the target iteration stall number are different.

[0166] It should be noted that the perturbation strategy is a method used in the optimization algorithm to break the stagnation of the algorithm, especially to jump out of the local optimal solution and explore the new solution space. The first perturbation strategy represents the minimum load rate vehicle path removal strategy, and the second perturbation strategy represents the random vehicle path removal strategy.

[0167] In addition, it can be understood that the fifth nectar source refers to a new optimal solution obtained by iteratively updating the fourth nectar source.

[0168] Step S39: When the updated number of iterations is greater than the target number of iterations, the fifth nectar source is used as the target value of the objective function.

[0169] It can be understood that when the updated number of iterations is greater than the target number of iterations, the current optimal nectar source, that is, the fifth nectar source, is used as the final solution, and the value of its objective function is used as the output result.

[0170] Reference Figure 3 , Figure 3 This is a flow chart of the hybrid artificial bee colony strategy for the time-dependent vehicle path planning method of this application, such as Figure 3 As shown:

[0171] After initializing the algorithm parameters, the improved nearest neighbor method is used to generate the initial nectar source set, and the fitness value of each nectar source is calculated to select the optimal nectar source. Then, the algorithm allows the employed bees to perform an adaptive large neighborhood search on the nectar sources ranked in the top 50% of the fitness value, that is, iterate according to the preset mining strategy to update the optimal nectar source. The follower bees select the nectar sources with differences ranked in the top 5% of the fitness value to further update the optimal nectar source. The hybrid artificial bee colony strategy also includes a decision-making process to determine whether the neighborhood search number limit or the mining number limit is reached. If the mining number does not reach the limit, the iteration continues; if the limit is reached or the number of consecutive iterations exceeds the threshold, the perturbation strategy is executed to update the optimal nectar source. Finally, if the number of consecutive iterations exceeds the maximum number of consecutive stagnations or the current number of iterations exceeds the maximum number of iterations, the optimal nectar source is output as the final result of the hybrid artificial bee colony strategy.

[0172] In a feasible implementation manner, step S33 may include steps S3311 to S3323:

[0173] Step S3311, obtaining the weight reset cycle number, target weight reset cycle number, preset annealing temperature, removal operator score, removal operator usage times, insertion operator score, and insertion operator usage times of the preset mining strategy;

[0174] It should be noted that the weight reset cycle number is the cycle number used to update the operator weight. Correspondingly, the target weight reset cycle number is a cycle number threshold, indicating that when the target weight reset cycle number is reached, the operator weight will be reset.

[0175] In addition, it should be noted that the preset annealing temperature is the initial temperature used to control the acceptance probability of inferior solutions in the simulated annealing algorithm.

[0176] It should be noted that the operator score refers to the cumulative performance score of the operator during the search process. Correspondingly, the removal operator score refers to the cumulative performance score of the removal operator during the search process, and the insertion operator score refers to the cumulative performance score of the insertion operator during the search process.

[0177] In addition, it should be noted that the operator usage count refers to the number of times the operator is used. Correspondingly, the removal operator usage count refers to the number of times the removal operator is used during the search process, and the insertion operator usage count refers to the number of times the insertion operator is used during the search process.

[0178] It is understandable that the operator score and the number of times the operator is used will affect the weight of the operator, and the weight of the operator will affect the probability of the operator being selected.

[0179] Step S3312, constructing a first current solution according to the first difference nectar source set by the employed bees of the hybrid artificial bee colony strategy, and updating the number of weight reset cycles;

[0180] It should be noted that the first current solution is the solution being improved in the current iteration process.

[0181] It can be understood that constructing the first current solution according to the first difference nectar source set means assigning the nectar sources ranked in the top 50% of fitness values, namely the first difference nectar source set, to the current solution, namely the first current solution.

[0182] In addition, it can be understood that updating the weight reset cycle number means updating the weight reset cycle number according to the calculation rule of the weight reset cycle count k=k+1.

[0183] Step S3313, calculating a removal operator weight and an insertion operator weight according to the removal operator score, the number of times the removal operator is used, the insertion operator score, and the number of times the insertion operator is used;

[0184] It can be understood that the initial weight of the operator is set to 0, and the weight is dynamically adjusted by the historical score of the operator. Removing the operator weight and inserting the operator weight can be considered as the calculation of the operator weight. The operator weight calculation formula is as follows:

[0185]

[0186] Where ω is the update coefficient, ranging from 0 to 1; θ orign and θ new are the initial weight and updated weight respectively, ε is the accumulated score of the operator, and δ is the number of times the operator is used.

[0187] Step S3314, selecting a target removal operator from the preset removal operators according to the removal operator weight and the roulette strategy, and obtaining a destructive solution according to the target removal operator and the first current solution;

[0188] It should be noted that the preset removal operator means preset different removal operators, and the removal operator will select certain elements in the current solution for removal. These elements can be customer points in the path planning problem, delivery tasks, or certain components in other problems. In this embodiment, the preset removal operators include:

[0189] Random removal: randomly remove Sr customers from the current solution and add them to the removal list Removelist.

[0190] Cost-saving removal: Calculate the cost savings of each customer, that is, calculate the cost difference after removing customers one by one. Select the first Sr customers with the largest cost difference after removal and add them to the removal list Removelist.

[0191] Remove the longest waiting time, calculate the waiting time of each customer and sort them, remove the first Sr customers with the longest waiting time and add them to the removal list Removelist.

[0192] Correlation removal. First, randomly select customer point c as the starting point. Then calculate its correlation with other customer points and sort them, remove the customer point with the largest correlation, and add it to the removal list Removelist. Then randomly select the starting point again in Removelist, and repeat this process until Sr customer points are removed.

[0193] Minimum Vehicle path removal, Defined as the cost of a subpath divided by the number of customers on that path, according to the present invention, a poor path has a higher value, select the minimum The vehicle paths are removed.

[0194] The maximum load increment is removed. The vehicle load is an important factor in the fuel consumption and carbon emission costs. The fuel consumption and carbon emission costs of heavy-loaded vehicles are high during driving. Define the maximum load increment Δm i =(d i -p i ), the customer with a larger Δm should be served as early as possible, and the customers with the maximum load increment Sr are removed and added to the removal list Removelist.

[0195] It can be understood that the correlation between the starting point c and other customer points in the correlation removal is as follows:

[0196] sim(k)=A*D(c,n(k))+B*(|d c -d n(k) |+|p c -p n(k) |)+C*(tw ce -tw n(k)e |+|tw cl -tw n(k)l |)+E*V

[0197] (Formula 2)

[0198] In the formula, A, B, C, E are correlation coefficients, D represents the distance matrix, n(k) (k = 1, 2, ... n) is the set of other customer points except the starting point, tw ce and tw cl represents the left time window and right time window of customer point c, tw n(k)e and tw n(k)lRepresents the left time window and right time window of other customer points. When the reference point and the removed node are serving the same vehicle, V is 0, otherwise it is 1. The smaller sim is, the greater the correlation is. The influencing factors include distance, pickup volume, delivery volume, time window, and whether it is the same vehicle.

[0199] It should be noted that after selecting the target removal operator from the preset removal operators, a removal operation is performed in the first current solution according to the target removal operator to generate a destructive solution, and further iterations are subsequently performed according to the destructive solution to find the optimal solution.

[0200] Step S3315, selecting a target insertion operator from preset insertion operators according to the insertion operator weight and the roulette strategy, and obtaining a repair solution according to the target insertion operator and the damage solution;

[0201] It should be noted that the preset insertion operator means preset different insertion operators, and the insertion operator will select one or more customer points or tasks that have not been scheduled and prepare to insert them into the existing path. In this embodiment, the preset insertion operator includes:

[0202] N-random insertion, calculate the first N feasible positions with the smallest incremental cost for the customer to be inserted, and then randomly select one of these positions as the insertion point for customer insertion.

[0203] Regret criterion insertion: pre-calculate the regret value of all customers to be inserted, that is, the sum of the difference between the optimal position increase cost and the c suboptimal position increase costs. Sort by regret value from large to small, and insert the customer with the largest regret value first.

[0204] Travel time - Greedy insertion, randomly select the customer to be inserted from the Removelist and insert it to the position where the travel time increases the least;

[0205] Target cost - Greedy insertion, randomly select customers in the Removelist and insert them to the position that minimizes the target cost.

[0206] In addition, it should be noted that after selecting the target insertion operator from the preset insertion operators, an insertion operation is performed in the damaged solution according to the target insertion operator to generate a repair solution, and further iterations will be performed based on this repair solution to find the optimal solution.

[0207] It can be understood that the roulette strategy of steps S3314 to S3315 selects removal operators and insertion operators. Each operator is adaptively selected according to different solution instances. In order to better distribute weights, a roulette strategy is required to select operators to be used. There are 6 removal operators and 5 insertion operators in total.

[0208] In addition, it can be understood that the operator selection probability calculation formula is:

[0209]

[0210] In the formula, the removal operator set Y = {y|1,2,3,4,…,k}, the insertion operator set C = {c|1,2,3,4,…,k}, where k is the number of operators; let θ(y i ) and θ(c i ) represent the weights of the i-th removal operator and the i-th insertion operator respectively.

[0211] Step S3316, obtaining the cost of the first current solution and the cost of the repair solution;

[0212] It can be understood that obtaining the cost of the first current solution and the cost of the repair solution means recording the objective function values ​​of the first current solution and the repair solution, which are usually related to the cost.

[0213] Step S3317, when the cost of the repair solution is greater than or equal to the cost of the first current solution, obtaining a preset inferior solution acceptance value, and obtaining an inferior solution acceptance probability according to the cost of the repair solution, the cost of the first current solution, and a preset annealing temperature;

[0214] It should be noted that the preset inferior solution acceptance value represents a set value for selecting whether to accept cracking. In this embodiment, a random number rand(0,1) is used as the preset inferior solution acceptance value.

[0215] In addition, it should be noted that the inferior solution acceptance probability is calculated based on the simulated annealing principle to determine whether to accept the inferior solution based on the inferior solution acceptance probability and the preset inferior solution acceptance value.

[0216] It can be understood that the calculation formula for the acceptance probability of the inferior solution is as follows:

[0217]

[0218] Where P is the acceptance probability of the inferior solution, dF = F(x′)-F(x), F(x′) is the target cost of the new solution x′, F(x) is the target cost of the original solution x, and t is the current temperature value.

[0219] In addition, it can be understood that in the process of calculating the acceptance probability of the inferior solution, the repair solution is taken as the new solution and the first current solution is taken as the original solution. When the cost of the repair solution is greater than or equal to the cost of the first current solution, it indicates that the repair solution is inferior to the first current solution.

[0220] Step S3318, when the inferior solution acceptance probability is greater than the preset inferior solution acceptance value, updating the first current solution according to the repair solution to obtain a second current solution;

[0221] It can be understood that when the acceptance probability of the inferior solution is greater than the preset inferior solution acceptance value, it means accepting the inferior solution, that is, when the repair solution is not as good as the first current solution, the repair solution is also accepted. Accepting the inferior solution can jump out of the current local optimal solution and explore new solution space areas, thereby making it possible to find the global optimal solution or a better local optimal solution.

[0222] Step S3319, when the cost of the repair solution is less than the cost of the first current solution, updating the first current solution according to the repair solution to obtain a second current solution;

[0223] It can be understood that when the cost of the repair solution is less than the cost of the first current solution, it means that the repair solution is better than the first current solution, and the repair solution is accepted at this time.

[0224] Step S3320, when the cost of the second current solution is greater than or equal to the cost of the first honey source or the inferior solution acceptance probability is less than or equal to the preset inferior solution acceptance value, update the removal operator score, the insertion operator score, the removal operator usage count, and the insertion operator usage count according to the cost of the first honey source, the cost of the repair solution, the cost of the first current solution, and the cost of the second current solution;

[0225] It can be understood that when the cost of the second current solution is greater than or equal to the cost of the first nectar source, it means that the updated solution is not as good as the current optimal solution when the cost of the updated solution is compared with the cost of the current optimal solution.

[0226] Step S3321, when the updated weight reset cycle number is not equal to the target weight reset cycle number, the weight reset cycle number is updated, the second current solution is used as the first current solution, and the step of calculating the removal operator weight and the insertion operator weight according to the removal operator score, the number of times the removal operator is used, the insertion operator score, and the number of times the insertion operator is used is returned;

[0227] Step S3322, when the updated weight reset cycle number is equal to the target weight reset cycle number, the target removal operator and the target insertion operator are updated according to the updated removal operator score, the updated removal operator usage count, the updated insertion operator score, and the updated insertion operator usage count, and the target removal operator and the target insertion operator are replaced according to the updated result;

[0228] Steps S3321 to S3322 indicate that when the updated weight reset cycle number is not equal to the target weight reset cycle number, continuous iterations will be performed.

[0229] It can be understood that replacing the target removal operator and the target insertion operator according to the updated result means recalculating the weights of the operators according to Formula 1.

[0230] Step S3323, when the cost of the second current solution is less than the cost of the first honey source, update the removal operator score, the insertion operator score, the number of times the removal operator is used, and the number of times the insertion operator is used, and update the first honey source according to the second current solution to obtain a second honey source.

[0231] It can be understood that updating the remove operator score, the insert operator score, the remove operator usage count, and the insert operator usage count represents updating the properties of the operators to be applicable to the next iteration.

[0232] In addition, it can be understood that updating the first honey source according to the second current solution to obtain the second honey source means that when the second current solution is better than the current optimal solution, namely the first honey source, the second current solution is used as the new optimal solution for subsequent iterations.

[0233] Reference Figure 4 , Figure 4 The flowchart of the adaptive large neighborhood search of the time-dependent vehicle path planning method of this application is as follows: Figure 4 As shown:

[0234] Initialize the current solution X cu The search process is controlled by updating the weight reset cycle count k. In each iteration, the roulette wheel strategy is first used to select the removal operator to destroy the current solution X. cu , generating the destructive solution X de Next, we use the roulette wheel strategy again to select the insertion operator to repair X de , get the repair solution X re , evaluate the new solution X re Is it better than the current solution X? cu If yes, update the current solution. If not, accept the inferior solution according to the Metropolis criterion. If the current solution is updated, evaluate the current solution X cu Is it better than the optimal solution X? gb At the same time, the score and usage count of the removal and insertion operators are recorded and updated, and the operator weights are updated adaptively. In addition, the algorithm also checks whether the current solution is the global optimal solution and updates it when appropriate. Finally, it determines whether the weight reset cycle K has been reached. If so, the operator weights are updated. Otherwise, the next round of iterations is continued until the stop condition is met, and the optimal solution X after the update iteration is finally output. gb .

[0235] In a feasible implementation manner, step S3320 may include steps A3321 to A3324:

[0236] Step A3321, obtaining the current score, the first score, the second score, and the third score of the preset removal operator and the preset insertion operator, wherein the third score is greater than the second score, and the second score is greater than the first score;

[0237] It should be noted that the current score is the score of the operator in the current iteration, and the first score, the second score, and the third score are preset scores used to update the operator score in different situations. In this embodiment, the first score is 1 point, the second score is 3 points, and the third score is 6 points.

[0238] Step A3322: when the cost of the repair solution is equal to the cost of the first current solution, update the removal operator score and the insertion operator score according to the first score and the current score;

[0239] It can be understood that updating the removal operator score according to the first score and the current score means adding the score corresponding to the first score to the input removal operator score and the insertion operator score, and the operator score will be updated by accumulation operation during the iteration process.

[0240] Step A3323: when the cost of the second current solution is greater than or equal to the cost of the first honey source, updating the removal operator score and the insertion operator score according to the second score and the current score;

[0241] It can be understood that updating the removal operator score and the insertion operator score according to the second score and the current score means adding the score corresponding to the second score to the input removal operator score and the insertion operator score.

[0242] Step A3324: when the cost of the second current solution is less than the cost of the first honey source, update the removal operator score and the insertion operator score according to the third score and the current score.

[0243] It can be understood that updating the removal operator score and the insertion operator score according to the third score and the current score means adding the score corresponding to the third score to the input removal operator score and the insertion operator score.

[0244] In a feasible implementation, step S38 may include steps S3811 to S3820:

[0245] Step S3811, obtaining initial loading information of the vehicle in the fourth honey source, and obtaining a first disturbance path corresponding to the initial loading information and a first customer position on the first disturbance path according to a first disturbance strategy;

[0246] It should be noted that the initial loading information refers to the information of the goods loaded on the vehicle at the beginning of the delivery route.

[0247] It can be understood that obtaining the first disturbance path corresponding to the initial loading information and the first customer position on the first disturbance path according to the first disturbance strategy means obtaining the path corresponding to the initial loading information and the customer information corresponding to the path.

[0248] Step S3812, recording the first customer information on the first disturbance path, and clearing the first customer position;

[0249] It can be understood that recording the first customer information on the first disturbance path and clearing the first customer position means selecting all customers in the vehicle path with the lowest loading rate for removal.

[0250] Step S3813, placing the customers in the first customer information into the first disturbance path in sequence, recording the target loading information corresponding to the position of the first customer, and obtaining the first disturbance honey source according to the target loading information;

[0251] Step S3814: when the cost of the first disturbed nectar source is less than the cost of the fourth nectar source, the fourth nectar source is updated according to the first disturbed nectar source to obtain a fifth nectar source;

[0252] It should be noted that the first perturbation nectar source is the output obtained after the first perturbation strategy, and will be compared based on the cost of the first perturbation nectar source and the cost of the current optimal solution, that is, the cost of the fourth nectar source.

[0253] It can be understood that step S3813 to step S3814 represent greedy repair, which is a search strategy usually used to improve the solution in the optimization problem. The core idea of ​​this strategy is to take the currently optimal decision in each selection step, in the hope of reaching the global optimal or approximate optimal solution through continuous local optimal selection.

[0254] The process of greedy repair is: put the customer loop that needs to be inserted into all positions in the path in turn, record the position that makes the objective function change best, and then insert the nodes in turn into the "best" position calculated by greedy insertion.

[0255] Step S3815, when the cost of the first disturbance nectar source is greater than or equal to the cost of the fourth nectar source, obtaining the disturbance times of the second disturbance strategy and the target disturbance times of the second disturbance strategy;

[0256] It can be understood that when the output of the first disturbance strategy, that is, the first disturbance nectar source, does not improve on the current optimal solution, the second disturbance strategy is entered.

[0257] Step S3816, selecting a target vehicle from the fourth honey source, and obtaining a second disturbance path of the target vehicle;

[0258] It should be noted that a vehicle is randomly selected as the target vehicle in the current optimal solution, and the transportation path of the target vehicle is obtained.

[0259] Step S3817, recording the second customer information on the second disturbance path, and clearing the second customer position on the second disturbance path;

[0260] It can be understood that recording the second customer information on the second disturbance path and clearing the second customer position on the second disturbance path means removing all customers in the vehicle path.

[0261] Step S3818, sequentially placing the customers in the second customer information into the second disturbance path, recording the second customer positions, obtaining the second disturbance honey source, and updating the disturbance times;

[0262] Step S3819, when the updated disturbance number is not equal to the target disturbance number and the cost of the second disturbance source is greater than the cost of the fourth source, return to the step of selecting a target vehicle in the fourth source and obtaining a second disturbance path of the target vehicle;

[0263] Step S3820: When the number of disturbances is equal to the target number of disturbances or the cost of the second disturbance nectar source is less than or equal to the cost of the fourth nectar source, the fourth nectar source is updated according to the second disturbance nectar source to obtain a fifth nectar source.

[0264] Steps S3818 to S3820 are to perform greedy repair on the path after removing all customers, and obtain the result of greedy repair, that is, the second perturbation nectar source. If the second perturbation nectar source solution is worse than the fourth nectar source, that is, the current global optimal solution, the second perturbation nectar source is used as the current optimal solution to update the fourth nectar source. If the second perturbation nectar source solution is not as good as or equal to the fourth nectar source, the number of perturbations is accumulated, and the perturbation method is continued until the target number of perturbations is reached. When the target number of perturbations is reached, the fourth nectar source is updated according to the current second perturbation nectar source.

[0265] Based on the first embodiment of the present application, in the second embodiment of the present application, the same or similar contents as those in the above-mentioned embodiment 1 can be referred to the above introduction, and will not be repeated in the following. Figure 5After step S10, the time-dependent vehicle path planning method further includes steps S11 to S15:

[0266] Step S11, obtaining vehicle information and time-varying road network information;

[0267] Step S12, constructing a time-dependent function and a comprehensive emission model according to the vehicle information and the time-varying road network information;

[0268] It can be understood that steps S11 to S12 represent building a time-dependent function and a comprehensive emission model based on the acquired vehicle information and time-varying road network information after obtaining the vehicle information and time-varying road network information. In the subsequent description, i represents the current customer and j represents the next customer.

[0269] In addition, it can be understood that the formula of the comprehensive emission model is as follows:

[0270]

[0271] Where λ is the calorific value of the fuel, is the driving distance of vehicle k from node i to node j in the tth time period, E f is the engine friction coefficient, E v is the engine speed, E d is the engine displacement, is the speed of vehicle k from node i to node j in the tth time period, γ is the powertrain coefficient, g is the gravity constant, m0 is the vehicle weight, y ijk is the load of vehicle k from customer i to customer j, and β is the aerodynamic drag coefficient.

[0272] It should be noted that the time-dependent function consists of the steps of calculating the driving distance, determining the time period, and updating the driving distance.

[0273] It can be understood that if the acceleration a of the current customer i in time period t t =0, the calculation formula of the driving distance is as follows:

[0274] a j =b i +d ij / V t (Formula 6)

[0275] In the formula, a j is the time when the vehicle arrives at node j, b i is the time when the vehicle leaves node i, V t is the speed at each time period when the speed is uniform, d ij is the distance between customer i and customer j.

[0276] If the acceleration a of the current customer i in time period t t When ≠0, the calculation formula of the driving distance is as follows:

[0277] a j =b i +(-2V+(4V 2 +8a t d ij ) 1 / 2 / 2a t ), V = V t-1 +a t (b i -T t-1 )(Formula 7)

[0278] In the formula, a j is the time when the vehicle arrives at node j, b i is the time when the vehicle leaves node i, V t is the speed at each time period when the speed is uniform, d ij is the distance between customer i and customer j.

[0279] It can be understood that, if the time period is determined, and the time when the vehicle leaves node j and the time when the vehicle leaves node i are in the same time period m, the arrival time a is output. j , otherwise, update the driving distance.

[0280] The steps of updating the driving distance include: if the acceleration a in the previous time period t-1 =0, then press d=d ij -V t (T t -b i ) Update the distance and proceed to the step of calculating the driving distance; otherwise, press d = d ij -(V(T t -b i )+1 / 2a t (T t -b i ) 2 ) is updated and the step of calculating the driving distance is entered.

[0281] Reference Figure 6 , Figure 6 A schematic diagram of a time-dependent function is provided for an embodiment of a time-dependent vehicle path planning method of the present application, such as Figure 6 As shown:

[0282] The continuous changes in road speed during the day are approximated by a continuous linear piecewise function, thereby constructing a speed time-dependent function. In different time periods of the day, from T1 to T9, there are different traffic conditions. The horizontal axis is time, and the vertical axis represents the vehicle's speed. The figure shows the periodic changes of the time-dependent function within a day, such as traffic congestion during peak hours in the morning and evening and smooth traffic during non-peak hours.

[0283] Step S13, constructing a cost function according to the vehicle information, the comprehensive emission model and the time-dependent function;

[0284] It should be noted that the cost function is composed of the vehicle use cost, the vehicle fuel consumption cost and the vehicle carbon emission cost, among which the vehicle use cost is composed of the vehicle rental cost and the vehicle use time cost.

[0285] It can be understood that the calculation formula of the vehicle use cost is as follows:

[0286]

[0287] In the formula, C F is the vehicle rental cost, C T is the unit time cost of the vehicle, If vehicle k passes arc (i, j) in time period t, it is 1, otherwise it is 0. ijk represents the load of vehicle k from customer i to customer j, wt i represents the waiting time of node i, wt i =max{(e i -a ik ),0},y ik If vehicle k serves i, it is 1, otherwise it is 0. i Indicates customer service hours.

[0288] It can be understood that the calculation formula of the vehicle's fuel consumption cost is as follows 9:

[0289]

[0290] In the formula, C R is the unit fuel consumption cost, is a comprehensive emission model, It means that if vehicle k passes through arc (i, j) in time period t, it is 1, otherwise it is 0.

[0291] It can be understood that the calculation formula of the carbon emission cost of a vehicle is as follows:

[0292]

[0293] In the formula, CP is the unit carbon emission cost, is a comprehensive emission model, It means that if vehicle k passes through arc (i, j) in time period t, it is 1, otherwise it is 0.

[0294] Step S14, constructing constraint conditions according to the vehicle information and the time-dependent function;

[0295] It can be understood that the vehicle number constraint is to ensure that the number of vehicles in use does not exceed the maximum number of vehicles allowed. The formula for the vehicle number constraint is as follows:

[0296]

[0297] In the formula, It means that if vehicle k passes through arc (i, j) in time period t, it is 1, otherwise it is 0, and m represents the number of vehicles.

[0298] It can be understood that the path connection constraint ensures that each customer is visited only once. The formula of the path connection constraint is as follows:

[0299]

[0300] In the formula, It means that if vehicle k passes through arc (i, j) in time period t, it is 1, otherwise it is 0, and n represents the number of customers.

[0301] It can be understood that the formula for eliminating the sub-loop constraint is as follows:

[0302]

[0303] In the formula, It means that if vehicle k passes through arc (i, j) in time period t, it is 1, otherwise it is 0, and m represents the number of vehicles.

[0304] It can be understood that the in-out balance constraint is the flow balance constraint, which ensures that the same vehicle arrives and leaves the customer it serves. When i = 0, it means that the vehicle starts from the distribution center and eventually returns to the distribution center. The formula for the in-out balance constraint is as follows:

[0305]

[0306] In the formula, It means that if vehicle k passes through arc (i, j) in time period t, it is 1, otherwise it is 0, and m represents the number of vehicles.

[0307] It can be understood that there are three types of load balancing constraints. The first load balancing constraint ensures that when the vehicle leaves the distribution center, the goods carried by the vehicle can meet the delivery needs of all customers to be served and will not be overloaded during the distribution process. The formula of the first load balancing constraint is as follows:

[0308]

[0309] In the formula, y ik If vehicle k serves i, it is 1, otherwise it is 0. It means that if vehicle k passes through arc (i, j) in time period t, it is 1, otherwise it is 0, m represents the number of vehicles, and Q represents the maximum load of the vehicle.

[0310] The second load balancing constraint ensures that when the vehicle returns to the distribution center, the cargo it carries is equal to the total amount of cargo picked up by the visited customers and will not be overloaded during the entire distribution process. The formula for the second load balancing constraint is as follows:

[0311]

[0312] In the formula, y ik If vehicle k serves i, it is 1, otherwise it is 0. It means that vehicle k passes through arc (i, j) in time period t, which is 1, otherwise it is 0, m is the number of vehicles, Q is the maximum load of the vehicle, and p is i Indicates the customer's pickup quantity.

[0313] The third load balancing constraint is that the change in the customer load served is equal to the difference between the demand and the pickup volume. The formula for the third load balancing constraint is as follows:

[0314] Formula 17:

[0315]

[0316] In the formula, y ik If vehicle k serves i, it is 1, otherwise it is 0. It means that vehicle k passes arc (i, j) in time period t, which is 1, otherwise it is 0, m represents the number of vehicles, p i represents the customer's pickup quantity, d j Indicates customer demand.

[0317] It can be understood that there are three types of time relationship constraints. The formula of the first time constraint is as follows:

[0318]

[0319] In the formula, a ik represents the arrival time of vehicle k to node i, then T 0 =e0,T w= l0, [e i ,l i ] represents the customer time window, [T t-1 ,T t ] represents the start and end time of the tth time period, then T 0 =e0,T w =l0, w represents the number of time periods that divide the distribution center working time [e0,l0] into equal parts, s i represents the customer service time, t ij represents the travel time of a vehicle from customer i to customer j, It means that if vehicle k passes through arc (i, j) in time period t, it is 1, otherwise it is 0, and m represents the number of vehicles.

[0320] The second time constraint represents the relationship between the arrival time and departure time of the vehicle to customer i. The formula of the second time constraint is as follows:

[0321]

[0322] In the formula, represents the time when vehicle k leaves node i in time period t, a ik represents the arrival time of vehicle k to node i, then T 0 =e0,T w =l0,wt i represents the waiting time of node i, wt i =max{(e i -a ik ),0},s i Indicates customer service hours, It means that if vehicle k passes through arc (i, j) in time period t, it is 1, otherwise it is 0, and m represents the number of vehicles.

[0323] The third time constraint ensures that the departure time is within the time period. The formulas of the third time constraint are as follows:

[0324]

[0325] In the formula, represents the time when vehicle k leaves node i in time period t, T represents the specific time, It means that if vehicle k passes through arc (i, j) in time period t, it is 1, otherwise it is 0, m represents the number of vehicles, and w represents the number of time periods that divide the distribution center working time [e0, l0] into equal parts.

[0326]

[0327] In the formula, represents the time when vehicle k leaves node i in time period t, T represents the specific time, It means that if vehicle k passes through arc (i, j) in time period t, it is 1, otherwise it is 0, m represents the number of vehicles, and w represents the number of time periods that divide the distribution center working time [e0, l0] into equal parts.

[0328] It can be understood that the decision variable attribute constraints are as follows:

[0329]

[0330] In the formula, It means that if vehicle k passes through arc (i, j) in time period t, it is 1, otherwise it is 0, m represents the number of vehicles, and w represents the number of time periods that divide the distribution center working time [e0, l0] into equal parts.

[0331] y ijk ≥0,i=1,2,...,n; j=1,2,...,n; k=1,2,...,m; t=1,2,...,w (Formula 23)

[0332] In the formula, y ik It is 1 if vehicle k serves i, otherwise it is 0, m is the number of vehicles, and w is the number of time periods that divide the distribution center working time [e0,l0] into equal parts.

[0333] Step S15, constructing an objective function according to the cost function.

[0334] It can be understood that the formula of the objective function is as follows:

[0335] minC=C1+C2+C3(Formula 24)

[0336] Where C1 is the vehicle use cost, C2 is the vehicle fuel consumption cost, C3 is the vehicle carbon emission cost, and minC indicates that the objective of the objective function is to minimize the sum of the vehicle use cost, the vehicle fuel consumption cost, and the vehicle carbon emission cost.

[0337] Reference Figure 7 , Figure 7 This is a schematic diagram of the effect provided by an embodiment of the time-dependent vehicle path planning method of the present application, such as Figure 7 As shown:

[0338] Iteration curves of the standard artificial bee colony algorithm (ABC), the hybrid artificial bee colony algorithm combined with large neighborhood search (ABC-LNS), and the HABC algorithm of the present invention are compared. As can be seen from the figure, the HABC algorithm has the fastest convergence speed, reaching convergence in the early stage of the iteration process, and performs well. It is closely followed by the ABC-LNS algorithm, which exhibits better convergence performance than the standard ABC algorithm due to the use of random removal operators and cost-greedy insertion operators. Although the ABC algorithm has a faster iteration speed in the early stage, it quickly falls into a local optimum and stops improving before 40 iterations. This shows that the use of a large neighborhood search operator, as embodied in ABC-LNS, can effectively improve the global search capability of the algorithm, and the HABC algorithm performs best among these algorithms. The X-axis on the chart represents the number of iterations, and the Y-axis represents the objective function value or cost. As the number of iterations increases, the objective function value of each algorithm shows a downward trend, but the HABC algorithm decreases the fastest and reaches the lowest cost first.

[0339] Reference Figure 8 , Figure 8 The optimal delivery path diagram provided by the first embodiment of the time-dependent vehicle path planning method of the present application is as follows: Figure 8 As shown:

[0340] The order of the delivery vehicles’ routes when serving 100 customers. The numbers in the figure represent the customer numbers, and the color gradient indicates the priority of delivery. The optimal delivery route can be obtained based on the colors and connecting lines.

[0341] It should be noted that the above examples are only used to understand the present application and do not constitute a limitation on the time-dependent vehicle path planning method of the present application. More simple transformations based on this technical concept are all within the scope of protection of the present application.

[0342] This application also provides a time-dependent vehicle path planning device, please refer to Fig. 9 , the time-dependent vehicle path planning device comprises:

[0343] The construction module 10 is used to obtain vehicle information and time-varying road network information, and to construct constraint conditions and objective functions according to the vehicle information and the time-varying road network information.

[0344] The obtaining module 20 is used to obtain a honey source set according to the constraint conditions, the time-varying road network information and the vehicle information, and to construct a hybrid artificial bee colony strategy according to the honey source set.

[0345] The solution module 30 is used to obtain the target value of the objective function according to the constraint conditions, the nectar source set and the hybrid artificial bee colony strategy, and plan the path of the vehicle according to the target value.

[0346] The time-dependent vehicle path planning device provided by the present application adopts the time-dependent vehicle path planning method in the above embodiment, which can solve the technical problem that most existing vehicle path planning studies fail to fully consider the time-varying characteristics of urban traffic and the impact of vehicle speed changes on energy consumption, and cannot accurately simulate the low distribution efficiency and increased transportation costs caused by traffic congestion and speed changes in the real world. Compared with the prior art, the beneficial effects of the time-dependent vehicle path planning device provided by the present application are the same as the beneficial effects of the time-dependent vehicle path planning method provided by the above embodiment, and the other technical features in the time-dependent vehicle path planning device are the same as the features disclosed in the above embodiment method, which will not be repeated here.

[0347] The present application provides a time-dependent vehicle path planning device, which includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the time-dependent vehicle path planning method in the above-mentioned embodiment one.

[0348] Reference below Fig.10 , which shows a schematic diagram of the structure of a time-dependent vehicle path planning device suitable for implementing the embodiment of the present application. The time-dependent vehicle path planning device in the embodiment of the present application may include but is not limited to mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Descriptions), PMPs (Portable Media Players), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Fig.10 The time-dependent vehicle path planning device shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0349] like Fig.10As shown, the time-dependent vehicle path planning device may include a processing device 1001 (e.g., a central processing unit, a graphics processing unit, etc.), which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM: Read Only Memory) 1002 or a program loaded from a storage device 1003 to a random access memory (RAM: Random Access Memory) 1004. In RAM1004, various programs and data required for the operation of the time-dependent vehicle path planning device are also stored. The processing device 1001, ROM1002, and RAM1004 are connected to each other via a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Typically, the following systems can be connected to the I / O interface 1006: input devices 1007 including, for example, a touch screen, a touchpad, a keyboard, a mouse, an image sensor, a microphone, an accelerometer, a gyroscope, etc.; output devices 1008 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; storage devices 1003 including, for example, a magnetic tape, a hard disk, etc.; and communication devices 1009. The communication device 1009 can allow the time-dependent vehicle path planning device to communicate wirelessly or wired with other devices to exchange data. Although the figure shows a time-dependent vehicle path planning device with various systems, it should be understood that it is not required to implement or have all the systems shown. More or fewer systems may be implemented or have instead.

[0350] In particular, according to the embodiments disclosed in the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the embodiments disclosed in the present application include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through a communication device, or installed from a storage device 1003, or installed from a ROM 1002. When the computer program is executed by the processing device 1001, the above-mentioned functions defined in the method of the embodiment disclosed in the present application are executed.

[0351] The time-dependent vehicle path planning device provided by the present application adopts the time-dependent vehicle path planning method in the above embodiment, which can solve the technical problem that most existing vehicle path planning studies fail to fully consider the time-varying characteristics of urban traffic and the impact of vehicle speed changes on energy consumption, and cannot accurately simulate the low distribution efficiency and increased transportation costs caused by traffic congestion and speed changes in the real world. Compared with the prior art, the beneficial effects of the time-dependent vehicle path planning device provided by the present application are the same as the beneficial effects of the time-dependent vehicle path planning method provided by the above embodiment, and the other technical features in the time-dependent vehicle path planning device are the same as the features disclosed in the method of the previous embodiment, which will not be repeated here.

[0352] It should be understood that the various parts disclosed in this application can be implemented by hardware, software, firmware or a combination thereof. In the description of the above embodiments, specific features, structures, materials or characteristics can be combined in any one or more embodiments or examples in a suitable manner.

[0353] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0354] The present application provides a computer-readable storage medium having computer-readable program instructions (ie, computer programs) stored thereon, wherein the computer-readable program instructions are used to execute the time-dependent vehicle path planning method in the above-mentioned embodiment.

[0355] The computer-readable storage medium provided in the present application may be, for example, a USB flash drive, but is not limited to electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, systems or devices, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, 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 above. In this embodiment, the computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, system or device. The program code contained on the computer-readable storage medium may be transmitted using any appropriate medium, including but not limited to: wires, optical cables, RF (Radio Frequency), etc., or any suitable combination of the above.

[0356] The computer-readable storage medium may be included in the time-dependent vehicle path planning device; or may exist independently without being assembled into the time-dependent vehicle path planning device.

[0357] The computer-readable storage medium carries one or more programs. When the one or more programs are executed by the time-dependent vehicle path planning device, the time-dependent vehicle path planning device: obtains vehicle information and time-varying road network information, and constructs constraint conditions and objective functions according to the vehicle information and the time-varying road network information;

[0358] Obtaining a honey source set according to the constraint conditions, the time-varying road network information and the vehicle information, and constructing a hybrid artificial bee colony strategy according to the honey source set;

[0359] The target value of the objective function is obtained according to the constraint conditions, the nectar source set and the hybrid artificial bee colony strategy, and the path of the vehicle is planned according to the target value.

[0360] Computer program code for performing the operations of the present application may be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0361] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present application. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0362] The modules involved in the embodiments described in this application may be implemented by software or hardware, wherein the name of the module does not constitute a limitation on the unit itself in some cases.

[0363] The readable storage medium provided by the present application is a computer-readable storage medium, which stores computer-readable program instructions (i.e., computer programs) for executing the above-mentioned time-dependent vehicle path planning method, and can solve the technical problems that most existing vehicle path planning studies fail to fully consider the time-varying characteristics of urban traffic and the impact of vehicle speed changes on energy consumption, and cannot accurately simulate the low distribution efficiency and increased transportation costs caused by traffic congestion and speed changes in the real world. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided by the present application are the same as the beneficial effects of the time-dependent vehicle path planning method provided by the above-mentioned embodiment, and will not be repeated here.

[0364] The present application also provides a computer program product, including a computer program, which implements the steps of the time-dependent vehicle path planning method as described above when executed by a processor.

[0365] The computer program product provided by this application can solve the technical problem that most existing vehicle path planning studies fail to fully consider the time-varying characteristics of urban traffic and the impact of vehicle speed changes on energy consumption, and fail to accurately simulate the low distribution efficiency and increased transportation costs caused by traffic congestion and speed changes in the real world. Compared with the prior art, the beneficial effects of the computer program product provided by this application are the same as the beneficial effects of the time-dependent vehicle path planning method provided by the above embodiment, and will not be repeated here.

[0366] The above descriptions are only some embodiments of the present application, and are not intended to limit the patent scope of the present application. All equivalent structural changes made using the contents of the present application specification and drawings under the technical concept of the present application, or direct / indirect applications in other related technical fields are included in the patent protection scope of the present application.

Claims

1. A time-dependent vehicle path planning method, characterized in that: The method comprises: Acquire vehicle information and time-varying road network information, and construct constraint conditions and objective functions according to the vehicle information and the time-varying road network information; Obtaining a honey source set according to the constraint conditions, the time-varying road network information and the vehicle information, and constructing a hybrid artificial bee colony strategy according to the honey source set; The target value of the objective function is obtained according to the constraint conditions, the nectar source set and the hybrid artificial bee colony strategy, and the path of the vehicle is planned according to the target value.

2. The method according to claim 1, characterized in that The steps of acquiring vehicle information and time-varying road network information, and constructing constraint conditions and objective functions according to the vehicle information and the time-varying road network information include: Obtain vehicle information and time-varying road network information; Constructing a time-dependent function and a comprehensive emission model according to the vehicle information and the time-varying road network information; constructing a cost function based on the vehicle information, the integrated emission model, and the time-dependent function; constructing a constraint condition based on the vehicle information and the time-dependent function; An objective function is constructed based on the cost function.

3. The method according to claim 1, characterized in that The step of obtaining a honey source set according to the constraint conditions, the time-varying road network information and the vehicle information, and constructing a hybrid artificial bee colony strategy according to the honey source set includes: Initialize the hybrid artificial bee colony strategy according to the vehicle information and preset parameters to obtain an initialized hybrid artificial bee colony strategy; Obtaining a set of candidate customers within a preset distance of the current position of the vehicle according to the time-varying road network information; Obtaining a visit record of the vehicle according to the vehicle information, and obtaining a non-visited customer set according to the visit record, the constraint condition, and the candidate customer set; When the unvisited client set is not an empty set, obtaining a preset initialization mechanism setting value; Determining a target initialization mechanism according to the preset initialization mechanism setting value; Through the target initialization mechanism, the unvisited customer set is added to the honey source set of the initialized hybrid artificial bee colony strategy, and the hybrid artificial bee colony strategy is obtained according to the honey source set.

4. The method according to claim 1, characterized in that The step of obtaining the target value of the objective function according to the constraint conditions, the nectar source set and the hybrid artificial bee colony strategy includes: Obtaining the number of exploitations, the preset number of exploitations, the number of iteration stagnations, the target number of iteration stagnations, the number of iterations, and the target number of iterations of the first difference nectar source; According to the nectar source set, a fitness value is obtained, and according to the fitness value and the nectar source set, a first nectar source, a first difference nectar source and a second difference nectar source are obtained; The employed bees based on the hybrid artificial bee colony strategy update the first nectar source according to the first differential nectar source, the roulette strategy and the preset mining strategy to obtain a second nectar source; The follower bees based on the hybrid artificial bee colony strategy update the second nectar source according to the second difference nectar source, the difference elite individual strategy and the preset exploitation strategy to obtain a third nectar source, and update the number of exploitations; When the updated number of mining times does not reach the preset number of mining times, returning to the step of updating the second nectar source according to the second difference nectar source, the difference elite individual strategy and the preset mining strategy by the follower bee based on the hybrid artificial bee colony strategy, obtaining a third nectar source, and updating the number of mining times; When the number of mining times reaches the preset number of mining times, the number of iterative stagnation times is updated, and the third nectar source is updated by the scout bees of the hybrid artificial bee colony strategy to obtain a fourth nectar source; When the updated number of iteration stagnation times is less than or equal to the target number of iteration stagnation times, returning to the step of updating the second nectar source according to the second difference nectar source, the difference elite individual strategy and the preset exploitation strategy by the follower bee based on the hybrid artificial bee colony strategy, obtaining a third nectar source, and updating the number of exploitation times, and updating the number of iterations; When the updated iteration stagnation number is greater than the target iteration stagnation number, updating the fourth nectar source according to the first perturbation strategy or the second perturbation strategy to obtain a fifth nectar source; When the updated number of iterations is greater than the target number of iterations, the fifth nectar source is used as the target value of the objective function.

5. The method according to claim 4, characterized in that The employed bees based on the hybrid artificial bee colony strategy update the first nectar source according to the first differential nectar source, the roulette strategy and the preset mining strategy to obtain the second nectar source, comprising: Obtaining the number of weight reset cycles, the number of target weight reset cycles, the preset annealing temperature, the removal operator score, the number of times the removal operator is used, the insertion operator score, and the number of times the insertion operator is used of the preset mining strategy; The employed bees of the hybrid artificial bee colony strategy construct a first current solution according to the first differential nectar source set, and update the number of weight reset cycles; Calculating a removal operator weight and an insertion operator weight according to the removal operator score, the number of times the removal operator is used, the insertion operator score, and the number of times the insertion operator is used; According to the removal operator weight and the roulette strategy, a target removal operator is selected from the preset removal operators, and a destructive solution is obtained according to the target removal operator and the first current solution; According to the insertion operator weight and the roulette strategy, a target insertion operator is selected from the preset insertion operators, and a repair solution is obtained according to the target insertion operator and the damage solution; Obtaining a cost of the first current solution and a cost of the repair solution; When the cost of the repair solution is greater than or equal to the cost of the first current solution, obtaining a preset inferior solution acceptance value, and obtaining an inferior solution acceptance probability according to the cost of the repair solution, the cost of the first current solution, and a preset annealing temperature; When the inferior solution acceptance probability is greater than the preset inferior solution acceptance value, updating the first current solution according to the repair solution to obtain a second current solution; When the cost of the repair solution is less than the cost of the first current solution, updating the first current solution according to the repair solution to obtain a second current solution; When the cost of the second current solution is greater than or equal to the cost of the first honey source or the inferior solution acceptance probability is less than or equal to the preset inferior solution acceptance value, the removal operator score, the insertion operator score, the removal operator usage count, and the insertion operator usage count are updated according to the cost of the first honey source, the cost of the repair solution, the cost of the first current solution, and the cost of the second current solution; When the updated weight reset cycle number is not equal to the target weight reset cycle number, the weight reset cycle number is updated, the second current solution is used as the first current solution, and the step of calculating the removal operator weight and the insertion operator weight according to the removal operator score, the number of times the removal operator is used, the insertion operator score, and the number of times the insertion operator is used is returned; When the updated weight reset cycle number is equal to the target weight reset cycle number, the target removal operator and the target insertion operator are updated according to the updated removal operator score, the updated removal operator usage count, the updated insertion operator score, and the updated insertion operator usage count, and the target removal operator and the target insertion operator are replaced according to the updated result; When the cost of the second current solution is less than the cost of the first honey source, the removal operator score, the insertion operator score, the number of times the removal operator is used, and the number of times the insertion operator is used are updated, and the first honey source is updated according to the second current solution to obtain a second honey source.

6. The method according to claim 5, characterized in that The step of updating the removal operator score and the insertion operator score according to the cost of the first honey source, the cost of the repair solution, the cost of the first current solution, and the cost of the second current solution comprises: Obtaining a current score, a first score, a second score, and a third score of the preset removal operator and the preset insertion operator, wherein the third score is greater than the second score, and the second score is greater than the first score; When the cost of the repair solution is equal to the cost of the first current solution, updating the removal operator score and the insertion operator score according to the first score and the current score; When the cost of the second current solution is greater than or equal to the cost of the first honey source, updating the removal operator score and the insertion operator score according to the second score and the current score; When the cost of the second current solution is less than the cost of the first honey source, the removal operator score and the insertion operator score are updated according to the third score and the current score.

7. The method according to claim 4, characterized in that The step of updating the fourth nectar source according to the first disturbance strategy or the second disturbance strategy to obtain the fifth nectar source includes: Acquire initial loading information of the vehicle in the fourth honey source, and acquire a first disturbance path corresponding to the initial loading information and a first customer position on the first disturbance path according to a first disturbance strategy; Recording the first customer information on the first disturbance path, and clearing the first customer position; placing the customers in the first customer information into the first disturbance path in sequence, recording the target loading information corresponding to the position of the first customer, and obtaining the first disturbance honey source according to the target loading information; When the cost of the first disturbed nectar source is less than the cost of the fourth nectar source, updating the fourth nectar source according to the first disturbed nectar source to obtain a fifth nectar source; When the cost of the first disturbance nectar source is greater than or equal to the cost of the fourth nectar source, obtaining the number of disturbances of the second disturbance strategy and the target number of disturbances of the second disturbance strategy; Selecting a target vehicle from the fourth honey source, and obtaining a second disturbance path of the target vehicle; Recording the second customer information on the second disturbance path, and clearing the second customer position on the second disturbance path; Put the customers in the second customer information into the second disturbance path in sequence, record the second customer positions, obtain the second disturbance honey source, and update the disturbance times; When the updated disturbance number is not equal to the target disturbance number and the cost of the second disturbance nectar source is greater than the cost of the fourth nectar source, returning to the step of selecting a target vehicle in the fourth nectar source and obtaining a second disturbance path of the target vehicle; When the number of disturbances is equal to the target number of disturbances or the cost of the second disturbance nectar source is less than or equal to the cost of the fourth nectar source, the fourth nectar source is updated according to the second disturbance nectar source to obtain a fifth nectar source.

8. A time-dependent vehicle path planning device, characterized in that: The device comprises: A construction module, used to obtain vehicle information and time-varying road network information, and to construct constraint conditions and objective functions according to the vehicle information and the time-varying road network information; An obtaining module, used for obtaining a honey source set according to the constraint conditions, the time-varying road network information and the vehicle information, and constructing a hybrid artificial bee colony strategy according to the honey source set; A solution module is used to obtain a target value of an objective function according to the constraint conditions, the nectar source set and the hybrid artificial bee colony strategy, and to plan a path for the vehicle according to the target value.

9. A time-dependent vehicle path planning device, characterized in that: The device comprises: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program is configured to implement the steps of the time-dependent vehicle path planning method according to any one of claims 1 to 7.

10. A storage medium, characterized in that: The storage medium is a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the time-dependent vehicle path planning method according to any one of claims 1 to 7 are implemented.

Images