Dynamic path planning method and device based on scenario prediction

Abstract

The application provides a dynamic path planning method and device based on scene prediction, comprising the following steps: constructing a static vehicle path planning model; establishing a target function of dynamic vehicle path planning; calculating the influence degree of potential customers; constructing a predicted customer set according to the influence degree of potential customers; and solving the dynamic vehicle path problem based on the predicted customer set. According to the change degree of potential customers to the path of known customers, the influence degree of each potential customer is calculated; the influence degree is used to generate a predicted scene through roulette; the path information is updated according to the predicted scene in a fixed time slice, so that the planning efficiency and the quality of the solution are improved; the scene state is adjusted based on the periodic re-optimization and continuous re-optimization method; and a taboo list is set for the to-be-visited and visited routes, so that the influence on the established route in the local search and greedy insertion process is avoided. The method effectively solves the path optimization problem in dynamic path planning.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of vehicle path planning, in particular to a dynamic path planning method and device based on scene prediction. BACKGROUND

[0002] With the rapid development of the logistics industry, dynamic vehicle routing problem (DVRP) has become an important research direction for solving intracity distribution and real-time transportation scheduling. The core of DVRP is to plan the optimal vehicle path to meet the real-time customer demand with the lowest travel cost. However, the dynamic and uncertainty of new customer requests pose a serious challenge to path optimization, which requires frequent adjustment of the established travel route to complete the distribution task.

[0003] Existing methods mainly focus on coping with dynamic changes by improving the global search ability of the algorithm or enhancing the diversity of the solution, but there are relatively few studies on the prediction of new customer requests. Traditional prediction methods usually rely on historical data, but in practical applications, these data are often difficult to obtain, limiting their application range. SUMMARY

[0004] The dynamic path planning method and device based on scene prediction and the storage medium provided by the present application can at least solve one of the technical problems in the background art.

[0005] To achieve the above-mentioned purpose, the present application adopts the following technical solutions:

[0006] A dynamic path planning method based on scene prediction, executed by a computer device, comprising the following steps:

[0007] S1, constructing a static vehicle path planning model;

[0008] S2, establishing a target function of dynamic vehicle path planning;

[0009] S3, calculating the influence degree of potential customers;

[0010] S4, constructing a predicted customer set according to the influence degree of potential customers;

[0011] S5, solving the dynamic vehicle routing problem based on the predicted customer set.

[0012] Further, applied to a scene composed of 1 warehouse , n customers , and m transport vehicles, the node set is composed of the 1 warehouse and n customers , , wherein, The customers are divided into known customers and potential customers, thus obtaining the set of known customers. and potential customer base ;

[0013] Node set The straight-line paths between each node are used as the edge set. ;in, Represents any number of Nodes and the Nodes The straight path between; let the straight path be... The straight-line distance is denoted as Let the maximum load capacity of each transport vehicle be... Assume all vehicles are warehouses For the start and end points;

[0014] Step S1 specifically includes,

[0015] Step 1.1: Given a path planning time of T, and set T to... Given three time slices of equal length, define the index of any time slice as q, and initialize... =1;

[0016] Step 1.2, the previous... All customers appearing in a given time slice are considered as the known customer set. , will the current number All customers appearing in a given time slot are considered as a new customer group. ;

[0017] Step 1.3: Use equation (1) to establish the current number of... The objective function of the static vehicle path planning model in a time slice :

[0018] (1)

[0019] Equation (1) represents the total distance traveled by all vehicles. Indicates the first The first time slice The first transport vehicle The node and the first The straight-line distance between nodes Indicates the first Did the u-th car pass through the i-th time slice? The node and the first A straight path between nodes, if it passes through, then let 1. Otherwise, let ;

[0020] Step 1.2: Construct the current number using equations (2)-(7). Constraints of the static vehicle routing model under each time slice:

[0021] (2)

[0022] (3)

[0023] (4)

[0024] (5)

[0025] (6)

[0026] (7)

[0027] (2) and (3) indicate that each customer must access the site once and only once;

[0028] In equation (4), m is the total number of vehicles located in the warehouse. This indicates vehicles departing from the warehouse. Constraint (4) ensures that the number of dispatched vehicles should not exceed the total number of vehicles available in the warehouse.

[0029] Constraint (5) is a sub-loop elimination constraint. Represents a set The number of customers in the country;

[0030] In equations (6) and (7) Indicates customer The demand and constraint (6) ensure that the total demand on each route does not exceed the vehicle capacity.

[0031] Furthermore, in step 2, the objective function for dynamic vehicle routing is established using equation (8). :

[0032] (8).

[0033] Further, step 3 includes knowing the customer set. Add a potential customer Thus constituting the first l temporary customer group According to the current number A static vehicle routing model for a given time slice, for Solving for the problem yields the following results: Corresponding customer access path sequence Therefore, based on Calculate potential customers degree of influence ;

[0034] Step 3.1: Calculate the capacity of the delivery vehicle using formula (9). :

[0035] (9)

[0036] In equation (9), Represents a known customer set The number of known customers in the database;

[0037] Step 3.2, according to Using ant colony algorithm The customer set Solve the problem to obtain the customer access sequence. And calculate the length of the customer access sequence. Using ant colony optimization algorithm Solve the problem to obtain the customer access sequence. And calculate the length of the customer access sequence. ;

[0038] Step 3.3: Calculate the degree of influence of potential customers according to formulas (10)-(14) based on the obtained customer visit sequence and the length of the customer visit sequence;

[0039] (10)

[0040] (11)

[0041] (12)

[0042] (13)

[0043] (14)

[0044] The function calculation method is as follows: [This refers to] potential customers. from Delete it, and get the deletion result. The customer access sequence, compared and The different numbers of customers in the same location Function representation The solution includes the number of customers; α and β are control parameters that adjust the importance of disturbances to travel costs and routes, in equation (14). This indicates the potential customers acquired through calculation. The extent of the impact.

[0045] Further, step S4 includes step 4.2, calculating the number of customers that need to be added to the predicted customer set:

[0046] (15)

[0047] In formula (15) express The number of customers in the country;

[0048] Step 4.3: Based on the degree of impact For potential customers Sort the customers in descending order;

[0049] Step 4.4, from potential customers Before the election A customer is added to the known customer set. In this context, the predicted customer set is formed. At the same time from potential customers Delete the corresponding customer;

[0050] Step 4.5: Optimize the access route for the predicted scenario based on the static path problem model. .

[0051] Furthermore, step S5 specifically includes,

[0052] Step 5.2, the At the beginning of each time segment, new customers will emerge. Add to the currently known customer set ;

[0053] Step 5.3: Determine the size of the predicted customer set. Did the expected quantity reach the target?

[0054] (16)

[0055] In formula (16) express The number of customers in the country;

[0056] Step 5.4, if According to the degree of impact From potential customer set Select the customer with the greatest influence. Add to the predicted customer set and merge it into the new customer set. In, and from Delete;

[0057] like According to the degree of impact From predicting customer sets Select the most influential potential customer from among them. Meanwhile, from predicting customer sets Delete Join the potential customer group ,exist Middle connection The previous customer and the next customer;

[0058] Step 5.5, repeat step 5.4 until... ;

[0059] Step 5.5: Use a greedy insertion method to add new customer sets. Insert into path superior;

[0060] Step 5.6, Local Search:

[0061] Step 5.7: Repeat steps 5.2 to 5.6 until the time slice ends, and return the solution of dynamic path planning.

[0062] Further, step 5.6, local search, specifically includes,

[0063] Step 5.6.1: Define five local search operators Perform the following operation:

[0064] 1. Randomly select a route One of the customers , randomly move its position;

[0065] 2. Randomly select routes Two clients and Exchange customer locations;

[0066] 3. Randomly select routes Two routes and ,from Randomly select a customer ,from Randomly select a customer ,exchange from after, from The route afterward;

[0067] 4. Randomly select routes One of the routes ,from Two customers were randomly selected. and ,Will and The order of customers is reversed;

[0068] 5. Randomly select routes The route in ,from Two customers were randomly selected. and ,Will and Customers in between are randomly inserted into other positions;

[0069] Step 5.6.2: Randomly select a search operator for optimization, and continue optimizing until the time reaches the allocated time for the current scene. .

[0070] In another aspect, the present invention also discloses a computer-readable storage medium storing a computer program, which, when executed by a processor, causes the processor to perform the steps of the method described above.

[0071] In another aspect, the present invention also discloses a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor performs the steps of the method described above.

[0072] As can be seen from the above technical solution, this invention proposes a scenario prediction method based on the degree of impact. By roughly estimating the location and demand of potential customers, it generates a complete customer network within a dynamic planning time frame, quickly processing new customer requests and reducing operating costs. This method does not rely on detailed historical data, is suitable for rapid response needs in dynamic environments, and improves the efficiency and accuracy of path planning while reducing computational complexity.

[0073] The present invention discloses a dynamic path planning method and system based on scenario prediction, comprising: calculating the impact of each potential customer on the path formed by known customers based on the degree of change of potential customers; generating a predicted scenario based on roulette wheel using the impact degree; designing a multi-scenario prediction strategy, updating path information by selecting the current optimal solution and synchronizing it with other scenarios; performing multi-scenario synchronization on fixed time slices to improve the efficiency and solution quality in the planning process; adjusting the scenario state based on a path planning method of periodic re-optimization and continuous re-optimization; and setting a taboo list for routes to be visited and routes already visited to avoid affecting the determined routes in local search and greedy insertion. This method effectively solves the multi-scenario synchronization problem in dynamic path planning and can continuously optimize the path execution performance of vehicles in complex scenarios.

[0074] Specifically, this invention provides a dynamic path planning method and system based on scenario prediction, which has the following advantages: This invention solves the problem of lacking historical data in actual dynamic vehicle path planning by using an innovative disturbance level scenario prediction method, without relying on detailed historical data of customer requests, but only utilizing rough information such as the location of potential customers and the expected number of requests. By analyzing the disturbance level of customers offline, prioritizing high-disturbance customers and pre-compiling their data into the initial customer network, it effectively reduces online computational complexity and achieves rapid response to dynamic changes. The system designs an efficient multi-scenario collaboration mechanism, improving path planning efficiency and solution quality through solution synchronization and information sharing between scenarios. Combining periodic re-optimization and continuous re-optimization methods, it can continuously optimize path planning in dynamic environments, ensuring stability and adaptability in complex scenarios. Furthermore, this invention sets a taboo list during the planning process to avoid the impact of local search and greedy insertion on the established route, thereby further ensuring the effectiveness of the planning results. Experimental results show that this invention exhibits significant superiority in multiple dynamic path planning instances, providing a low-cost, high-efficiency path planning scheme, with significant optimization effects and practical value compared to existing technologies. Attached Figure Description

[0075] Figure 1 This is a flowchart of the present invention;

[0076] Figure 2 This is a comparison chart of the results of the present invention and other methods. Detailed Implementation

[0077] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are some embodiments of the present invention, but not all embodiments.

[0078] like Figure 1 As shown in this embodiment, the multi-scenario collaborative dynamic path planning method based on population evolution algorithm is applied to a single warehouse. n customers In a scenario consisting of m transport vehicles, the 1 warehouse and n customers constituting a node set , ,in, The customers are divided into known customers and potential customers, thus obtaining the set of known customers. and potential customer base ;

[0079] Node set The straight-line paths between each node are used as the edge set. ;in, Represents any number of Nodes and the Nodes The straight path between; let the straight path be... The straight-line distance is denoted as Let the maximum load capacity of each transport vehicle be... Assume all vehicles are warehouses The multi-scenario collaborative dynamic path planning method proceeds as follows: (The starting and ending points are defined as follows:)

[0080] Step 1: Construct a static vehicle routing model:

[0081] Step 1.1: Given a path planning time of T, and set T to... Given three time slices of equal length, define the index of any time slice as q, and initialize... =1;

[0082] Step 1.2, the previous... All customers appearing in a given time slice are considered as the known customer set. , will the current number All customers appearing in a given time slot are considered as a new customer group. ;

[0083] Step 1.3: Use equation (1) to establish the current number of... The objective function of the static vehicle path planning model in a time slice :

[0084] (1)

[0085] Equation (1) represents the total distance traveled by all vehicles. Indicates the first The first time slice The first transport vehicle The node and the first The straight-line distance between nodes Indicates the first Did the u-th car pass through the i-th time slice? The node and the first A straight path between nodes, if it passes through, then let 1. Otherwise, let ;

[0086] Step 1.2: Construct the current number using equations (2)-(7). Constraints of the static vehicle routing model under each time slice:

[0087] (2)

[0088] (3)

[0089] (4)

[0090] (5)

[0091] (6)

[0092] (7)

[0093] (2) and (3) indicate that each customer must access the site once and only once;

[0094] In equation (4), m is the total number of vehicles located in the warehouse. This indicates vehicles departing from the warehouse. Constraint (4) ensures that the number of dispatched vehicles should not exceed the total number of vehicles available in the warehouse.

[0095] Constraint (5) is a sub-loop elimination constraint. Represents a set The number of customers in the country;

[0096] In equations (6) and (7) Indicates customer demand, Constraint (6) ensures that the total demand on each route does not exceed the vehicle capacity.

[0097] Step 2: Use equation (8) to establish the objective function for dynamic vehicle path planning. :

[0098] (8)

[0099] Step 3, Known customer set Add a potential customer Thus constituting the first l temporary customer group According to the current number A static vehicle routing model for a given time slice, for Solving for the problem yields the following results: Corresponding customer access path sequence Therefore, based on Calculate potential customers degree of influence ;

[0100] Step 3.1: Calculate the capacity of the delivery vehicle using formula (9). :

[0101] (9)

[0102] In equation (10), Represents a known customer set The number of known customers in the database;

[0103] Step 3.2, according to Using ant colony algorithm The customer set Solve the problem to obtain the customer access sequence. And calculate the length of the customer access sequence. Using ant colony optimization algorithm Solve the problem to obtain the customer access sequence. And calculate the length of the customer access sequence. ;

[0104] Step 3.3: Calculate the degree of influence of potential customers according to formulas (10)-(14) based on the obtained customer visit sequence and the length of the customer visit sequence;

[0105] (10)

[0106] (11)

[0107] (12)

[0108] (13)

[0109] (14)

[0110] The function calculation method is as follows: [This refers to] potential customers. from Delete it, and get the deletion result. The customer access sequence, compared and The different numbers of customers in the same location Function representation The solution includes the number of customers; α and β are control parameters that adjust the importance of disturbances to travel costs and routes, in equation (14). This indicates the potential customers acquired through calculation. The extent of the impact;

[0111] Step 4: Construct the predicted customer set:

[0112] Step 4.2: Calculate the number of customers that need to be added to the predicted customer set.

[0113] (15)

[0114] In formula (15) express The number of customers in the country;

[0115] Step 4.3: Based on the degree of impact For potential customers Sort the customers in descending order;

[0116] Step 4.4, from potential customers Before the election A customer is added to the known customer set. In this context, the predicted customer set is formed. At the same time from potential customers Delete the corresponding customer;

[0117] Step 4.5: Optimize the access route for the predicted scenario based on the static path problem model. ;

[0118] Step 5: Solve the dynamic vehicle routing problem based on the predicted customer set.

[0119] Step 5.1: Initialize the online prediction scenario;

[0120] Step 5.2, the At the beginning of each time segment, new customers will emerge. Add to the currently known customer set ;

[0121] Step 5.3: Determine the size of the predicted customer set. Did the expected quantity reach the target?

[0122] (16)

[0123] In formula (16) express The number of customers in the country;

[0124] Step 5.4, if According to the degree of impact From potential customer set Select the customer with the greatest influence. Add to the predicted customer set and merge it into the new customer set. In, and from Delete;

[0125] like According to the degree of impact From predicting customer sets Select the most influential potential customer from among them. Meanwhile, from predicting customer sets Delete Join the potential customer group ,exist Middle connection The previous customer and the next customer;

[0126] Step 5.5, repeat step 5.4 until... ;

[0127] Step 5.5: Use a greedy insertion method to add new customer sets. Insert into path superior;

[0128] Step 5.6, Local Search:

[0129] Step 5.6.1: Define five local search operators Perform the following operation:

[0130] 1. Randomly select a route One of the customers , randomly move its position;

[0131] 2. Randomly select routes Two clients and Exchange customer locations;

[0132] 3. Randomly select routes Two routes and ,from Randomly select a customer ,from Randomly select a customer ,exchange from after, from The route afterward;

[0133] 4. Randomly select routes One of the routes ,from Two customers were randomly selected. and ,Will and The order of customers is reversed;

[0134] 5. Randomly select routes The route in ,from Two customers were randomly selected. and ,Will and Customers in between are randomly inserted into other positions;

[0135] Step 5.6.2: Randomly select a search operator for optimization, and continue optimizing until the time reaches the allocated time for the current scene. ;

[0136] Step 5.7: Repeat steps 5.2 to 5.6 until the time slice ends, and return the solution of dynamic path planning.

[0137] As shown in Table 1, this embodiment of the invention generated ten datasets, GoldenD, containing potential customers, based on the Golden dataset. The algorithm proposed in this invention was compared with the three most advanced algorithms for solving dynamic vehicle routing problems. The results show that the algorithm proposed in this invention achieves better results on all ten datasets. In Table 1, PB-ILS stands for Population-Based Iterated Local Search Approach for Dynamic Vehicle Routing Problems; RDMOVR stands for Robust Dynamic Multi-Objective Vehicle Routing Optimization Method; and AGA-ES-TW stands for Elastic Strategy-Based Adaptive Genetic Algorithm for Solving Dynamic Vehicle Routing Problem With Time Windows.

[0138] Table 1

[0139]

[0140] To verify the effectiveness of the method of this invention, while keeping other parts unchanged, the scene prediction module was removed, and experiments were conducted on the GoldenD dataset under the same operating environment. Figure 2 As shown, the results demonstrate that the scene prediction module can significantly reduce the length of the generated path, verifying its effectiveness.

[0141] In another aspect, the present invention also discloses a computer-readable storage medium storing a computer program, which, when executed by a processor, causes the processor to perform the steps of the method described above.

[0142] In another aspect, the present invention also discloses a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor performs the steps of the method described above.

[0143] In another embodiment provided in this application, a computer program product containing instructions is also provided, which, when run on a computer, causes the computer to execute any of the multi-scenario collaborative dynamic path planning methods based on population evolution algorithms in the above embodiments.

[0144] It is understood that the systems, devices, and storage media provided in the embodiments of the present invention correspond to the methods provided in the embodiments of the present invention, and the explanations, examples, and beneficial effects of the relevant content can be referred to the corresponding parts of the above methods.

[0145] In the above embodiments, implementation can be achieved entirely or partially through software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented entirely or partially as a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., solid state disk (SSD)).

[0146] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0147] The various embodiments in this specification are described in a related manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the system embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0148] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A dynamic path planning method based on scene prediction, characterized in that, Includes the following steps, S1. Construct a static vehicle routing planning model; S2. Establish the objective function for dynamic vehicle routing planning; S3. Calculate the degree of impact on potential customers; S4. Construct a predicted customer set based on the degree of influence of potential customers; S5. Solve the dynamic vehicle routing problem based on the predicted customer set; Applied to a warehouse n customers In a scenario consisting of m transport vehicles, the 1 warehouse and n customers constituting a node set , ,in, The customers are divided into known customers and potential customers, thus obtaining the set of known customers. and potential customer base ; Node set The straight-line paths between each node are used as the edge set. ;in, Represents any number of Nodes and the Nodes The straight path between; let the straight path be... The straight-line distance is denoted as Let the maximum load capacity of each transport vehicle be... Assume all vehicles are warehouses For the start and end points; Step S1 specifically includes, Step 1.1: Given a path planning time of T, and set T to... Given three time slices of equal length, define the index of any time slice as q, and initialize... =1; Step 1.2, the previous... All customers appearing in a given time slice are considered as the known customer set. , will the current number All customers appearing in a given time slot are considered as a new customer group. ; Step 1.3: Use equation (1) to establish the current number of... The objective function of the static vehicle path planning model in a time slice : (1) Equation (1) represents the total distance traveled by all vehicles. Indicates the first The first time slice The first transport vehicle The node and the first The straight-line distance between nodes Indicates the first Did the u-th car pass through the i-th time slice? The node and the first A straight path between nodes, if it passes through, then let 1. Otherwise, let ; Step 1.2: Construct the current number using equations (2)-(7). Constraints of the static vehicle routing model under each time slice: (2) (3) (4) (5) (6) (7) Equations (2) and (3) show that each customer must access the site once and only once; In equation (4), m is the total number of vehicles located in the warehouse. This indicates vehicles departing from the warehouse. Constraint (4) ensures that the number of dispatched vehicles should not exceed the total number of vehicles available in the warehouse. Equation (5) is the sub-loop constraint elimination. Represents a set The number of customers in the country; In equations (6) and (7) Indicates customer Equation (6) ensures that the total demand on each route does not exceed the vehicle's capacity. In step 2, the objective function for dynamic vehicle routing is established using equation (8). : (8)； Step 3 includes the known customer set. Add a potential customer Thus constituting the first l temporary customer group According to the current number A static vehicle routing model for a given time slice, for Solving for the problem yields the following results: Corresponding customer access path sequence Therefore, based on Calculate potential customers degree of influence ; Step 3.1: Calculate the capacity of the delivery vehicle using formula (9). : (9) In equation (9), Represents a known customer set The number of known customers in the database; Step 3.2, according to Using ant colony algorithm The customer set Solve the problem to obtain the customer access sequence. And calculate the length of the customer access sequence. Using ant colony optimization algorithm Solve the problem to obtain the customer access sequence. And calculate the length of the customer access sequence. ; Step 3.3: Calculate the degree of influence of potential customers according to formulas (10)-(14) based on the obtained customer visit sequence and the length of the customer visit sequence; (10) (11) (12) (13) (14) The function calculation method is as follows: [This refers to] potential customers. from Delete it, and get the deletion result. The customer access sequence, compared and The different numbers of customers in the same location Function representation The solution includes the number of customers; α and β are control parameters that adjust the importance of disturbances to travel costs and routes, in equation (14). This indicates the potential customers acquired through calculation. The extent of the impact.

2. The dynamic path planning method based on scene prediction according to claim 1, characterized in that: Step S4 includes step 4.2, calculating the number of customers that need to be added to the predicted customer set: (15) In formula (15) express The number of customers in the country; Step 4.3: Based on the degree of impact For potential customers Sort the customers in descending order; Step 4.4, from potential customers Before the election A customer is added to the known customer set. In this context, the predicted customer set is formed. At the same time from potential customers Delete the corresponding customer; Step 4.5: Optimize the access route for the predicted scenario based on the static path problem model. .

3. The dynamic path planning method based on scene prediction according to claim 2, characterized in that: Step S5 specifically includes, Step 5.2, the At the beginning of each time segment, new customers will emerge. Add to the currently known customer set ; Step 5.3: Determine the size of the predicted customer set. Did the expected quantity reach the target? (16) In formula (16) express The number of customers in the country; Step 5.4, if According to the degree of impact From potential customer set Select the customer with the greatest influence. Add to the predicted customer set and merge it into the new customer set. In, and from Delete; like According to the degree of impact From predicting customer sets Select the most influential potential customer from among them. Meanwhile, from predicting customer sets Delete Join the potential customer group ,exist Middle connection The previous customer and the next customer; Step 5.5, repeat step 5.4 until... ; Step 5.5: Use a greedy insertion method to add new customer sets. Insert into path superior; Step 5.6, Local Search: Step 5.7: Repeat steps 5.2 to 5.6 until the time slice ends, and return the solution of dynamic path planning.

4. The dynamic path planning method based on scene prediction according to claim 3, characterized in that: Step 5.6, Local Search, specifically including: Step 5.6.1: Define five local search operators Perform the following operation: 1) Randomly select routes One of the customers , randomly move its position; 2) Randomly select routes Two clients and Exchange customer locations; 3) Randomly select routes Two routes and ,from Randomly select a customer ,from Randomly select a customer ,exchange from after, from The route afterward; 4) Randomly select routes One of the routes ,from Two customers were randomly selected. and ,Will and The order of customers is reversed; 5) Randomly select routes The route in ,from Two customers were randomly selected. and ,Will and Customers in between are randomly inserted into other positions; Step 5.6.2: Randomly select a search operator for optimization, and continue optimizing until the time reaches the allocated time for the current scene. .

5. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the computer program is executed by the processor, it causes the processor to perform the steps of the method as described in any one of claims 1 to 4.

Images