Multi-distribution center oriented unmanned vehicle logistics transportation scheduling method and system
By calculating customer affiliation and intimacy in multi-distribution-center autonomous vehicle logistics transportation and using an improved ant colony algorithm to optimize routes, the problems of low scheduling efficiency and poor path optimization ability are solved, achieving more efficient logistics transportation scheduling.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- GUANGDONG UNIV OF TECH
- Filing Date
- 2022-12-08
- Publication Date
- 2026-05-22
AI Technical Summary
Existing technologies suffer from low scheduling efficiency and poor path optimization capabilities in the scheduling of autonomous vehicle logistics transportation across multiple distribution centers.
By constructing a logistics transportation scheduling model, calculating the degree of belonging and intimacy among customers, and using an improved ant colony algorithm to optimize the matching relationship between customer groups and distribution centers, the optimal delivery route is obtained.
It has improved the efficiency and quality of logistics transportation scheduling and enabled more efficient route planning.
Smart Images
Figure CN116307964B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of logistics and transportation, and more specifically, to a method and system for scheduling unmanned vehicle logistics transportation for multiple distribution centers. Background Technology
[0002] The application of autonomous vehicles in logistics transportation is an important development direction for the transportation industry, and transportation scheduling is one of the key issues that urgently needs to be addressed. Multi-distribution-center autonomous vehicle logistics transportation scheduling refers to a situation where, within a regional transportation network, there are multiple distribution centers, each using multiple identical vehicles to deliver goods to several customers within that network. Each customer has a specific demand for goods, and each customer's goods are delivered by only one vehicle from a particular distribution center, fulfilling all customers' needs. The number of vehicles at each distribution center is fixed, and each vehicle has a load capacity limit. Each vehicle departs from its assigned distribution center, delivers goods to its corresponding customer, and finally returns to its assigned distribution center.
[0003] There is a multi-delivery-center logistics transportation scheduling method. In the process of realizing multi-delivery-center logistics transportation scheduling, the multi-delivery-center logistics transportation scheduling problem is transformed into multiple single-delivery-center logistics transportation scheduling problems based on the clustering analysis strategy. Then, the optimal vehicle path is searched using the harmony search algorithm based on the artificial fish swarm algorithm to obtain the optimal delivery path.
[0004] However, the above methods do not improve the matching of delivery customer groups and still suffer from low scheduling efficiency and poor route optimization capabilities. Summary of the Invention
[0005] To overcome the shortcomings of existing technologies, such as low scheduling efficiency and poor path optimization capabilities, this invention provides a method and system for scheduling unmanned vehicle logistics transportation for multiple distribution centers.
[0006] To solve the above-mentioned technical problems, the technical solution of the present invention is as follows:
[0007] Firstly, this invention proposes a method for scheduling unmanned vehicle logistics transportation for multiple distribution centers, comprising:
[0008] S1: Construct a logistics transportation scheduling model. The logistics transportation scheduling model describes how vehicles from multiple distribution centers complete delivery tasks for multiple customers within a preset time frame.
[0009] S2: Calculate the degree of belonging among customers, and establish the customer group of the logistics transportation scheduling model based on the degree of belonging.
[0010] S3: Calculate the affinity between the customer group and the distribution center, and allocate the customer group to the distribution center based on the affinity.
[0011] S4: The improved ant colony algorithm is used to solve the logistics transportation scheduling model that completes the allocation of customer groups, and the optimal delivery route for logistics transportation scheduling is obtained.
[0012] Secondly, this invention also proposes an unmanned vehicle logistics transportation scheduling system for multiple distribution centers, comprising:
[0013] The model building module is used to construct a logistics transportation scheduling model. This model describes how vehicles from multiple distribution centers complete delivery tasks for multiple customers within a preset time frame.
[0014] The customer group establishment module is used to calculate the degree of belonging among customers and establish the customer group of the logistics transportation scheduling model based on the degree of belonging.
[0015] The allocation module is used to calculate the affinity between a customer group and a distribution center, and allocate the customer group to the distribution center based on the affinity.
[0016] The solution module is used to solve the logistics transportation scheduling model that completes customer group allocation using an improved ant colony algorithm, so as to obtain the optimal delivery route for logistics transportation scheduling.
[0017] Compared with existing technologies, this invention improves the matching relationship between customer groups and distribution centers by calculating the degree of belonging among customers, establishing customer groups with high degree of belonging, and calculating the intimacy between customer groups and distribution centers, and assigning the customer groups to distribution centers with high intimacy. Furthermore, by introducing an improved ant colony algorithm, the practicality of the model can be further improved, and the efficiency and quality of logistics transportation scheduling can be enhanced. Attached Figure Description
[0018] Figure 1 This is a flowchart of a logistics transportation scheduling method according to an embodiment of this application.
[0019] Figure 2 This is a flowchart illustrating the establishment of a customer group according to an embodiment of this application.
[0020] Figure 3 This application describes the process of using the ant colony algorithm to solve the optimal delivery route for logistics transportation scheduling in an embodiment of the present application.
[0021] Figure 4 This is a delivery trajectory diagram of the optimal delivery route in an embodiment of this application.
[0022] Figure 5 This is an architecture diagram of the logistics transportation scheduling system according to an embodiment of this application. Detailed Implementation
[0023] The accompanying drawings are for illustrative purposes only and should not be construed as limiting the scope of this patent.
[0024] The technical solution of the present invention will be further described below with reference to the accompanying drawings and embodiments.
[0025] Example 1
[0026] Please see Figure 1 This embodiment proposes a method for scheduling unmanned vehicle logistics transportation for multiple distribution centers, including the following steps:
[0027] S1: Construct a logistics transportation scheduling model. The logistics transportation scheduling model describes how vehicles from multiple distribution centers complete delivery tasks for multiple customers within a preset time frame.
[0028] S2: Calculate the degree of belonging among customers, and establish the customer group of the logistics transportation scheduling model based on the degree of belonging.
[0029] S3: Calculate the affinity between the customer group and the distribution center, and allocate the customer group to the distribution center based on the affinity.
[0030] S4: The improved ant colony algorithm is used to solve the logistics transportation scheduling model that completes the allocation of customer groups, and the optimal delivery route for logistics transportation scheduling is obtained.
[0031] The autonomous vehicle logistics transportation scheduling method proposed in this embodiment for multiple distribution centers establishes a customer group with high affiliation by calculating the affiliation degree between customers, and calculates the intimacy between the customer group and the distribution center. The customer group is then assigned to the distribution center with high intimacy, thus improving the matching relationship between the customer group and the distribution center. Furthermore, an improved ant colony algorithm is introduced, which can further improve the practicality of the model and enhance the efficiency and quality of logistics transportation scheduling.
[0032] Example 2
[0033] This embodiment is an improvement upon the unmanned vehicle logistics transportation scheduling method for multiple distribution centers proposed in Embodiment 1.
[0034] To verify the feasibility and superiority of the present invention, this embodiment will apply the method of the present invention to solve the case delivery analysis of 3 given distribution centers and 30 random customers, and clearly and completely describe the technical solution of the present invention.
[0035] In this embodiment, the objective function of the logistics transportation scheduling model is expressed as follows:
[0036]
[0037] Where P is the total delivery cost, V is the set of all customers and distribution centers, and w ij x represents the delivery cost between customer i and customer j. ijk This is a binary decision variable. When its value is 1, it means that vehicle k delivers goods from customer i to customer j. When its value is 0, it means that vehicle k does not deliver goods from customer i to customer j.
[0038] The constraints of the objective function include:
[0039]
[0040]
[0041]
[0042]
[0043]
[0044]
[0045]
[0046]
[0047]
[0048] Equation (1) represents the maximum load limit of the delivery vehicle; Equation (2) indicates that each customer is served by only one vehicle once; Equation (3) indicates that there is no connecting path between the same customers; Equation (4) indicates that a vehicle serves only one path and must return to the starting point from the distribution center; Equation (5) ensures that all customers have vehicle service; Equations (6) and (7) indicate that when a customer is served, there must be a path connected to them; Equation (8) indicates the elimination of sub-loops; Equation (9) indicates the value of the decision variable.
[0049] Where I = {1, 2, ..., N} represents the set of all customers, N is a positive integer representing the number of customers, and g i Let y represent the required load capacity of customer i, T be the maximum load capacity of the vehicle, k = {1, 2, ..., K} represent the set of available vehicles at distribution centers, K is a positive integer representing the number of available vehicles, and m = {1, 2, ..., M} represent the set of all distribution centers, M is a positive integer representing the number of distribution centers; ik Let be a binary decision variable. When its value is 1, it means that customer i is served by vehicle k. When its value is 0, it means that customer i is not served by vehicle k. S is the set of customers served by vehicle k.
[0050] Table 1 Initialization Information
[0051]
[0052]
[0053] In the specific implementation process, the initial information such as the coordinates of the distribution centers, the coordinates of the logistics delivery customers, and the customer's required load capacity is shown in Table 1. Among them, the number of distribution centers M = 3, K... m =2 represents the number of delivery vehicles in distribution center m, T = 60 (unit: tons), and the total number of customers N = 30.
[0054] In this embodiment, as Figure 2 As shown, the establishment of the customer group for the logistics transportation scheduling model specifically includes the following steps:
[0055] S2.1: Randomly select the locations of C customers as the initial customer group centers, C = M;
[0056] S2.2: Calculate the degree of affiliation between each customer and the C customer group centers; the expression is as follows:
[0057]
[0058] Where AEF(c,i) represents the degree of affiliation between customer i and customer group center c, N is the total number of customers, λ is the distance weighting coefficient, β is the load weighting coefficient, and x c The x-coordinate represents the center c of the customer group. i The x-coordinate of customer i is represented by y. c The y-coordinate represents the center c of the customer group. i Represents the ordinate of customer i, n c This represents the number of customers assigned to customer group center c.
[0059] S2.3: Assign each customer to the customer group center with the highest current affinity, and update the customer group center; the specific expression for the update is as follows:
[0060]
[0061] Where, x c and y c Let n represent the x-coordinate and y-coordinate of the updated customer group center c, respectively. c n represents the number of customers in the c-th customer group. i c Let i represent customer object i in the c-th customer group;
[0062] If the customer base approaches the capacity constraint, no new customers will be accepted. The expression for the capacity constraint is as follows:
[0063]
[0064] G c =K m T
[0065] Where, N c Let G represent the maximum number of customers in the c-th customer group, σ be the margin coefficient and 0 ≤ σ ≤ 0.5. c K represents the maximum carrying capacity of the c-th customer group. m This indicates the number of delivery vehicles in distribution center m.
[0066] S2.4: Determine whether the customer group center has converged. If the customer group center has not converged, proceed to S2.2. If the customer group center has converged, end the process and obtain C customer groups and customer group centers.
[0067] In this embodiment, the process of allocating the customer group to the distribution center specifically includes the following steps:
[0068] S3.1: Calculate the affinity between each distribution center and each customer group center, and construct a sequence of affinity objects for customer group centers based on the affinity; the calculation expression for the affinity is as follows:
[0069]
[0070] Where DOI(m,c) represents the proximity between distribution center m and customer group center c, n c K represents the number of customers in customer group center c. m g represents the number of customer clusters for distribution center m. i For customer i, the required load capacity is given by μ, where μ represents the load capacity influence coefficient, γ represents the distance influence coefficient, and x... c The x-coordinate of the customer group center c is represented by y. c The ordinate of the customer group center c is x. m The x-coordinate of distribution center m is represented by y. m The vertical coordinate of distribution center m is represented.
[0071] S3.2: Based on the customer group center's affinity object sequence, assign the currently unmatched customer group center to the distribution center with the highest affinity with it, and determine whether the distribution center has already established a match with other customer group centers;
[0072] If the distribution center is not matched with other customer group centers, then the currently unmatched customer group centers will be directly matched with the distribution center.
[0073] If the distribution center has already been matched with other customer group centers, then determine whether the affinity between the currently unmatched customer group center and the distribution center is higher than the affinity between the currently matched other customer group centers and the distribution center; if it is higher, then remove the original match between the distribution center and other customer group centers, and match the currently unmatched customer group center with the distribution center; if it is lower than or equal to, then proceed to step S3.3.
[0074] S3.3: Update the sequence of intimacy objects in the customer group center;
[0075] S3.4: Determine whether all customer group centers are matched with distribution centers. If there are customer group centers that are not matched with distribution centers, then proceed to S3.2; otherwise, end the process.
[0076] In this embodiment, as Figure 3 As shown, the improved ant colony algorithm is used to solve the logistics transportation scheduling model to obtain the optimal delivery route for logistics transportation scheduling. The specific steps include:
[0077] S4.1: Initialize control parameters; the control parameters include the maximum number of iterations Max_Iter, the current algorithm iteration number Iter, the total number of ants A, the heuristic factor α, the pheromone evaporation coefficient ρ, and the pheromone intensities Q, τ, and Δτ; the ants represent vehicles;
[0078] S4.2: Set the starting point of all ants' access paths to the distribution center;
[0079] S4.3: Construct the access path for ant 'a'; the specific steps include:
[0080] S4.3.1: Calculate the probability that ant a will visit the next customer. Its expression is as follows:
[0081]
[0082] Where i represents customer i, j represents customer j, c is the customer group center, and τ ij (a) represents the pheromone intensity of the access path formed by ant a between customer i and customer j, and α is the heuristic factor.
[0083] S4.3.2: If customer j meets the constraints, then move ant a to customer j and add customer j to the tabu list set; the tabu list set is the set of target customers that have been visited.
[0084] S4.3.3: Determine if all customers have visited. If yes, proceed to S4.4; otherwise, proceed to S4.3.1.
[0085] S4.4: Determine whether the access paths for all ants have been constructed. If yes, execute S4.5; otherwise, jump to execute S4.3.
[0086] S4.5: Update the pheromone of the access path; the specific expression is as follows:
[0087]
[0088] Where, τ max ρ represents the maximum intensity of the pheromone, ρ is the pheromone evaporation coefficient, and τ is the pheromone evaporation coefficient. ij (Iter) represents the pheromone intensity of the path formed by customer i and customer j after the Iter-th iteration, Δτ ij (Iter) represents the pheromone intensity increment of the path formed by customer i and customer j after the Iter-th iteration, τ. min This represents the minimum intensity of the pheromone.
[0089] Pheromonium intensity increment Δτ ij The expression for calculating (Iter) is as follows:
[0090]
[0091] Among them, L a l represents the length of the path ant a travels. ij Let a represent the path formed by customer i and customer j. ij For the path l ij The number of ants, A is the total number of ants, L gh The value represents the length of the current optimal solution path, L represents the current optimal solution path, and rand(0,1) is a random number uniformly distributed on [0,1].
[0092] S4.6: Obtain the optimal access path with the lowest cost from the constructed access paths;
[0093] S4.7: Set the number of local optimizations z, randomly select two unconnected customers, reverse the path between the two unconnected customers to obtain a new access path, determine whether the cost of the new access path is less than the current optimal access path, if so, then take the new access path as the optimal access path and directly execute S4.8; if not, randomly select two unconnected customers again to construct a new access path, until the number of random selections is equal to the number of local optimizations z, and execute S4.8.
[0094] S4.8: Determine if the current ant colony algorithm iteration count (Iter) is less than the maximum iteration count (Max_Iter); if yes, proceed to S4.2; otherwise, end the process and use the optimal access path obtained in S4.7 as the optimal delivery path for logistics transportation scheduling. The optimal delivery path is as follows: Figure 4 As shown in Table 2.
[0095] In practical implementation, the autonomous vehicle logistics transportation scheduling problem for multiple distribution centers can be described as follows: In a regional transportation network, there are multiple distribution centers. Each distribution center uses multiple identical vehicles to deliver goods to several customers within the network. Each customer has a required quantity of goods, and each customer's goods are delivered by only one vehicle from a specific distribution center, fulfilling all customers' needs. The number of vehicles at each distribution center is fixed, and each vehicle has a load capacity limit. Each vehicle departs from its assigned distribution center, delivers goods to the corresponding customer, and finally returns to its assigned distribution center. The optimal delivery path is to assign delivery tasks to each vehicle at each distribution center and plan a route for each vehicle that minimizes the total delivery distance.
[0096] Table 2 Optimal Delivery Route
[0097]
[0098] exist Figure 4 As shown in Table 2, vehicle A1 departs from distribution center A, delivers goods to customers 4, 23, and 24 in sequence, and finally returns to distribution center A, covering a distance of 52km and carrying an actual load of 41 tons; vehicle A2 departs from distribution center A, delivers goods to customers 1, 20, 27, 11, and 28 in sequence, and finally returns to distribution center A, covering a distance of 102km and carrying an actual load of 32 tons; vehicle B1 departs from distribution center B, delivers goods to customers 10, 5, 6, 22, 12, 2, and 19 in sequence, and finally returns to distribution center B, covering a distance of 74km and carrying an actual load of 52 tons. Vehicle B2 departs from distribution center B, delivers goods to customers 14, 17, and 8 in sequence, and finally returns to distribution center B, with a mileage of 135km and an actual load of 32 tons; Vehicle C1 departs from distribution center C, delivers goods to customers 9, 26, 16, 18, and 25 in sequence, and finally returns to distribution center C, with a mileage of 111km and an actual load of 33 tons; Vehicle C2 departs from distribution center C, delivers goods to customers 30, 13, 29, 21, 15, 7, and 3 in sequence, and finally returns to distribution center C, with a mileage of 107km and an actual load of 45 tons.
[0099] Example 3
[0100] See Figure 5 This embodiment proposes an unmanned vehicle logistics transportation scheduling system for multiple distribution centers, including:
[0101] The model building module is used to construct a logistics transportation scheduling model. This model describes how vehicles from multiple distribution centers complete delivery tasks for multiple customers within a preset time frame.
[0102] The customer group establishment module is used to calculate the degree of belonging among customers and establish the customer group of the logistics transportation scheduling model based on the degree of belonging.
[0103] The allocation module is used to calculate the affinity between a customer group and a distribution center, and allocate the customer group to the distribution center based on the affinity.
[0104] The solution module is used to solve the logistics transportation scheduling model that completes customer group allocation using an improved ant colony algorithm, so as to obtain the optimal delivery route for logistics transportation scheduling.
[0105] The autonomous vehicle logistics transportation scheduling system for multiple distribution centers proposed in this embodiment establishes a customer group with high affiliation by calculating the affiliation degree between customers, and calculates the intimacy between the customer group and the distribution center. The system then assigns the customer group to the distribution center with high intimacy, improves the matching relationship between the customer group and the distribution center, and introduces an improved ant colony algorithm, which can further improve the practicality of the model and improve the efficiency and quality of logistics transportation scheduling.
[0106] The terms used to describe positional relationships in the accompanying drawings are for illustrative purposes only and should not be construed as limiting this patent.
[0107] Obviously, the above embodiments of the present invention are merely examples for clearly illustrating the present invention, and are not intended to limit the implementation of the present invention. Those skilled in the art can make other variations or modifications based on the above description. It is neither necessary nor possible to exhaustively describe all embodiments here. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the scope of protection of the claims of the present invention.
Claims
1. A method and system for scheduling unmanned vehicle logistics transportation for multiple distribution centers, characterized in that: include: S1: Construct a logistics transportation scheduling model; the logistics transportation scheduling model is a model that describes how vehicles from multiple distribution centers complete delivery tasks for multiple customers within a preset time range; S2: Calculate the degree of belonging among customers, and establish the customer group of the logistics transportation scheduling model based on the degree of belonging; S3: Calculate the affinity between the customer group and the distribution center, and allocate the customer group to the distribution center based on the affinity, including: S3.1: Calculate the affinity between each distribution center and each customer group center, and construct a sequence of affinity objects for customer group centers based on the affinity; the calculation expression for the affinity is as follows: in, Indicates distribution center m With Customer Center c The level of intimacy between them n c Indicates customer group center c The number of customers, K m Indicates distribution center m The number of customer group centers g i For customers i Required load capacity Indicates the load-bearing influence coefficient. This represents the distance influence coefficient. Indicates customer group center c x-coordinate Indicates customer group center c The ordinate, Indicates distribution center m x-coordinate Indicates distribution center m The ordinate; S3.2: Based on the customer group center's affinity object sequence, assign the currently unmatched customer group center to the distribution center with the highest affinity with it, and determine whether the distribution center has already established a match with other customer group centers; If the distribution center has not established a match with other customer group centers, then the currently unmatched customer group centers will directly establish a match with the distribution center and proceed to step S3.3; If the distribution center has already been matched with other customer group centers, then determine whether the affinity between the currently unmatched customer group center and the distribution center is higher than the affinity between the currently matched other customer group centers and the distribution center; if it is higher, then remove the original match between the distribution center and other customer group centers, and match the currently unmatched customer group center with the distribution center; if it is lower than or equal to, then proceed to step S3.
3. S3.3: Update the sequence of intimacy objects in the customer group center; S3.4: Determine whether all customer group centers are matched with distribution centers. If there are customer group centers that are not matched with distribution centers, proceed to S3.2; otherwise, end. S4: The improved ant colony algorithm is used to solve the logistics transportation scheduling model that completes the allocation of customer groups, and the optimal delivery route for logistics transportation scheduling is obtained.
2. The unmanned vehicle logistics transportation scheduling method and system for multiple distribution centers according to claim 1, characterized in that, The objective function of the logistics transportation scheduling model is expressed as follows: in, P For the total delivery cost, V For the collection of all customers and distribution centers, Indicates customer i and customers j Delivery costs between This is a binary decision variable; a value of 1 indicates a vehicle. k From the customer i Delivery to customer j When its value is 0, it indicates a vehicle. k Not from the customer i Delivery to customer j .
3. The unmanned vehicle logistics transportation scheduling method and system for multiple distribution centers according to claim 2, characterized in that, The constraints of the objective function include: in, I ={1,2,..., N } represents the set of all customers. N The number of customers is a positive integer. Indicates customer i The required load capacity T This is the vehicle's maximum load capacity. k ={1,2,... K } indicates the vehicle number available at the distribution center. K A positive integer represents the number of available vehicles. m ={1,2,..., M } represents the distribution center number. M The number of distribution centers is a positive integer. y ik This is a binary decision variable; a value of 1 indicates that the customer... i By vehicle k The service, when its value is 0, represents a customer. i Not by vehicle k Serve; S For vehicles k The collection of customers served.
4. The unmanned vehicle logistics transportation scheduling method and system for multiple distribution centers according to claim 3, characterized in that, The establishment of the customer group for the logistics transportation scheduling model specifically includes the following steps: S2.1: Random selection C The location of each customer is used as the initial customer group center. C = M ; S2.2: Calculate the relationship between each customer and the aforementioned... C The degree of belonging between customer group centers; its expression is as follows: in, Indicates customer i With Customer Center c Degree of belonging between them Total number of customers These are the weighting coefficients for distance. This is the load-weighted factor. Indicates customer group center c x-coordinate Indicates customer i x-coordinate Indicates customer group center c The ordinate, Indicates customer i The ordinate, n c Indicates customer group center c The number of customers already assigned; S2.3: Assign each customer to the customer group center with the highest current affinity, and update the customer group center; the specific expression for the update is as follows: in, x c and y c These represent the updated customer group center. c The x and y coordinates; n c Indicates the first c The number of customers included in a customer group n i c Indicates the first c Customers in a customer group i ; S2.4: Determine whether the customer group center has converged. If the customer group center has not converged, proceed to S2.2; if the customer group center has converged, end the process and obtain the result. C Individual customer groups and customer group center.
5. The unmanned vehicle logistics transportation scheduling method and system for multiple distribution centers according to claim 4, characterized in that, The improved ant colony algorithm is used to solve the logistics transportation scheduling model to obtain the optimal delivery route for logistics transportation scheduling. The specific steps include: S4.1: Initialize control parameters; the control parameters include the maximum number of iterations. Max_Iter Current number of algorithm iterations Iter Total number of ants A Heuristic factors pheromone volatility coefficient and pheromone intensity Q , and The ants represent vehicles. S4.2: Set the starting point of all ants' access paths to the distribution center; S4.3: Building Ants a The access path; S4.4: Determine whether the access paths for all ants have been constructed. If yes, execute S4.5; otherwise, jump to execute S4.
3. S4.5: Update the pheromone of the access path; S4.6: Obtain the optimal access path with the lowest cost from the constructed access paths; S4.7: Set the number of local optimizations z, randomly select two unconnected customers, reverse the path between the two unconnected customers to obtain a new access path, determine whether the cost of the new access path is less than the current optimal access path, if so, then take the new access path as the optimal access path and directly execute S4.8; if not, randomly select two unconnected customers again to construct a new access path, until the number of random selections is equal to the number of local optimizations z, and execute S4.
8. S4.8: Determine the current ant colony algorithm iteration number. Iter Is it less than the maximum number of iterations? Max_Iter If yes, then proceed to S4.2; otherwise, end the process and use the optimal access path obtained in S4.7 as the optimal delivery path for logistics transportation scheduling.
6. The unmanned vehicle logistics transportation scheduling method and system for multiple distribution centers according to claim 5, characterized in that, The specific steps in S4.3 include: S4.3.1: Calculating Ants a The probability of the next customer visit Its expression is as follows: in, i Indicates customer i , j Indicates customer j , Centered on the customer base For ants a In customers i and customers j The pheromone intensity of the access path formed between them As a heuristic factor; S4.3.2: If customer j meets the constraints, then the ant... a Move to customer j and customers j Add to the taboo list set; the taboo list set is the set of target customers that have been visited. S4.3.3: Determine if all customers have visited. If yes, proceed to S4.4; otherwise, proceed to S4.3.
1.
7. The unmanned vehicle logistics transportation scheduling method and system for multiple distribution centers according to claim 6, characterized in that, In S4.5, the specific expression for updating the pheromone of the access path is as follows: in, For the maximum pheromone intensity, The pheromone evaporation coefficient, Indicates after the first Iter After the next iteration, by the customer i With customers j The pheromone intensity of the formed path, For the first Iter After the next iteration, by the customer i With customers j The pheromone intensity increment of the formed path This represents the minimum intensity of the pheromone.
8. The unmanned vehicle logistics transportation scheduling method and system for multiple distribution centers according to claim 7, characterized in that, Pheromon intensity increment The calculation expression is as follows: in, L a Ants a Length of the path traversed l ij Indicated by the customer i With customers j The path formed a ij For the path l ij The number of ants, A This represents the total number of ants. L gh This indicates the length of the current optimal solution path. L Indicates the current optimal solution path. rand (0,1) is a random number uniformly distributed on [0,1].
9. A multi-distribution-center unmanned vehicle logistics transportation scheduling system, applied to the multi-distribution-center unmanned vehicle logistics transportation scheduling method as described in any one of claims 1 to 8, characterized in that, include: The model building module is used to build logistics transportation scheduling models; The logistics transportation scheduling model is a model that describes how vehicles from multiple distribution centers complete delivery tasks for multiple customers within a preset time range. The customer group establishment module is used to calculate the degree of belonging among customers and establish the customer group of the logistics transportation scheduling model based on the degree of belonging. The allocation module is used to calculate the affinity between a customer group and a distribution center, and allocate the customer group to the distribution center based on the affinity. The solution module is used to solve the logistics transportation scheduling model that completes customer group allocation using an improved ant colony algorithm, so as to obtain the optimal delivery route for logistics transportation scheduling.