Intelligent transportation vehicle path planning method and device, and electronic equipment

Abstract

The application provides a smart transport vehicle path planning method and device and electronic equipment, the method comprising: initializing a glowworm swarm optimization model based on transport vehicle related data; determining whether a set termination condition is reached, and outputting a smart transport vehicle path planning optimization result if the set termination condition is reached; otherwise, calculating the fitness value of each glowworm in the glowworm swarm, calculating the relative brightness and relative attractiveness of each glowworm based on the fitness value of each glowworm, updating the position of each glowworm based on the relative brightness and relative attractiveness of each glowworm in the glowworm swarm, and repeatedly calculating the fitness value of each glowworm in the glowworm swarm, calculating the relative brightness and relative attractiveness of each glowworm, and updating the position of each glowworm until the set termination condition is reached, and outputting the smart transport vehicle path planning optimization result, thereby improving the efficiency of transport and ensuring the economy of the transport process.

Description

Technical Field

[0001] This invention relates to the field of intelligent transportation technology, and in particular to an intelligent transportation vehicle route planning method, device, and electronic equipment. Background Technology

[0002] In today's transportation industry, route planning for transport vehicles is a critical issue. Traditional route planning methods are often based on pre-set routes or simple rules, which cannot adapt to complex road conditions and real-time delivery needs. This leads to problems such as low transportation efficiency, high delivery costs, and delivery delays.

[0003] Applying intelligent algorithms, such as genetic algorithms and simulated annealing, to vehicle routing problems has achieved some success. However, for complex traffic networks and dynamic road conditions, traditional methods often face problems such as high computational complexity, slow convergence speed, and susceptibility to getting trapped in local optima. Summary of the Invention

[0004] This invention provides an intelligent transportation vehicle route planning method, device, and electronic device to solve the shortcomings of existing technologies in vehicle route planning using intelligent algorithms, such as high computational complexity, slow convergence speed, and susceptibility to getting trapped in local optima. It achieves efficient solutions to problems related to complexity, dynamism, and multi-objective optimization in vehicle route planning.

[0005] This invention provides a method for intelligent transportation vehicle route planning, comprising:

[0006] The firefly algorithm model is initialized based on relevant data of transport vehicles;

[0007] Determine whether a set termination condition has been met. If the set termination condition has been met, output the intelligent transport vehicle route planning optimization result; otherwise, if the set termination condition has not been met, calculate the fitness value of each firefly in the firefly population, calculate the relative brightness and relative attractiveness of each firefly based on the fitness value of each firefly, update the position of each firefly based on the relative brightness and relative attractiveness of each firefly in the firefly population, and repeat the steps of calculating the fitness value of each firefly in the firefly population, calculating the relative brightness and relative attractiveness of each firefly, and updating the position of each firefly until the set termination condition is met, and output the intelligent transport vehicle route planning optimization result.

[0008] The constraints of the firefly algorithm model are determined based on the relevant data of the transport vehicle; each firefly in the firefly population represents an alternative path; and the fitness value of each firefly is determined based on the total transportation and delivery cost of the alternative path corresponding to the firefly.

[0009] According to a method for intelligent transport vehicle path planning provided by the present invention, updating the position of each firefly includes:

[0010] In h>R t In the case of h≤R, update the position of each firefly according to the Levy flight mode; t In the case of [the firefly], the position of each firefly is updated according to the cosine search method;

[0011] Where h represents a uniform random number in the range [0,1], and R t The adaptive parameters for the t-th iteration are determined according to the following formula:

[0012]

[0013] In the formula, R t The initial value is 0.5; It is the optimal fitness function value of the firefly population during the (t-1)th iteration; It is the optimal fitness function value of the firefly population during the (t-2)th iteration.

[0014] According to the intelligent transport vehicle path planning method provided by the present invention, updating the position of each firefly based on the Levy flight mode specifically includes:

[0015] The position of each firefly is updated according to the following formula:

[0016]

[0017] In the formula, y i,t+1 Let y be the position of the i-th firefly during the (t+1)-th iteration; i,t Let be the position of the i-th firefly during the t-th iteration; A represents the relative attraction between the i-th and j-th fireflies during the t-th iteration; β is the light intensity absorption coefficient; r ij Let y be the distance between the i-th firefly and the j-th firefly during the t-th iteration; j,t Let be the position of the i-th firefly during the t-th iteration; α represents the step size factor; rand is a random number uniformly distributed in the range [0,1]. This indicates dot product.

[0018] According to the intelligent transport vehicle route planning method provided by the present invention, the step size factor is determined according to the following formula:

[0019]

[0020] In the formula, the initial value of α is 0.5; g represents the number of iterations, G max This indicates the maximum number of iterations.

[0021] According to the intelligent transport vehicle route planning method provided by the present invention, updating the position of each firefly according to a cosine search method specifically includes:

[0022] The position of each firefly is updated according to the following formula:

[0023]

[0024] In the formula, y i,t+1 Let y be the position of the i-th firefly during the (t+1)-th iteration; i,t Let be the position of the i-th firefly during the t-th iteration; A represents the relative attraction between the i-th and j-th fireflies during the t-th iteration; β is the light intensity absorption coefficient; r ij Let y be the distance between the i-th firefly and the j-th firefly during the t-th iteration; j,t Let i be the position of the i-th firefly during the t-th iteration; represents dot product; b is a constant with a default value of 1; l is a d-dimensional uniform random vector of [-1, 1]*d.

[0025] According to the intelligent transport vehicle route planning method provided by the present invention, the initialization of the firefly algorithm model includes:

[0026] Initialize multiple fireflies, and generate chaotic sequences corresponding to the multiple fireflies based on the Chebyshev chaotic operator;

[0027] Map the chaotic sequences corresponding to the multiple fireflies into the solution space to obtain the positions of the multiple fireflies;

[0028] Based on the locations of the multiple fireflies, the fitness values ​​of the multiple fireflies are calculated respectively, and according to the fitness values ​​from high to low, the same number of fireflies as the number of client nodes are selected as the initial firefly population.

[0029] The present invention also provides an intelligent transport vehicle route planning device, comprising:

[0030] The initialization module is used to initialize the firefly algorithm model based on relevant data of transport vehicles;

[0031] The output module is used to determine whether a set termination condition has been met. If the set termination condition has been met, the module outputs the intelligent transport vehicle route planning optimization result. Otherwise, if the set termination condition has not been met, the module calculates the fitness value of each firefly in the firefly population, calculates the relative brightness and relative attractiveness of each firefly based on the fitness value of each firefly, updates the position of each firefly based on the relative brightness and relative attractiveness of each firefly in the firefly population, and repeats the steps of calculating the fitness value of each firefly in the firefly population, calculating the relative brightness and relative attractiveness of each firefly, and updating the position of each firefly until the set termination condition is met, and then outputs the intelligent transport vehicle route planning optimization result.

[0032] The constraints of the firefly algorithm model are determined based on the relevant data of the transport vehicle; each firefly in the firefly population represents an alternative path; and the fitness value of each firefly is determined based on the total transportation and delivery cost of the alternative path corresponding to the firefly.

[0033] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the intelligent transport vehicle path planning method as described above.

[0034] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the intelligent transport vehicle path planning method as described above.

[0035] The present invention also provides a computer program product, including a computer program that, when executed by a processor, implements the intelligent transport vehicle path planning method as described above.

[0036] The intelligent transport vehicle route planning method, device, and electronic device provided by this invention initialize a firefly algorithm model, the constraints of which are determined based on relevant transport vehicle data. Each firefly in the firefly population represents a candidate path. The fitness value of each firefly is determined based on the total transport and delivery cost of the candidate path corresponding to the firefly. Then, without reaching the set termination condition, the fitness value of each firefly is calculated, and then the relative brightness and relative attractiveness of each firefly are calculated. The position of each firefly is updated based on the relative brightness and relative attractiveness, and this process is iterated until the set termination condition is reached. At this point, the optimized intelligent transport vehicle route planning result can be output. This allows for full utilization of available transport resources, dynamic adjustment of route planning schemes, and cargo transport according to actual needs, thereby improving transport efficiency and ensuring the economy of the transport process. Attached Figure Description

[0037] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0038] Figure 1 A flowchart illustrating the intelligent transport vehicle route planning method provided by this invention;

[0039] Figure 2 A schematic diagram of the process for obtaining optimization results provided by the present invention;

[0040] Figure 3 A schematic diagram of the intelligent transport vehicle route planning device provided by the present invention;

[0041] Figure 4 This is a schematic diagram of the structure of the electronic device provided by the present invention. Detailed Implementation

[0042] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0043] Figure 1 This is a flowchart illustrating the intelligent transportation vehicle route planning method provided by the present invention, as shown below. Figure 1 As shown, the method includes the following steps:

[0044] Step 100: Initialize the firefly algorithm model based on the relevant data of the transport vehicles.

[0045] Step 101: Determine if the set termination condition has been met. If the set termination condition has been met, output the intelligent transport vehicle route planning optimization result; otherwise, if the set termination condition has not been met, calculate the fitness value of each firefly in the firefly population, calculate the relative brightness and relative attractiveness of each firefly based on the fitness value of each firefly, update the position of each firefly based on the relative brightness and relative attractiveness of each firefly in the firefly population, and repeat the steps of calculating the fitness value of each firefly in the firefly population, calculating the relative brightness and relative attractiveness of each firefly, and updating the position of each firefly until the set termination condition is met, and output the intelligent transport vehicle route planning optimization result.

[0046] The constraints of the firefly algorithm model are determined based on relevant data of the transport vehicles; each firefly in the firefly population represents an alternative path; and the fitness value of each firefly is determined based on the total transportation and delivery cost of the alternative path corresponding to the firefly.

[0047] Specifically, before planning the transportation vehicle routes, relevant data about the transportation vehicles can be collected. This data can include customer information, warehouse information, vehicle information, cargo origin, destination, cargo information, and time windows. Customer information can include the location of customer points (geographic coordinates), the total amount of cargo required by each customer point, the distance between customer points, and time window restrictions for reaching each customer point. Vehicle information includes the maximum distance a vehicle can travel on a full tank of fuel and its maximum capacity. Warehouse information includes the start and end times of warehouse operations.

[0048] In some implementations, after collecting data related to transport vehicles, the data can be cleaned and processed to prevent data loss and redundancy.

[0049] Based on relevant data about transport vehicles, the optimal path can be obtained through a multi-objective optimization model for transport vehicle route planning. This model can be defined as a complete directed graph A(V',E), where V' = V∪{v0} is a vertex set, and V = {V0,V1,……,V} m Let} represent a set of m customer nodes. v0 represents the departure point of the transport vehicle and its destination after completing the task. Each customer node i∈V is associated with each destination by the vehicle's arrival time, cargo quantity, maximum vehicle capacity, and time window. Each edge E represents the corresponding vehicle path. Path Z represents the sequence of nodes from the warehouse, visiting customer nodes in sequence, completing the task, and finally returning to the warehouse.

[0050] Optionally, the constraints of the multi-objective optimization model for transport vehicle route planning may include one or more of the following:

[0051] (1) Vehicle capacity constraint: The cargo carried by a transport vehicle shall not exceed the maximum capacity of the transport vehicle.

[0052] (2) Departure time constraint: For warehouses, the departure time of transport vehicles is zero.

[0053] (3) Return time constraint: The transport vehicle must return to the warehouse before the warehouse closes.

[0054] (4) Time window constraint: If the customer point v is reached ahead of schedule... iOr delay in reaching the customer point. i The transport vehicles will be punished.

[0055] (5) Vehicle type constraint: The same type of transport vehicle must be used when performing the same task.

[0056] (6) Single access constraint: The same customer point can only be accessed by one transport vehicle.

[0057] Based on the above constraints, in this embodiment of the invention, minimizing the total cost of transportation and distribution can be taken as the optimization objective, resulting in the following objective function:

[0058]

[0059] The objective function is divided into three parts, the first part... This represents the transportation cost of the vehicle, where C represents the transportation cost per unit distance, and d... ij The distance between customer node i and customer node j is represented by n, the number of vehicles is represented by m, and the number of customer nodes is represented by m. (Part Two) B represents the fixed cost, H represents the maximum value of the vehicle number, and sign(N) represents the fixed cost. h This indicates whether vehicle number h is dispatched; if dispatched, its fixed costs will be calculated. When N... h >0, sign(N) h The value is 1 when N h =0, then sign(N) h ) = 0; Part Three S represents the time penalty cost of the vehicle. i This refers to the actual time when the delivery vehicle arrives at customer node i.

[0060] The remaining parameters in the objective function are explained below:

[0061]

[0062] That is, if delivery vehicle h travels from customer point i to customer point j, then x h ij Assign a value of 1 if the value is 1, otherwise assign a value of 0.

[0063]

[0064] ET i LT represents the earliest time the delivery vehicle arrives at customer node i; i S represents the latest time that the delivery vehicle arrives at customer node i; i This refers to the actual time when the delivery vehicle arrives at customer node i. [ET] i ,LTi Let [i] be the time window for customer node i. Let l represent the cost of arriving early per unit time; t represent the penalty cost of arriving late per unit time. In summary, the penalty cost for the vehicle routing problem based on time windows is P. i (S i )=l*max(ET i -S i ,0)+t*max(S i -LT i ,0).

[0065] For each constraint condition, the formula can be as follows:

[0066]

[0067] This formula ensures that vehicles depart from the distribution center, complete their delivery tasks, and return to the distribution center, forming a complete loop.

[0068]

[0069]

[0070] The above two equations represent the relationship between 0-1 decision variables and x and y.

[0071]

[0072] This formula indicates that the distribution center dispatched a total of n vehicles during the entire delivery mission.

[0073]

[0074] This formula indicates that each customer node has one and only one delivery vehicle to complete the delivery task.

[0075]

[0076] This expression means that if the delivery task of customer node i is completed by vehicle number h, the value is 1; otherwise, it is 0.

[0077]

[0078] This formula indicates that the maximum carrying weight of vehicle number h for each delivery cannot exceed W, where W represents the maximum weight that each delivery vehicle can carry.

[0079]

[0080] This formula indicates that the delivery vehicle numbered h will not travel more than the maximum distance D when completing a single task.

[0081] In this embodiment of the invention, the multi-objective optimization model for transport vehicle route planning is implemented using the firefly algorithm. In the firefly algorithm, it is assumed that fireflies are genderless, and no firefly is attracted to other fireflies based on its gender. In short, all fireflies are considered neutral; for any firefly, brighter fireflies will attract brighter ones, and the brightest firefly will move randomly. That is, the attractiveness of fireflies is directly proportional to their brightness; the brightness of fireflies is influenced or determined by the terrain, which is the objective function. For maximization problems, brightness can simply be proportional to the value of the objective function.

[0082] Therefore, the firefly algorithm model can be initialized first. During initialization, the initial firefly population and the position of each firefly within it can be determined. Each firefly in the initial population represents an alternative path in the transportation vehicle route planning scheme. It is understood that the constraints and fitness function in the firefly algorithm model can be determined through the constraints and objective function described above. The fitness value of the fireflies represents the total cost of vehicle transportation; for the method provided in this embodiment, a smaller fitness value is better.

[0083] Brightness and attraction are two crucial parameters in the firefly algorithm. Brightness reflects the quality of a firefly's location and determines its movement direction, while attraction determines the distance it travels. In the method provided in this embodiment, it is necessary to determine whether the firefly population has reached a set termination condition. If the set termination condition is reached, the optimized path planning result for the intelligent transport vehicle is output. The set termination condition can be that the number of iterations reaches a preset value.

[0084] If the set termination condition is not met, the fitness value of each firefly in the firefly population can be calculated, and based on the fitness value of each firefly, the relative brightness and relative attractiveness of each firefly can be calculated. Based on the relative brightness and relative attractiveness of each firefly in the firefly population, the position of each firefly can be updated.

[0085] Then, the fitness value of each firefly in the firefly population is repeatedly calculated, as well as the relative brightness and relative attractiveness of each firefly, and the position of each firefly is updated, until a set termination condition is met, at which point the intelligent transport vehicle path planning optimization result is output. By continuously updating the brightness and attractiveness, and subsequently updating the position of each firefly, multi-objective vehicle path optimization can be achieved.

[0086] From a mathematical perspective, the main parameters of the firefly algorithm can be described as follows:

[0087] The definition of relative brightness B is:

[0088]

[0089] Where B0 represents the maximum brightness of the firefly and is related to the objective function value; the better the objective function value, the higher the brightness. β is the light absorption coefficient, and r is the distance between fireflies. Generally, a bright firefly will attract a less bright one, but light intensity is not the only factor in this mechanism. The brightness of a firefly gradually decreases with increasing distance and absorption by the propagation medium.

[0090] The definition of relative attraction A is:

[0091]

[0092] Where A0 is the maximum attraction; β is the light intensity absorption coefficient; and r is the distance between fireflies.

[0093] The distance r between fireflies can be calculated using Descartes' formula.

[0094]

[0095] Where, r ij Y is the distance between fireflies i and j; i,z It is the z-th component of the spatial coordinate Y of the i-th firefly, Y j,p d is the z-th component of the spatial coordinate Y of the j-th firefly, and d is the dimension of the spatial coordinate Y.

[0096] In some implementations, the firefly's position can be updated using the following formula:

[0097]

[0098] Among them, y i,t+1 Let y be the position of the i-th firefly during the (t+1)-th iteration; i,t Let be the position of the i-th firefly during the t-th iteration; A represents the relative attraction between the i-th and j-th fireflies during the t-th iteration; β is the light intensity absorption coefficient; r ij Let y be the distance between the i-th firefly and the j-th firefly during the t-th iteration; j,t Let be the position of the i-th firefly during the t-th iteration; α represents the step size factor, which generally takes values ​​in the range [0,1].

[0099] The intelligent transport vehicle route planning method provided by this invention initializes a firefly algorithm model, the constraints of which are determined based on relevant transport vehicle data. Each firefly in the firefly population represents a candidate path. The fitness value of each firefly is determined based on the total transport and delivery cost of the candidate path corresponding to the firefly. Then, without reaching the set termination condition, the fitness value of each firefly is calculated, followed by the calculation of its relative brightness and relative attractiveness. The position of each firefly is then updated based on its relative brightness and relative attractiveness. This process is iterated until the set termination condition is reached, at which point the optimized intelligent transport vehicle route planning result can be output. This allows for full utilization of available transport resources, dynamic adjustment of route planning schemes, and cargo transport based on actual needs, thereby improving transport efficiency and ensuring the economy of the transport process.

[0100] Optionally, based on updating the location of each firefly, including:

[0101] In h>R t In the case of h≤R, update the position of each firefly according to the Levy flight pattern; t In this case, the position of each firefly is updated according to the cosine search method.

[0102] Specifically, in this embodiment of the invention, fireflies can be made to fly using two models: Levy flight mode and cosine search method, thereby enhancing their flight capabilities and local space exploration capabilities.

[0103] To balance these two modes, embodiments of the present invention propose an adaptive approach to determine the superior mode in the current iteration, namely:

[0104]

[0105] Where h represents a uniform random number in the range [0,1], and R t This is calculated in the last iteration. To speed up the convergence of the computation, the global search mode needs to be selected with a higher probability compared to the local search mode. Therefore, a switch R can be defined. t The range is [0.5, 1]. R t The initial value is set to 0.5, and R is in the t-th iteration. t Determined according to the following formula:

[0106]

[0107] In the formula, R t The initial value is 0.5; It is the optimal fitness function value of the firefly population during the (t-1)th iteration; It is the optimal fitness function value of the firefly population during the (t-2)th iteration.

[0108] Many heuristic algorithms get stuck in local optima to varying degrees, affecting their accuracy in solving certain problems. To address this issue, this invention introduces the Levy flight strategy into the firefly algorithm model.

[0109] The Levy flight strategy, by introducing the randomness of the Levy distribution, provides a strategy for global search and stochastic simulation, as shown in the following formula:

[0110] Levy~ε=t -λ (1<λ<3)

[0111] Wherein, λ represents the stability parameter, which is usually taken as 1.5.

[0112]

[0113] Where η and ε conform to a standard normal distribution, and The calculation method is as follows.

[0114]

[0115] Where γ is the standard gamma function and λ takes the value 1.5.

[0116] To make the firefly position update more random, this embodiment of the invention optimizes the firefly position update formula.

[0117] Optionally, the location of each firefly is updated based on the Levy flight pattern, specifically including:

[0118] The position of each firefly is updated according to the following formula:

[0119]

[0120] In the formula, y i,t+1 Let y be the position of the i-th firefly during the (t+1)-th iteration; i,t Let be the position of the i-th firefly during the t-th iteration; A represents the relative attraction between the i-th and j-th fireflies during the t-th iteration; β is the light intensity absorption coefficient; r ij Let y be the distance between the i-th firefly and the j-th firefly during the t-th iteration; j,tLet be the position of the i-th firefly during the t-th iteration; α represents the step size factor; rand is a random number uniformly distributed in the range [0,1]; α·sign(rand-0.5) represents the perturbation term. If the brightest firefly is not attracted by other fireflies, its position update will be changed by random perturbation. This can prevent the algorithm from getting trapped in local optima too early. This indicates dot product.

[0121] In the firefly algorithm, the step size of the firefly's movement significantly impacts the optimization performance. If the step size is too large, the firefly may travel a distance greater than the distance between it and the optimal value, causing the algorithm to skip the optimal solution. Conversely, if the step size is too small, the firefly's movement speed and convergence speed become extremely slow, increasing the convergence time. Therefore, the step size setting in the firefly algorithm is crucial to its performance. This invention proposes a novel adaptive step size strategy, and by adaptively changing the step size, the algorithm's performance can be effectively improved.

[0122] Optionally, the step size factor is determined according to the following formula:

[0123]

[0124] In the formula, the initial value of α is 0.5; g represents the number of iterations, G max This indicates the maximum number of iterations.

[0125] Optionally, the position of each firefly is updated according to the cosine search method, specifically including:

[0126] The position of each firefly is updated according to the following formula:

[0127]

[0128] In the formula, y i,t+1 Let y be the position of the i-th firefly during the (t+1)-th iteration; i,t Let be the position of the i-th firefly during the t-th iteration; A represents the relative attraction between the i-th and j-th fireflies during the t-th iteration; β is the light intensity absorption coefficient; r ij Let y be the distance between the i-th firefly and the j-th firefly during the t-th iteration; j,t Let i be the position of the i-th firefly during the t-th iteration; represents dot product; b is a constant with a default value of 1; l is a d-dimensional uniform random vector of [-1, 1]*d.

[0129] Optionally, the firefly algorithm model is initialized, including:

[0130] Initialize multiple fireflies and generate chaotic sequences corresponding to the multiple fireflies based on the Chebyshev chaotic operator;

[0131] Map the chaotic sequences corresponding to multiple fireflies into the solution space to obtain the positions of the multiple fireflies;

[0132] Based on the locations of multiple fireflies, calculate the fitness value of each firefly, and select the same number of fireflies as the number of client nodes according to the fitness value from high to low as the initial firefly population.

[0133] Specifically, to better facilitate the understanding and resolution of this problem, the embodiments of this invention can employ real number encoding. Assuming the firefly population size is N, the total number of client nodes is M, and the optimized path scheme corresponding to each firefly contains m client nodes, where m <= M, then the firefly population (path) Z can be represented by the following formula:

[0134]

[0135] Where Z represents the firefly population, z n,m This represents the code of the m-th client node selected by the n-th firefly.

[0136] Since the initial firefly population does not consider the influence of other factors, and in order to make the initial position of the fireflies more random and enrich the diversity of the population, this embodiment of the invention uses the randomness of the Chebyshev chaotic sequence to initialize the population.

[0137] First, in a D-dimensional space, suppose there are N fireflies. For the first firefly, a D-dimensional vector Y can be randomly generated, where the value of each dimension ranges from -1 to 1. This vector is denoted as Y = (y1, y2, ..., y...). D ), and each y i This indicates the initial information of the first firefly.

[0138] Then, the chaotic sequence corresponding to the first firefly can be obtained by iterating N-1 times on each dimension of the Y vector using the following formula (i.e., the Chebyshev chaotic operator).

[0139]

[0140] Where ξ represents the chaos factor, and its value is 4.

[0141] By analogy, the chaotic sequences corresponding to the other N-1 fireflies can be obtained.

[0142] The following formula can be used to map these N individuals that have experienced a chaotic process into the solution space:

[0143] x id =ld+(1+y id )*(ud-ld) / 2

[0144] The d-th dimension in the solution space has an upper bound and a lower bound, denoted by ud and ld, respectively. Based on this formula, the value of the i-th firefly in the d-th dimension can be calculated, denoted as y. id After mapping, we can obtain the coordinates of the i-th firefly in the d-th dimension, denoted as x. id .

[0145] Finally, for N fireflies, their fitness functions can be calculated separately, they can be sorted, and M individuals with high fitness (i.e., the number of client nodes) can be selected as members of the initial firefly population.

[0146] The following examples, using specific application scenarios, further illustrate the intelligent transportation vehicle route planning method provided in this application.

[0147] The transportation vehicle route planning method provided in this embodiment includes the following steps:

[0148] Step 1: Collect real-time data, including customer information, warehouse information, vehicle information, cargo origin, destination, cargo information, time window, etc.

[0149] Step 2: Preprocess and clean the collected data to ensure its accuracy and integrity.

[0150] Step 3: Model the intelligent transportation vehicle routing problem.

[0151] Step 4: Based on the processed data, apply an appropriate path planning algorithm to generate a preliminary path plan.

[0152] Step 5: Optimize and schedule. Figure 2 This is a schematic diagram of the process for obtaining optimization results provided by the present invention, such as... Figure 2 As shown, this embodiment utilizes the Levy flight strategy-based chaotic adaptive firefly algorithm to optimize the initial path scheme, considering multiple objectives such as the shortest path and minimum time window violation, to improve the efficiency and accuracy of the path; and makes reasonable scheduling arrangements based on the actual situation of vehicles and goods to avoid resource waste and conflicts.

[0153] The main step 5 can be broken down in detail as follows:

[0154] Step 5.1: Encode the position X of each firefly in the chaotic adaptive firefly algorithm based on the Levy flight strategy into an integer according to the number of client nodes M. The length of X is M.

[0155] Step 5.2 initializes the parameters of the Levy-based Chaotic Adaptive Firefly Algorithm (LCAFA), setting the firefly population size to N and the number of iterations to G. Additionally, it initializes the firefly positions, where each firefly represents a candidate path. Through path encoding simplification, the firefly position can be represented by a vector, where each dimension of the vector represents the ordinate of a path point on a candidate path. To achieve this, the initial firefly positions can be initialized within the search space using the Chebyshev chaotic operator. If m client points need to be determined, the firefly position vector will be an m-dimensional vector.

[0156] Step 5.2 is broken down as follows:

[0157] Step 5.2.1: In a space with D dimensions, suppose there are N fireflies. Next, we randomly generate a D-dimensional vector Y, where the value of each dimension ranges from -1 to 1. We denote this vector as Y = (y1, y2, ..., yD), and each y... i This indicates the initial information of the first firefly.

[0158] Step 5.2.2: Using the following formula, iterate N-1 times on each dimension of the Y vector to obtain information about the other N-1 fireflies.

[0159]

[0160] Here, ξ represents the chaos factor. When ξ is greater than or equal to 2, regardless of the initial value, the sequences after chaotic iteration will not be correlated. This embodiment selects the case where ξ equals 4. By using the formula, a set of chaotic sequences is generated. These sequences are uncorrelated with each other, but all variables exhibit ergodic properties.

[0161] Step 5.2.3: According to the following formula, this embodiment maps N individuals that have experienced chaotic processes into the solution space.

[0162] x id =ld+(1+y id )*(ud-ld) / 2

[0163] The d-th dimension in the solution space has an upper bound and a lower bound, denoted by ud and ld, respectively. Based on the above formula, the value of the i-th firefly in the d-th dimension can be calculated, denoted as y. id After mapping, the coordinates of the i-th firefly in the d-th dimension are obtained, denoted as x. id .

[0164] Step 5.2.4: For N fireflies, calculate their fitness functions and sort them. Finally, select M individuals with high fitness as members of the initial population.

[0165] Step 5.3: Obtain the brightness of the firefly and update its position to find a solution space that is close to the optimal solution.

[0166] In the firefly algorithm, fireflies communicate by emitting light, and brightness indicates the quality of the solution represented by each firefly. Brighter fireflies attract dimmer fireflies, causing the entire population to move towards better solutions. In other words, brighter fireflies act as leaders, attracting other fireflies with stronger light signals and guiding the population towards more promising solutions. This process of information sharing and concentrated movement allows the algorithm to gradually adjust the positions of solutions to find the optimal solution or a solution space close to it.

[0167] The specific method for updating firefly locations is as follows:

[0168] First, the position vector of each firefly is applied sequentially to the objective function described above. Then, the merits of the paths are determined by evaluating these objective functions, and the feasibility of each path is judged. If a path is infeasible, this embodiment applies a penalty. This process yields the objective function vector for each firefly, enabling a comprehensive evaluation of the solution for each firefly across multiple objectives, thus helping to find a better solution.

[0169] This embodiment improves the firefly position update by proposing two modes: the Levy flight strategy mode and the cosine search mode, and designs an adaptive method to balance these two modes.

[0170]

[0171] Where h represents a uniform random number in [0,1], and l is a d-dimensional uniform random vector in [-1,1]*d.

[0172]

[0173] Where R t This is calculated in the last iteration. To accelerate the convergence speed, the cosine search mode needs to be selected with a higher probability compared to the Levy search mode. Therefore, this embodiment defines R... t The range is [0.5, 1], and its initial value is set to 0.5. This yields R. t The values ​​for the adaptive mode are shown below:

[0174]

[0175] in, It is the optimal fitness function value of the firefly population during the (t-1)th iteration; This is the optimal fitness function value for the firefly population during the (t-2)th iteration. The adaptive scaling parameter threshold θ is calculated as follows:

[0176]

[0177] To update the step size of the firefly, this embodiment designs an adaptive firefly step size formula:

[0178]

[0179] In the formula, the initial value of α is 0.5; g represents the number of iterations, G max This indicates the maximum number of iterations.

[0180] Step 5.4, determine whether the preset maximum number of iterations G has been reached:

[0181] If the maximum number of iterations G set in step 5.2 has been reached, proceed to step 6; otherwise, continue with step 5.3.

[0182] Step 6: Obtain the optimal path considering path length constraints, time window constraints, and vehicle capacity constraints; path planning is now complete.

[0183] In the vehicle routing problem, this embodiment of the invention uses four different optimization algorithms: Simulated Annealing (SA), Genetic Algorithm (GA), Hybrid Particle Swarm Optimization (GA-PSO), and LCAFA. This embodiment compares the routing of transport vehicles with 10, 17, and 35 customer points, respectively. Through comparison and simulation experiments, this embodiment evaluates their performance in solving the vehicle routing problem.

[0184] For 10 customer points, among the optimized paths provided by different algorithms, SA, GA, GA-PSO, and LCAFA dispatched 3, 4, 3, and 3 transport vehicles respectively, completing the tasks over total distances of 217.071km, 239.602km, 219.202km, and 204.653km. This shows that LCAFA's optimization effect is better. Although SA, GA-PSO, and LCAFA all dispatched only 3 transport vehicles, the total distance traveled by LCAFA to complete the task was 12.418km less than SA and 14.549km less than GA-PSO.

[0185] For 17 customer points, among the optimized paths provided by different algorithms, SA, GA, GA-PSO, and LCAFA dispatched 8, 8, 8, and 7 transport vehicles respectively for the 11 customer point paths optimized by them, with total distances of 482.5693km, 498.2376km, 490.1634km, and 451.5854km respectively. LCAFA not only dispatched fewer transport vehicles than other algorithms, but its total distance traveled was also 30.9839km, 46.6522km, and 38.578km less than SA, GA, and GA-PSO respectively.

[0186] For 35 customer points, among the optimized paths provided by different algorithms, SA, GA, GA-PSO, and LCAFA dispatched 13, 13, 13, and 9 transport vehicles respectively, with total distances of 482.5693km, 498.2376km, 490.1634km, and 451.5854km. LCAFA not only dispatched fewer transport vehicles than other algorithms, but its total distance traveled was also 139.8321km, 87.7192km, and 153.9443km less than SA, GA, and GA-PSO respectively.

[0187] The data analysis above shows that LCAFA optimizes transport vehicle routes better with an increase in customer points. While completing the same task, it not only effectively saves vehicle resources but also reduces the distance vehicles travel. Therefore, LCAFA has promising applications in intelligent transport vehicle route planning and can effectively conserve resources.

[0188] The intelligent transport vehicle route planning device provided by the present invention is described below. The intelligent transport vehicle route planning device described below and the intelligent transport vehicle route planning method described above can be referred to in correspondence.

[0189] Figure 3 This is a schematic diagram of the intelligent transport vehicle route planning device provided by the present invention, as shown below. Figure 3 As shown, the device includes:

[0190] Initialization module 300 is used to initialize the firefly algorithm model based on relevant data of transport vehicles;

[0191] Output module 310 is used to determine whether the set termination condition has been met. If the set termination condition has been met, it outputs the intelligent transport vehicle route planning optimization result; otherwise, if the set termination condition has not been met, it calculates the fitness value of each firefly in the firefly population, calculates the relative brightness and relative attractiveness of each firefly based on the fitness value of each firefly, updates the position of each firefly based on the relative brightness and relative attractiveness of each firefly in the firefly population, and repeats the steps of calculating the fitness value of each firefly in the firefly population, calculating the relative brightness and relative attractiveness of each firefly, and updating the position of each firefly until the set termination condition is met, and outputs the intelligent transport vehicle route planning optimization result.

[0192] The constraints of the firefly algorithm model are determined based on relevant data of the transport vehicles; each firefly in the firefly population represents an alternative path; and the fitness value of each firefly is determined based on the total transportation and delivery cost of the alternative path corresponding to the firefly.

[0193] Optionally, update the location of each firefly, including:

[0194] In h>R t In the case of h≤R, update the position of each firefly according to the Levy flight pattern; t In this case, update the position of each firefly according to the cosine search method;

[0195] Where h represents a uniform random number in the range [0,1], and R t The adaptive parameters for the t-th iteration are determined according to the following formula:

[0196]

[0197] In the formula, R t The initial value is 0.5; It is the optimal fitness function value of the firefly population during the (t-1)th iteration; It is the optimal fitness function value of the firefly population during the (t-2)th iteration.

[0198] Optionally, the location of each firefly is updated based on the Levy flight pattern, specifically including:

[0199] The position of each firefly is updated according to the following formula:

[0200]

[0201] In the formula, y i,t+1 Let y be the position of the i-th firefly during the (t+1)-th iteration; i,tLet be the position of the i-th firefly during the t-th iteration; A represents the relative attraction between the i-th and j-th fireflies during the t-th iteration; β is the light intensity absorption coefficient; r ij Let y be the distance between the i-th firefly and the j-th firefly during the t-th iteration; j,t Let be the position of the i-th firefly during the t-th iteration; α represents the step size factor; rand is a random number uniformly distributed in the range [0,1]. This indicates dot product.

[0202] Optionally, the step size factor is determined according to the following formula:

[0203]

[0204] In the formula, the initial value of α is 0.5; g represents the number of iterations, G max This indicates the maximum number of iterations.

[0205] Optionally, the position of each firefly is updated according to a cosine search method, specifically including:

[0206] The position of each firefly is updated according to the following formula:

[0207]

[0208] In the formula, y i,t+1 Let y be the position of the i-th firefly during the (t+1)-th iteration; i,t Let be the position of the i-th firefly during the t-th iteration; A represents the relative attraction between the i-th and j-th fireflies during the t-th iteration; β is the light intensity absorption coefficient; r ij Let y be the distance between the i-th firefly and the j-th firefly during the t-th iteration; j,t Let i be the position of the i-th firefly during the t-th iteration; represents dot product; b is a constant with a default value of 1; l is a d-dimensional uniform random vector of [-1, 1]*d.

[0209] Optionally, the firefly algorithm model is initialized, including:

[0210] Initialize multiple fireflies and generate chaotic sequences corresponding to the multiple fireflies based on the Chebyshev chaotic operator;

[0211] Map the chaotic sequences corresponding to multiple fireflies into the solution space to obtain the positions of the multiple fireflies;

[0212] Based on the locations of multiple fireflies, calculate the fitness value of each firefly, and select the same number of fireflies as the number of client nodes according to the fitness value from high to low as the initial firefly population.

[0213] Figure 4 This is a schematic diagram of the structure of the electronic device provided by the present invention, such as... Figure 4 As shown, the electronic device may include: a processor 410, a communication interface 420, a memory 430, and a communication bus 440, wherein the processor 410, the communication interface 420, and the memory 430 communicate with each other via the communication bus 440. The processor 410 can call logical instructions in the memory 430 to execute the intelligent transportation vehicle path planning method provided by the above methods, which includes:

[0214] The firefly algorithm model is initialized based on relevant data of transport vehicles;

[0215] Determine whether the set termination condition has been met. If the set termination condition has been met, output the intelligent transport vehicle route planning optimization result; otherwise, if the set termination condition has not been met, calculate the fitness value of each firefly in the firefly population, calculate the relative brightness and relative attractiveness of each firefly based on the fitness value of each firefly, update the position of each firefly based on the relative brightness and relative attractiveness of each firefly in the firefly population, and repeat the steps of calculating the fitness value of each firefly in the firefly population, calculating the relative brightness and relative attractiveness of each firefly, and updating the position of each firefly until the set termination condition is met, and output the intelligent transport vehicle route planning optimization result.

[0216] The constraints of the firefly algorithm model are determined based on relevant data of the transport vehicles; each firefly in the firefly population represents an alternative path; and the fitness value of each firefly is determined based on the total transportation and delivery cost of the alternative path corresponding to the firefly.

[0217] Furthermore, the logical instructions in the aforementioned memory 430 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, essentially, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0218] On the other hand, the present invention also provides a computer program product, the computer program product comprising a computer program that can be stored on a non-transitory computer-readable storage medium, wherein when the computer program is executed by a processor, the computer is able to execute the intelligent transport vehicle route planning method provided by the above methods, the method comprising:

[0219] The firefly algorithm model is initialized based on relevant data of transport vehicles;

[0220] Determine whether the set termination condition has been met. If the set termination condition has been met, output the intelligent transport vehicle route planning optimization result; otherwise, if the set termination condition has not been met, calculate the fitness value of each firefly in the firefly population, calculate the relative brightness and relative attractiveness of each firefly based on the fitness value of each firefly, update the position of each firefly based on the relative brightness and relative attractiveness of each firefly in the firefly population, and repeat the steps of calculating the fitness value of each firefly in the firefly population, calculating the relative brightness and relative attractiveness of each firefly, and updating the position of each firefly until the set termination condition is met, and output the intelligent transport vehicle route planning optimization result.

[0221] The constraints of the firefly algorithm model are determined based on relevant data of the transport vehicles; each firefly in the firefly population represents an alternative path; and the fitness value of each firefly is determined based on the total transportation and delivery cost of the alternative path corresponding to the firefly.

[0222] In another aspect, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to perform the intelligent transport vehicle path planning method provided by the methods described above, the method comprising:

[0223] The firefly algorithm model is initialized based on relevant data of transport vehicles;

[0224] Determine whether the set termination condition has been met. If the set termination condition has been met, output the intelligent transport vehicle route planning optimization result; otherwise, if the set termination condition has not been met, calculate the fitness value of each firefly in the firefly population, calculate the relative brightness and relative attractiveness of each firefly based on the fitness value of each firefly, update the position of each firefly based on the relative brightness and relative attractiveness of each firefly in the firefly population, and repeat the steps of calculating the fitness value of each firefly in the firefly population, calculating the relative brightness and relative attractiveness of each firefly, and updating the position of each firefly until the set termination condition is met, and output the intelligent transport vehicle route planning optimization result.

[0225] The constraints of the firefly algorithm model are determined based on relevant data of the transport vehicles; each firefly in the firefly population represents an alternative path; and the fitness value of each firefly is determined based on the total transportation and delivery cost of the alternative path corresponding to the firefly.

[0226] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0227] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0228] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; 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; and these 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 method for intelligent transport vehicle route planning, characterized in that, include: The firefly algorithm model is initialized based on relevant data of transport vehicles; Determine whether a set termination condition has been met. If the set termination condition has been met, output the intelligent transport vehicle route planning optimization result; otherwise, if the set termination condition has not been met, calculate the fitness value of each firefly in the firefly population, calculate the relative brightness and relative attractiveness of each firefly based on the fitness value of each firefly, update the position of each firefly based on the relative brightness and relative attractiveness of each firefly in the firefly population, and repeat the steps of calculating the fitness value of each firefly in the firefly population, calculating the relative brightness and relative attractiveness of each firefly, and updating the position of each firefly until the set termination condition is met, and output the intelligent transport vehicle route planning optimization result. The constraints of the firefly algorithm model are determined based on the relevant data of the transport vehicle; each firefly in the firefly population represents an alternative path; the fitness value of each firefly is determined based on the total transportation and delivery cost of the alternative path corresponding to the firefly. The feature is that updating the position of each firefly includes: exist In this case, the position of each firefly is updated according to the Levy flight mode; In the case of [the firefly], the position of each firefly is updated according to the cosine search method; in, h Represents a uniformly random number in the range [0,1]. R t For the first t The adaptive parameters in the iteration process are determined according to the following formula: ； In the formula, R t The initial value is 0.5; ; It is the first t The optimal fitness function value of the firefly population during the -1st iteration; It is the first t - The optimal fitness function value of the firefly population during the 2nd iteration; The step of updating the position of each firefly according to the cosine search method specifically includes: The position of each firefly is updated according to the following formula: ； In the formula, For the first t During the +1st iteration, the... i The location of the fireflies; For the first t During the nth iteration i The location of the fireflies; A Indicates the first t During the nth iteration i Only fireflies and the first j The relative attraction between individual fireflies; β The light intensity absorption coefficient; r ij For the first t During the nth iteration i Only fireflies and the first j The distance between fireflies; For the first t During the nth iteration j The location of the fireflies; Dot product; b It is a constant with a default value of 1; l It is a d-dimensional uniform random vector of [-1, 1]*d; The initialization of the firefly algorithm model includes: Initialize multiple fireflies, and generate chaotic sequences corresponding to the multiple fireflies based on the Chebyshev chaotic operator; Map the chaotic sequences corresponding to the multiple fireflies into the solution space to obtain the positions of the multiple fireflies; Based on the locations of the multiple fireflies, the fitness values ​​of the multiple fireflies are calculated respectively, and according to the fitness values ​​from high to low, the same number of fireflies as the number of client nodes are selected as the initial firefly population.

2. The intelligent transport vehicle route planning method according to claim 1, characterized in that, The step of updating the position of each firefly according to the Levy flight pattern specifically includes: The position of each firefly is updated according to the following formula: ； In the formula, For the first t During the +1st iteration, the... i The location of the fireflies; For the first t During the nth iteration i The location of the fireflies; A Indicates the first t During the nth iteration i Only fireflies and the first j The relative attraction between individual fireflies; β The light intensity absorption coefficient; r ij For the first t During the nth iteration i Only fireflies and the first j The distance between fireflies; For the first t During the nth iteration j The location of the fireflies; α Indicates the step size factor; rand A random number uniformly distributed in the range [0,1]. Dot product; Let represent a random step size vector generated by the Lévy flight probability distribution, wherein the random step size vector follows a heavy-tailed distribution.

3. The intelligent transport vehicle route planning method according to claim 2, characterized in that, The step size factor is determined according to the following formula: ； In the formula, The initial value is 0.5; Indicates the number of iterations. This indicates the maximum number of iterations.

4. An intelligent transport vehicle route planning device, characterized in that, include: The initialization module is used to initialize the firefly algorithm model based on relevant data of transport vehicles; The output module is used to determine whether the set termination conditions have been met. If the set termination conditions have been met, the intelligent transportation vehicle route planning optimization results are output. Otherwise, if the set termination condition is not met, calculate the fitness value of each firefly in the firefly population, calculate the relative brightness and relative attractiveness of each firefly based on the fitness value of each firefly, update the position of each firefly based on the relative brightness and relative attractiveness of each firefly in the firefly population, and repeat the steps of calculating the fitness value of each firefly in the firefly population, calculating the relative brightness and relative attractiveness of each firefly, and updating the position of each firefly until the set termination condition is met, and output the intelligent transport vehicle route planning optimization result; The constraints of the firefly algorithm model are determined based on the relevant data of the transport vehicle; each firefly in the firefly population represents an alternative path; the fitness value of each firefly is determined based on the total transportation and delivery cost of the alternative path corresponding to the firefly. The updating of the position of each firefly includes: exist In this case, the position of each firefly is updated according to the Levy flight mode; In the case of [the firefly], the position of each firefly is updated according to the cosine search method; in, h Represents a uniformly random number in the range [0,1]. R t For the first t The adaptive parameters in the iteration process are determined according to the following formula: ； In the formula, R t The initial value is 0.5; ; It is the first t The optimal fitness function value of the firefly population during the -1st iteration; It is the first t - The optimal fitness function value of the firefly population during the 2nd iteration; The step of updating the position of each firefly according to the cosine search method specifically includes: The position of each firefly is updated according to the following formula: ； In the formula, For the first t During the +1st iteration, the... i The location of the fireflies; For the first t During the nth iteration i The location of the fireflies; A Indicates the first t During the nth iteration i Only fireflies and the first j The relative attraction between individual fireflies; β The light intensity absorption coefficient; r ij For the first t During the nth iteration i Only fireflies and the first j The distance between fireflies; For the first t During the nth iteration i The location of the fireflies; Dot product; b It is a constant with a default value of 1; l It is a d-dimensional uniform random vector of [-1, 1]*d; The initialization of the firefly algorithm model includes: Initialize multiple fireflies, and generate chaotic sequences corresponding to the multiple fireflies based on the Chebyshev chaotic operator; Map the chaotic sequences corresponding to the multiple fireflies into the solution space to obtain the positions of the multiple fireflies; Based on the locations of the multiple fireflies, the fitness values ​​of the multiple fireflies are calculated respectively, and according to the fitness values ​​from high to low, the same number of fireflies as the number of client nodes are selected as the initial firefly population.

5. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the program, it implements the intelligent transport vehicle path planning method as described in any one of claims 1 to 3.

6. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the intelligent transport vehicle path planning method as described in any one of claims 1 to 3.

7. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the intelligent transport vehicle path planning method as described in any one of claims 1 to 3.

Images