Highway efficient passing system and method based on ant colony algorithm

Abstract

The invention discloses a highway efficient passing system and method based on an ant colony algorithm. The system comprises a vehicle user terminal and a highway cloud platform management terminal. The vehicle user side provides vehicle information and driving information for the cloud platform management side through the intelligent terminal and obtains a next road section recommended by the cloud platform management side; the cloud platform management terminal sets an observation window to analyze and process the driving information of the vehicle user within a set number of days before the current day, and analyzes and processes the driving information of the vehicle user on the current day at a set time interval at the same time; whether a certain road section is an important road section or not is judged by setting a threshold value of vehicle passing times, a pheromone volatilization rate in an ant colony algorithm, a control pheromone concentration and a weight for controlling a heuristic information effect, and selection recommendation of a next road section is completed through the ant colony algorithm according to driving information provided by a vehicle user. And the result is transmitted to the vehicle user side. According to the invention, highway congestion is avoided, and the management level and the operation efficiency of the highway are improved.

Description

Technical Field

[0001] The present invention relates to an efficient highway traffic system and method based on an ant colony algorithm, and belongs to the technical field of traffic management and control. Background Art

[0002] With the rapid development of my country's economy and society, the number of motor vehicles has continued to increase, leading to a surge in highway traffic and exacerbating congestion. Traffic management departments typically implement measures to alleviate congestion after it occurs, but this approach cannot fundamentally resolve the problem. Furthermore, while existing navigation software on smart terminals can provide users with a variety of route options, the routes are relatively fixed and users cannot communicate with each other, hindering the provision of effective and reasonable driving information.

[0003] The rapid development of internet and artificial intelligence technologies in recent years has provided new ideas for efficient highway travel. Leveraging the internet and cloud computing to provide route guidance to on-the-go users is crucial for improving highway efficiency and service levels. Summary of the Invention

[0004] In response to the problems existing in the above-mentioned prior art, the present invention provides an efficient highway traffic system and method based on ant colony algorithm, which can improve road utilization and driving efficiency, avoid highway congestion, and enhance the management level and operation efficiency of highways.

[0005] In order to achieve the above-mentioned object, the present invention provides a system for efficient highway traffic based on ant colony algorithm, comprising a vehicle user terminal and a highway cloud platform management terminal;

[0006] The vehicle user terminal provides vehicle information and driving information to the highway cloud platform management terminal through the smart terminal and obtains the next road section recommended by the highway cloud platform management terminal;

[0007] The highway cloud platform management terminal sets a historical day observation window to facilitate analysis and processing of vehicle users' driving information within a certain number of days K before the current day, and sets a time interval M for the current day to facilitate analysis and processing of vehicle users' driving information every M hours. The highway cloud platform management terminal sets a vehicle passing number threshold H as a criterion for determining whether a certain road section is an important road section, and also sets the pheromone volatility rate ρ, the weight α for controlling the effect of pheromone concentration, and the weight β for controlling the effect of heuristic information in the ant colony algorithm. The highway cloud platform management terminal also completes the selection recommendation of the next road section based on the driving information provided by the vehicle user through the ant colony algorithm, and transmits the recommendation result to the intelligent terminal of the vehicle user.

[0008] A method for efficient highway traffic based on an ant colony algorithm comprises the following steps:

[0009] Step 1: Before a vehicle user starts a certain starting and ending road section, the vehicle user submits the starting and ending road section information and vehicle user information to the highway cloud platform management terminal through the vehicle user terminal. The highway cloud platform management terminal uses the information to generate the header node of the vehicle user's driving information linear table. The highway cloud platform management terminal groups the vehicle user's driving information linear table according to the different starting and ending road sections, that is, the driving information linear tables of vehicle users with the same starting and ending sections are divided into one group.

[0010] Step 2: When a vehicle user travels between the starting and ending sections, the vehicle user terminal continuously submits the pheromone vector, distance, start time, and end time of the section passed through to the highway cloud platform management terminal. The highway cloud platform management terminal inserts these nodes as nodes at the end of the corresponding vehicle user's driving information linear table in order. This is equivalent to the ants in the ant colony algorithm releasing pheromones and recording driving information along their respective routes. The pheromone vector is composed of two components: traffic congestion and traffic comfort. Traffic congestion is represented by the average speed of vehicle users traveling on a certain section of road, and traffic comfort is represented by vehicle users' evaluation of the road surface quality, service facilities, etc.

[0011] Step 3: For each pair of start and end road segments, the highway cloud platform management terminal includes the linear table of driving information of users who have newly entered the starting and ending road segments on the same day, and removes the linear table of driving information of users who have been in the past for more than K days but do not include the current day;

[0012] Step 4: For each pair of start and end road segments, the highway cloud platform management terminal uses the linear table of vehicle user driving information on the current day and within the previous K days to generate the pheromone concentration for each road segment between the start and end road segments. This stage is equivalent to pheromone update, that is, completing the volatilization and enhancement of pheromones. The generation process includes the following steps:

[0013] 4-1. For a certain section R on the route between the starting and ending sections, the highway cloud platform management terminal finds the corresponding pheromone vector from the linear table of driving information of all vehicle users passing through section R, and normalizes the two components of these pheromone vectors to obtain the modulus. The normalization process uses the z-score normalization method, which performs data normalization based on the mean and standard deviation of the original data. In the z-score normalization, the new data is calculated using the following formula: new data = (original data - mean) / standard deviation. The modulus of a vector is the length of the vector, calculated as the square root of the sum of the squares of its components.

[0014] 4-2. The highway cloud platform management terminal uses the pheromone vector modulus released by all vehicle users within the previous K days to generate the historical pheromone concentration of the road section R. The calculation formula is: Among them, ρ k represents the k-th power of the pheromone volatility rate ρ, represents the set of pheromone vector moduli released by all vehicle users passing through road section R in the first k days, τ k Representing a collection The modulus of the pheromone vector released by a certain vehicle user;

[0015] 4-3. The highway cloud platform management terminal uses the pheromone vector modulus released by all vehicle users on the day to generate the pheromone concentration of the road section R on the day at each interval M. The calculation formula is: in, represents the set of pheromone vector moduli released by all vehicle users up to the current time on that day, τ0 represents the set The modulus of the pheromone vector released by a certain vehicle user;

[0016] 4-4. Calculate the pheromone concentration of road section R according to the following formula: Where N represents the number of all vehicle users passing through road section R;

[0017] Step 5: When a vehicle user is traveling on a certain section D of the highway and is about to enter the next section, the highway cloud platform management terminal finds the corresponding group according to the starting and ending sections of the vehicle user, finds the linear table of driving information of all vehicle users that have traveled on section D in the group, and uses the next section of section D in the linear table of each vehicle user's driving information to construct a set S = {W1, W2, ..., W n};

[0018] Step 6: Count the number of times vehicles have passed through the road sections in the linear table of vehicle user travel information between road section D and the end road section, excluding road section D and the road sections in set S. If the number of times vehicles have passed through a road section exceeds a threshold value H, the road section is determined to be an important road section.

[0019] Step 7: For each road segment W in the set S i , construct the road segment D, road segment W i , terminal road sections and important road sections as nodes of the connected graph, the connection weight of the edge between the nodes is expressed by the sum of the pheromone concentrations on multiple road sections connecting the two nodes; if there are two disconnected nodes, the weight of the connection between them is set to 0;

[0020] Step 8: Based on the connected graph constructed in step 7, use Kruskal algorithm to construct a spanning tree with the largest sum of the weights of the edges in the connected graph. The sum of the weights is recorded as F. i ;

[0021] Step 9: For each road segment W in the set S i , construct the road segment D, road segment W i , the terminal road section and the important road section are used as the connected graph of the nodes. The connection weight of the edge between the nodes is expressed by the inverse of the sum of the distances of multiple road sections connecting the two nodes. If there are two disconnected nodes, the weight of the connection between them is set to 0;

[0022] Step 10: Based on the connected graph constructed in step 9, use Dijkstra's algorithm to find a shortest path from segment D to the end segment such that the sum of the weights of the edges along this path reaches the maximum. The sum of the weights is denoted as V. i ;

[0023] Step 11: For the vehicle user traveling on the current road segment D, for the road segment W in the set S i , the probability of selecting it as the next segment is: Among them, α and β are the weights controlling the effect of pheromone concentration and the weight controlling the effect of heuristic information, respectively;

[0024] Step 12: Select W i =max{P i}As the next road section to be selected by the vehicle user traveling on the current road section D, the highway cloud platform management end sends the final analysis results to the vehicle user end.

[0025] Furthermore, the construction process of using the Kruskal algorithm in step 8 to construct a spanning tree with the maximum sum of the weights of the edges in the connected graph is:

[0026] 8-1. Initialization: Sort all edges of the connected graph by weight from large to small, and construct an empty maximum spanning tree;

[0027] 8-2. Check each edge in order. If adding the edge will not form a cycle in the maximum spanning tree, then add the edge; otherwise, ignore the edge.

[0028] 8-3. Repeat step 8-2 until there are G-1 edges in the maximum spanning tree, and obtain the spanning tree with the largest sum of the weights of the edges in the connected graph; where G is the number of vertices in the connected graph.

[0029] Furthermore, the specific process of step 10 is as follows:

[0030] 10-1. Initialization: Set the distance from the node D of the road segment to the end node of the road segment to infinity, the distance from the node D of the road segment to other road segment nodes to 0, and mark all road segment nodes as unvisited;

[0031] 10-2. Select the node U on the road segment that has the largest path weight from the current node D and has not been visited, and mark the node U on the road segment as visited.

[0032] 10-3. For all adjacent nodes E of node U, if the weight of the path from node D through node U to node E is greater than the weight of the currently known path from node D to node E, then update the distance from node D to node E.

[0033] 10-4. Repeat steps 10-2 to 10-3 until all the road segment nodes are visited;

[0034] 10-5. Based on the shortest path from the node of segment D to other segment nodes obtained in step 10-4, access the node of the end segment in reverse order until the node of segment D, and obtain a path from segment D to the end segment so that the sum of the weights of the edges along this path reaches the maximum shortest path.

[0035] The present invention provides a vehicle user terminal with vehicle information and driving information to a highway cloud platform management terminal via an intelligent terminal, and obtains the next road segment recommended by the highway cloud platform management terminal. The highway cloud platform management terminal sets a historical day observation window to facilitate analysis and processing of vehicle user driving information within a certain number of days K before the current day, and sets a daily time interval M to facilitate analysis and processing of vehicle user driving information every M hours. The highway cloud platform management terminal sets a vehicle passing number threshold H as a criterion for determining whether a road segment is an important road segment, and also sets a pheromone volatility rate ρ, a weight α for controlling the effect of pheromone concentration, and a weight β for controlling the effect of heuristic information in an ant colony algorithm. The highway cloud platform management terminal also uses an ant colony algorithm to complete the selection and recommendation of the next road segment based on the driving information provided by the vehicle user, and transmits the recommendation result to the intelligent terminal of the vehicle user terminal. The present invention improves road utilization and driving efficiency, avoids highway congestion, and enhances the management level and operation efficiency of highways. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 It is a workflow diagram of the present invention. DETAILED DESCRIPTION

[0037] The present invention will be further described below with reference to the accompanying drawings.

[0038] like Figure 1 As shown, a system for efficient highway traffic based on ant colony algorithm includes a vehicle user end and a highway cloud platform management end;

[0039] The vehicle user terminal provides vehicle information and driving information to the highway cloud platform management terminal through the smart terminal and obtains the next road section recommended by the highway cloud platform management terminal;

[0040] The highway cloud platform management terminal sets a historical day observation window to facilitate analysis and processing of vehicle users' driving information within a certain number of days K before the current day, and sets a time interval M for the current day to facilitate analysis and processing of vehicle users' driving information every M hours. The highway cloud platform management terminal sets a vehicle passing number threshold H as a criterion for determining whether a certain road section is an important road section, and also sets the pheromone volatility rate ρ, the weight α for controlling the effect of pheromone concentration, and the weight β for controlling the effect of heuristic information in the ant colony algorithm. The highway cloud platform management terminal also completes the selection recommendation of the next road section based on the driving information provided by the vehicle user through the ant colony algorithm, and transmits the recommendation result to the intelligent terminal of the vehicle user.

[0041] Example: A method for efficiently passing through a highway based on an ant colony algorithm, specifically comprising the following steps:

[0042] (1) Before starting a route, a vehicle user submits the starting and ending route information and vehicle user information to the cloud platform management terminal, which then uses it to generate the header node of the vehicle user's driving information linear table. Based on the different starting and ending routes, the highway cloud platform management terminal groups the vehicle user's driving information linear table, that is, the driving information linear tables of vehicle users with the same starting and ending routes are grouped together;

[0043] (2) When a vehicle user is traveling between the starting and ending sections, the vehicle user will continuously submit the pheromone vectors of each section passed by, which are the two components consisting of traffic congestion and traffic comfort, distance, start time and end time, to the cloud platform management end. The highway cloud platform management end will insert them as nodes at the end of the linear table of the corresponding vehicle user's driving information in order. This is equivalent to the ants in the ant colony algorithm releasing pheromones and recording driving information on their respective routes. Among them, traffic congestion is described by the average speed of vehicle users traveling on a certain section of road, and traffic comfort is described by the vehicle users' evaluation of the road surface quality, service facilities and other aspects of a certain section of road;

[0044] (3) For each pair of start and end road segments, the highway cloud platform management terminal will include the linear table of vehicle users who have newly entered the start and end road segments on the day, and at the same time remove the linear table of vehicle users whose history of days exceeds K (excluding the day) at the end of the day. In this implementation case, K is set to 7;

[0045] (4) For each pair of start and end sections, the highway cloud platform management terminal uses the linear table of vehicle user driving information on the current day and the previous K days (from the 1st day to the Kth day) to generate the pheromone concentration of each section on the route between the start and end sections. This stage is equivalent to updating the pheromone, that is, completing the volatilization and enhancement of the pheromone. The generation process includes the following steps:

[0046] 1) For a certain section R on the route between the starting and ending sections, the highway cloud platform management end finds the corresponding pheromone vector from the linear table of driving information of all vehicle users passing through the section R, and normalizes the two components of these pheromone vectors to obtain the modulus.

[0047] 2) The highway cloud platform management terminal uses the pheromone vector modulus released by all vehicle users within the previous K days to generate the historical pheromone concentration of the road section R. The calculation formula is as follows: Among them, ρ k represents the k-th power of the pheromone volatility rate ρ, represents the set of pheromone vector moduli released by all vehicle users passing through road section R in the first k days, τ k Representing a collection The modulus of the pheromone vector released by a certain vehicle user; in this embodiment, ρ is set to 0.5;

[0048] 3) The highway cloud platform management terminal continuously uses the pheromone vector modulus released by all vehicle users on the day at a certain time interval M to generate the pheromone concentration of the road section R on the day. The calculation formula is as follows: in, represents the set of pheromone vector moduli released by all vehicle users up to the current time on that day, τ0 represents the set The modulus of the pheromone vector released by a certain vehicle user; in this embodiment, M is set to 2 hours;

[0049] 4) Calculate the pheromone concentration of the road section R according to the following formula: Where N represents the number of all vehicle users passing through road section R;

[0050] (5) When a vehicle user is driving on a certain section D of the highway and is about to drive to the next section, the highway cloud platform management terminal will find the corresponding group according to the starting and ending sections of the vehicle user, find the linear table of driving information of all vehicle users who have driven on section D in the group, and use the next section of section D in the linear table of driving information of each vehicle user to construct the set S = {W1, W2, ..., W n};

[0051] (6) The number of vehicle passes in the linear table of user travel information of vehicles that have traveled from segment D to the end segment (excluding segment D and segments in set S) is counted. If the number of vehicle passes of a segment exceeds a threshold H, the segment is determined to be an important segment. In this embodiment, H is set to 10,000 vehicles.

[0052] (7) For each road segment W in the set S i , construct the road segment D, road segment W i , terminal road segments and important road segments as nodes. The connection weight of the edge between nodes is expressed as the sum of the pheromone concentrations on multiple road segments connecting the two nodes. If there are two disconnected nodes, the weight of the connection between them is set to 0;

[0053] (8) Based on the connected graph constructed above, the Kruskal algorithm is used to construct a spanning tree with the largest sum of the weights of the edges in the graph. The sum of the weights is denoted as F. i The construction process includes the following steps:

[0054] 1) Initialization: Sort all edges of the graph by weight from large to small and construct an empty maximum spanning tree;

[0055] 2) Check each edge in order. If adding the edge will not form a cycle in the maximum spanning tree, then add the edge; otherwise, ignore the edge.

[0056] 3) Repeat step 2) until there are G-1 edges in the maximum spanning tree, where G is the number of vertices in the graph;

[0057] 4) Based on the results obtained above, we get the spanning tree with the largest sum of the weights of the edges in the graph;

[0058] (9) For each road segment W in the set S i , construct the road segment D, road segment W i , terminal road segments and important road segments as nodes. The connection weight of the edge between nodes is expressed as the inverse of the sum of the distances of multiple road segments connecting the two nodes. If there are two disconnected nodes, the weight of the connection between them is set to 0;

[0059] (10) Based on the connected graph constructed above, Dijkstra's algorithm is used to find a path from segment D to the end segment so that the sum of the weights of the edges along this path reaches the maximum shortest path. The sum of the weights is recorded as V. i The construction process includes the following steps:

[0060] 1) Initialization: Set the distance from the node D of the segment to the node of the end segment in the graph to infinity, the distance from the node D of the segment to other segment nodes to 0, and all segment nodes are marked as unvisited;

[0061] 2) Select the node U of the current segment that has the largest path weight from the node D and has not been visited, and mark the node U of the segment as visited;

[0062] 3) For all neighboring nodes E of node U, if the path weight from node D through node U to node E is greater than the currently known path weight from node D to node E, then update the distance from node D to node E.

[0063] 4) Repeat steps 2) to 3) until all the road nodes are visited;

[0064] 5) Based on the shortest path from the node of segment D to other segment nodes obtained above, access the node of the terminal segment in reverse order until the node of segment D, so that a path from segment D to the terminal segment is obtained so that the sum of the weights of the edges along this path reaches the maximum shortest path.

[0065] (11) For a vehicle user traveling on the current road segment D, for the road segment W in the set S, i , the probability of selecting it as the next segment is: Among them, α and β are used to control the weight relationship between the heuristic information and the pheromone concentration; in this embodiment, α=1, β=3;

[0066] (12) Select W i =max{P i}As the next road section to be selected by the vehicle user traveling on the current road section D, the cloud platform management end will send the final analysis results to the vehicle user end.

Claims

1. An efficient highway traffic system based on ant colony algorithm, characterized by: Including vehicle user end and highway cloud platform management end; The vehicle user terminal provides vehicle information and driving information to the highway cloud platform management terminal through the smart terminal and obtains the next road section recommended by the highway cloud platform management terminal; The highway cloud platform management terminal sets a historical day observation window to facilitate analysis and processing of vehicle users' driving information within K days before the current day, and sets a time interval M for the current day to facilitate analysis and processing of vehicle users' driving information every M hours. The highway cloud platform management terminal sets a vehicle passing number threshold H as a criterion for determining whether a road section is an important road section, and also sets the pheromone volatility rate ρ, the weight α for controlling the effect of pheromone concentration, and the weight β for controlling the effect of heuristic information in the ant colony algorithm. The highway cloud platform management terminal also completes the selection recommendation of the next road section based on the driving information provided by the vehicle user through the ant colony algorithm, and transmits the recommendation result to the intelligent terminal of the vehicle user.

2. An efficient highway traffic method based on ant colony algorithm, characterized in that: The steps include: Step 1: Before a vehicle user starts a certain starting and ending road section, the vehicle user submits the starting and ending road section information and vehicle user information to the highway cloud platform management terminal through the vehicle user terminal. The highway cloud platform management terminal uses the information to generate the header node of the vehicle user's driving information linear table. The highway cloud platform management terminal groups the vehicle user's driving information linear table according to the different starting and ending road sections, that is, the driving information linear tables of vehicle users with the same starting and ending sections are divided into one group. Step 2: When a vehicle user travels between a starting and ending road section, the vehicle user terminal continuously submits the pheromone vector, distance, start time, and end time of the road section passed through to the highway cloud platform management terminal through the vehicle user terminal. The highway cloud platform management terminal inserts the pheromone vectors as nodes at the end of the corresponding vehicle user's driving information linear table in order. The pheromone vector is composed of two components: traffic congestion and traffic comfort. Traffic congestion is represented by the average speed of the vehicle user traveling on a certain road section, and traffic comfort is represented by the vehicle user's evaluation of the road surface quality, service facilities, etc. of the certain road section. Step 3: For each pair of start and end road segments, the highway cloud platform management terminal includes the linear table of driving information of users who have newly entered the starting and ending road segments on the same day, and removes the linear table of driving information of users who have been in the past for more than K days but do not include the current day; Step 4: For each pair of start and end road segments, the highway cloud platform management terminal uses the linear table of vehicle user driving information on the current day and within the previous K days to generate the pheromone concentration of each road segment on the route between the start and end road segments. The generation process includes the following steps: 4-1. For a certain road section R between the starting and ending sections, the highway cloud platform management terminal finds the corresponding pheromone vector from the linear table of driving information of all vehicles and users passing through this road section R, and normalizes the two components of these pheromone vectors to obtain the modulus. 4-2. The highway cloud platform management terminal uses the pheromone vector modulus released by all vehicle users within the previous K days to generate the historical pheromone concentration of the road section R. The calculation formula is: Among them, ρ k represents the k-th power of the pheromone volatility rate ρ, represents the set of pheromone vector moduli released by all vehicle users passing through road section R in the first k days, τ k Representing a collection The modulus of the pheromone vector released by a certain vehicle user; 4-3. The highway cloud platform management terminal uses the pheromone vector modulus released by all vehicle users on the day to generate the pheromone concentration of the road section R on the day at each interval M. The calculation formula is: in, represents the set of pheromone vector moduli released by all vehicle users up to the current time on that day, τ0 represents the set The modulus of the pheromone vector released by a certain vehicle user; 4-4. Calculate the pheromone concentration of road section R according to the following formula: Where N represents the number of all vehicle users passing through road section R; Step 5: When a vehicle user is traveling on a certain section D of the highway and is about to enter the next section, the highway cloud platform management terminal finds the corresponding group according to the starting and ending sections of the vehicle user, finds the linear table of driving information of all vehicle users that have traveled on section D in the group, and uses the next section of section D in the linear table of each vehicle user's driving information to construct a set S = {W1, W2, ..., W n }; Step 6: Count the number of times vehicles have passed through the road sections in the linear table of vehicle user travel information between road section D and the end road section, excluding road section D and the road sections in set S. If the number of times vehicles have passed through a road section exceeds a threshold value H, the road section is determined to be an important road section. Step 7: For each road segment W in the set S i , construct the road segment D, road segment W i , terminal road sections and important road sections as nodes of the connected graph, the connection weight of the edge between the nodes is expressed by the sum of the pheromone concentrations on multiple road sections connecting the two nodes; if there are two disconnected nodes, the weight of the connection between them is set to 0; Step 8: Based on the connected graph constructed in step 7, use Kruskal algorithm to construct a spanning tree with the largest sum of the weights of the edges in the connected graph. The sum of the weights is recorded as F. i ; Step 9: For each road segment W in the set S i , construct the road segment D, road segment W i , the terminal road section and the important road section are used as the connected graph of the nodes. The connection weight of the edge between the nodes is expressed by the inverse of the sum of the distances of multiple road sections connecting the two nodes. If there are two disconnected nodes, the weight of the connection between them is set to 0; Step 10: Based on the connected graph constructed in step 9, use Dijkstra's algorithm to find a shortest path from segment D to the end segment such that the sum of the weights of the edges along this path reaches the maximum. The sum of the weights is denoted as V. i ; Step 11: For the vehicle user traveling on the current road segment D, for the road segment W in the set S i , the probability of selecting it as the next segment is: Among them, α and β are the weights controlling the effect of pheromone concentration and the weight controlling the effect of heuristic information, respectively; Step 12: Select W i =max{P i }As the next road section to be selected by the vehicle user traveling on the current road section D, the highway cloud platform management end sends the final analysis results to the vehicle user end.

3. The efficient highway traffic method based on ant colony algorithm according to claim 2 is characterized in that: The construction process of using Kruskal algorithm in step 8 to construct a spanning tree with the largest sum of edge weights in a connected graph is as follows: 8-1. Initialization: Sort all edges of the connected graph by weight from large to small, and construct an empty maximum spanning tree; 8-2. Check each edge in order. If adding the edge will not form a cycle in the maximum spanning tree, then add the edge; otherwise, ignore the edge. 8-3. Repeat step 8-2 until there are G-1 edges in the maximum spanning tree, and obtain the spanning tree with the largest sum of the weights of the edges in the connected graph; where G is the number of vertices in the connected graph.

4. The efficient highway traffic method based on ant colony algorithm according to claim 2 is characterized in that: The specific process of step 10 is as follows: 10-1. Initialization: Set the distance from the node D of the road segment to the end node of the road segment to infinity, the distance from the node D of the road segment to other road segment nodes to 0, and mark all road segment nodes as unvisited; 10-2. Select the node U on the road segment that has the largest path weight from the current node D and has not been visited, and mark the node U on the road segment as visited. 10-3. For all adjacent nodes E of node U, if the weight of the path from node D through node U to node E is greater than the weight of the currently known path from node D to node E, then update the distance from node D to node E. 10-4. Repeat steps 10-2 to 10-3 until all the road segment nodes are visited; 10-5. Based on the shortest path from the node of segment D to other segment nodes obtained in step 10-4, access the node of the end segment in reverse order until the node of segment D, and obtain a path from segment D to the end segment so that the sum of the weights of the edges along this path reaches the maximum shortest path.

Images