A path real-time optimization method based on GPS vehicle positioning

Abstract

This invention provides a real-time route optimization method based on GPS vehicle positioning, belonging to the field of route planning and optimization technology. It acquires GPS positioning data, vehicle load data, hazardous material type data, and road restriction data of hazardous materials transport vehicles, and combines this with feasibility assessment and real-time risk assessment of candidate routes to achieve dynamic route optimization and priority ranking. When temporary traffic control is detected in sensitive areas, the route can be quickly replanned. This invention can significantly improve the safety and flexibility of hazardous materials transportation route planning, effectively reduce potential risks during transportation, and further enhance transportation safety while ensuring transportation efficiency.

Description

Technical Field

[0001] This invention relates to the field of route planning and optimization technology, and more specifically, to a real-time route optimization method based on GPS vehicle positioning. Background Technology

[0002] With the rapid development of the global logistics and transportation industry, the transportation of hazardous materials, as a crucial component, has received widespread attention. Hazardous materials transport vehicles involve high-risk factors during transport, and their safety directly impacts public safety, property security, and environmental protection. Currently, GPS (Global Positioning System)-based vehicle positioning technology has become an indispensable part of the logistics and transportation sector, providing a foundation for monitoring and optimizing the transportation process by acquiring real-time vehicle location data. However, in the field of hazardous materials transportation, GPS positioning data alone is insufficient to meet the safety requirements of complex transportation environments. Traditional route planning methods typically rely on fixed road network models for calculations, failing to fully consider the specific characteristics of hazardous materials transportation, such as load capacity, hazardous material type, and road restrictions. Furthermore, existing route planning technologies have weak responsiveness to real-time road conditions, lack support for dynamic route adjustments, and cannot flexibly respond to emergencies or temporary traffic control measures.

[0003] In existing technologies, some studies have attempted to incorporate factors such as vehicle load and type of hazardous materials into the route planning process. For example, by analyzing the characteristics of different types of hazardous materials, transport vehicles can avoid passing through densely populated or ecologically sensitive areas to reduce potential risks. However, these methods are often based on static route planning and lack comprehensive consideration of real-time traffic conditions and dynamic regulations. Furthermore, the evaluation indicators for route selection and optimization are relatively singular, typically considering only single factors such as distance or time, failing to prioritize routes from a comprehensive risk perspective. These shortcomings limit the effectiveness of existing technologies in the transportation of hazardous materials, especially when transport vehicles need to respond quickly to dynamic changes in the external environment; existing route planning methods cannot provide efficient and safe solutions in such cases. Summary of the Invention

[0004] To address the aforementioned technical problems, this invention is proposed. This invention provides a real-time route optimization method based on GPS vehicle positioning.

[0005] According to one aspect of the present invention, a real-time route optimization method based on GPS vehicle positioning is provided, comprising:

[0006] Obtain GPS location data, vehicle load data, hazardous materials type data, and road restriction data for vehicles transporting hazardous materials;

[0007] Based on the GPS positioning data, multiple candidate routes between the current location of the hazardous materials transport vehicle and its destination are determined;

[0008] Using the vehicle load data and the hazardous materials type data as constraints, the feasibility of the candidate routes is evaluated, and a set of compliant routes that meet the constraints is selected.

[0009] Real-time risk assessment is performed on the paths in the set of compliant paths, a comprehensive risk value is calculated, and the paths are prioritized based on real-time traffic data. When temporary traffic control is detected in the sensitive area, candidate paths are re-determined.

[0010] Furthermore, the multiple candidate paths are determined using a hierarchical road network search algorithm, which divides the electronic map road network data into a highway layer, a national and provincial highway layer, and an urban road layer, and establishes the connection relationship between road segment nodes in each layer;

[0011] The main road segments are divided based on the distance between the current geographic coordinates and the target geographic coordinates, and the repetition rate is determined to obtain multiple candidate paths that meet the requirements.

[0012] Furthermore, the hierarchical road network search algorithm can be expressed as follows:

[0013]

[0014] in, Let p be the search function for path p. Let x_s be the straight-line distance from the starting point x_s to the ending point x_e. Let k be the weight of the k-th segment of the highway layer. This represents the actual length of the k-th segment of the highway layer. The speed limit for the kth segment of national, provincial, and urban roads. This represents the grade coefficient of the k-th segment of national, provincial, and urban roads. The congestion index for the k-th segment of national and provincial highways and urban roads. This represents the total number of road segments on the highway level. This refers to the total number of road sections in the national, provincial, and urban road layers.

[0015] Furthermore, when the main road segment is a highway, multiple different highway combination paths are calculated using the Dijkstra algorithm, and the path is divided into sub-intervals for separate search based on the distance. Then, these sub-interval paths are connected to form a complete path.

[0016] When the number of search nodes exceeds the preset number, a bidirectional Dijkstra algorithm is used, starting the search from both the starting point and the ending point simultaneously. The search is completed when the vertex sets of the two search directions intersect.

[0017] Furthermore, the Dijkstra algorithm includes the following steps:

[0018] Initialize the distance from the starting point to all vertices except itself to infinity, and the distance from the starting point to itself to zero;

[0019] Establish a priority queue. When the priority queue is not empty, perform the following operations:

[0020] Take the first element of the priority queue as the current processing vertex;

[0021] Obtain all neighboring vertices of the currently processed vertex;

[0022] Calculate the new distance from the starting point through the currently processed vertex to each of the adjacent vertices;

[0023] When the new distance is less than the currently recorded distance of the corresponding adjacent vertex, the distance value of the adjacent vertex is updated to the new distance, and the adjacent vertex is added to the priority queue;

[0024] Record the predecessor vertex corresponding to the updated distance;

[0025] Once the target vertex has been processed, the shortest path is obtained by reversing the process from the target vertex through the predecessor vertex.

[0026] Furthermore, the determination of the intersection of the vertex sets of the two search directions is achieved by maintaining the start and end point access sets. When the same vertex appears in both sets, the shortest total distance through that vertex is calculated and updated, and the search is terminated early when the threshold condition is met.

[0027] Furthermore, the constraints include hard constraints and soft constraints;

[0028] The hard constraints include vehicle load constraints, hazardous materials type constraints, and time constraints.

[0029] The soft constraints are evaluated using a fuzzy evaluation method, which calculates the overall score of the path by setting weight coefficients.

[0030] Furthermore, the soft constraint evaluation score is calculated as follows:

[0031]

[0032] in, The path is evaluated and scored. Let i be the road grade coefficient of the i-th road segment. Let be the length of the i-th road segment. This is the total path length. For ideal travel time, This is the actual estimated travel time. The straight-line distance between the start and end points. The congestion impact coefficient, The average congestion index for the route. As the benchmark congestion index, This represents the number of road segments contained in the path.

[0033] Furthermore, all candidate paths that pass the hard constraint assessment are sorted from high to low according to their soft constraint assessment scores, and the three paths with the highest scores are selected as the set of compliant paths.

[0034] Furthermore, the comprehensive risk value is calculated by dividing the path into multiple assessment units, calculating the basic risk value and dynamic risk value of each unit, weighting the two to obtain the unit risk value, summing the risk values ​​of all units, calculating the transition risk value based on the difference in risk values ​​between adjacent assessment units, and adding the sum of the unit risk values ​​to the transition risk value to obtain the comprehensive risk value of the path.

[0035] Compared with existing technologies, the GPS-based vehicle positioning-based real-time route optimization method provided by this invention acquires GPS positioning data, vehicle load data, hazardous material type data, and road restriction data of hazardous materials transport vehicles. Combined with feasibility assessments and real-time risk assessments of candidate routes, it achieves dynamic route optimization and priority ranking. When temporary traffic control is detected in sensitive areas, the route can be quickly replanned. This significantly improves the safety and flexibility of hazardous materials transport route planning, effectively reduces potential risks during transportation, and ensures transportation efficiency while further enhancing transportation safety. Attached Figure Description

[0036] To more clearly illustrate the technical solutions in the embodiments of the present 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 the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. In the drawings:

[0037] Figure 1 This is a flowchart of a real-time route optimization method based on GPS vehicle positioning according to an embodiment of the present invention.

[0038] Figure 2 This is a flowchart of the bidirectional Dijkstra algorithm in the GPS vehicle positioning-based real-time path optimization method according to an embodiment of the present invention. Detailed Implementation

[0039] Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings. It is obvious that the described embodiments are merely some embodiments of the present invention, and not all embodiments of the present invention, and it should be understood that the present invention is not limited to the exemplary embodiments described herein.

[0040] Figure 1 This is a flowchart of a real-time route optimization method based on GPS vehicle positioning according to an embodiment of the present invention. Figure 1 As shown, the real-time path optimization method based on GPS vehicle positioning includes:

[0041] S1: Obtain GPS location data, vehicle load data, hazardous materials type data, and road restriction data for hazardous materials transport vehicles, wherein the road restriction data includes information on dedicated hazardous materials transport lanes, weight-restricted roads, and time-based traffic restrictions.

[0042] Obtaining GPS positioning data, vehicle load data, hazardous materials type data, and road restriction data for hazardous materials transport vehicles specifically includes: collecting GPS positioning data through a GPS positioning terminal installed on the hazardous materials transport vehicle, wherein the GPS positioning data continuously records the vehicle's longitude, latitude, and direction of travel at a sampling period of 10 seconds; retrieving the vehicle load data from vehicle management, wherein the vehicle load data includes the vehicle's empty weight, current loaded weight, and approved maximum load capacity; and extracting the hazardous materials type data from hazardous materials transport management, wherein the hazardous materials type data includes the UN number, hazard class, packaging specifications, and transport requirements of the hazardous materials; from... The traffic management department obtains the road restriction data from a real-time updated database. This road restriction data includes information on dedicated lanes for hazardous materials transportation, weight-restricted roads, and time-based restrictions. The dedicated lane information identifies the designated lanes and times during which hazardous materials vehicles are permitted to pass. The weight-restricted road information records bridges, tunnels, and road sections with weight limits on vehicles. The time-based restrictions include traffic control regulations for hazardous materials transport vehicles at different times. The GPS positioning data, vehicle load data, hazardous materials type data, and road restriction data are all stored in a unified data format and updated in real time.

[0043] S2: Based on the GPS positioning data, determine multiple candidate routes between the current location of the hazardous materials transport vehicle and its destination, and mark the restricted areas on each candidate route based on the road restriction data to generate a restricted area distribution map;

[0044] The GPS positioning data is parsed into the current geographic coordinates of the hazardous materials transport vehicle, and the preset destination information is converted into the target geographic coordinates. A hierarchical road network search algorithm is used to search the electronic map road network data, starting from the current geographic coordinates and ending at the target geographic coordinates, for the top five candidate paths with the shortest distance. Each candidate path consists of several road segment nodes. For each candidate path, road segment analysis is performed, and the information on hazardous materials transport lanes, weight-restricted roads, and time-based traffic restrictions in the road restriction data is compared with the information on each road segment node. The process involves matching traffic restrictions on road segments. When such restrictions apply, the restricted area, time period, and type of restriction are marked on the electronic map. All marked restricted areas are then integrated to generate a restricted area distribution map. In this map, different graphic symbols are used to distinguish between hazardous materials restricted areas, weight-restricted areas, and time-based restricted areas. Connecting lines are used to indicate the spatial relationships between the restricted areas, forming a complete restricted area network covering the candidate routes. The restricted area distribution map is overlaid on the electronic map base in vector layer form, supporting dynamic scaling and real-time updates.

[0045] The hierarchical road network search algorithm specifically includes: firstly, dividing the electronic map road network data into three levels: expressway layer, national and provincial highway layer, and urban road layer, and establishing road segment node connection relationships in each level; when the straight-line distance between the current geographic coordinate point and the target geographic coordinate point is greater than 100 kilometers, priority is given to searching for main road segments in the expressway layer, and multiple different expressway combination paths are calculated using the Dijkstra algorithm; after determining the expressway combination path, supplementary road segments connecting the current geographic coordinate point to the expressway entrance and the expressway exit to the target geographic coordinate point are searched in the national and provincial highway layer; when the straight-line distance between the current geographic coordinate point and the target geographic coordinate point is less than 100 kilometers, path searches are performed simultaneously in the national and provincial highway layer and the urban road layer, and road segment weights are assigned to each selectable path according to the road segment's traffic speed limit, road grade, and congestion index; if the overlap rate between a searched path and a selected path exceeds 60%, the path is eliminated, and the search continues for the next path until five candidate paths with an overlap rate of less than 60% are obtained. The hierarchical road network search algorithm is shown in the following formula:

[0046]

[0047] in, Let p be the search function for path p. x_s represents the straight-line distance (in kilometers) from the starting point x_s to the ending point x_e. Let k be the weight of the k-th segment of the highway layer. This represents the actual length of the k-th segment of the highway layer. The speed limit for the kth segment of national, provincial, and urban roads. This represents the grade coefficient of the k-th segment of national, provincial, and urban roads. The congestion index (range 0-1) is the congestion index of the kth segment of national and provincial highways and urban roads. This represents the total number of road segments on the highway level. This refers to the total number of road sections in the national, provincial, and urban road layers.

[0048] The specific steps of Dijkstra's algorithm are as follows: First, initialize the distance from the starting vertex_start to all other vertices to infinity, and the distance from the starting vertex to itself to 0; establish a priority queue_queue to store vertices to be processed, and sort the elements in the queue according to their distance to the starting vertex; when the priority queue is not empty, take the first element of the queue, vertex_current, as the current vertex to be processed; for each adjacent vertex_adjacent of the current vertex_current, calculate the distance dist_new from the starting vertex through vertex_current to vertex_adjacent; when dist_new is less than the currently recorded distance dist_old of vertex_adjacent, update the distance value of vertex_adjacent to dist_new, and add vertex_adjacent to the priority queue; at the same time, record the predecessor vertex_current when the distance of vertex_adjacent is updated, for subsequent path reconstruction; after the target vertex_end is processed, start from vertex_end and reverse the path through the predecessor vertex to obtain a shortest path path_shortest.

[0049] To obtain multiple different combinations of highway paths, the basic Dijkstra algorithm is extended as follows: after finding a shortest path path_shortest, the weight of all edges on that path is increased by 50%, and Dijkstra's algorithm is executed again to obtain the second shortest path path_second; and so on, by dynamically adjusting the weight of the used road segments, 3-5 different combinations of highway paths are finally obtained; if the total weight of a certain path exceeds 1.5 times the total weight of the shortest path, the search is stopped.

[0050] The following optimization mechanisms were implemented during algorithm execution: For long-distance transportation with a straight-line distance exceeding 500 kilometers, a segmented search strategy was adopted, dividing the path into several sub-intervals of approximately 300 kilometers each for separate searches, and then connecting the sub-interval paths to form a complete path; when the number of search nodes exceeded 1000, a bidirectional Dijkstra's algorithm was used, simultaneously searching from both the starting and ending points, and the search was completed when the vertex sets of the two search directions intersected. The specific implementation of the intersection judgment criterion for the bidirectional Dijkstra's algorithm is as follows:

[0051] During the bidirectional search process, two sets are maintained, such as... Figure 2 As shown: `forward_visited` and `backward_visited` record the vertices visited and their distance information when the search starts from the beginning and end points, respectively. The distance information for each vertex includes: the distance from the beginning point to the vertex `dist_from_start` and the distance from the end point to the vertex `dist_from_end`. The specific judgment process is as follows: When a vertex_current appears in both the forward_visited and backward_visited sets, calculate the total distance of the path through that vertex: total_distance = dist_from_start + dist_from_end; maintain a global minimum distance variable min_total_distance, initially set to infinity; when total_distance < min_total_distance, update the value of min_total_distance and record the current vertex_current as the optimal intersection point vertex_best; to avoid prematurely stopping the search, set an early termination threshold threshold = min_total_distance × 0.8; when the sum of the minimum distances in forward_visited and backward_visited is greater than threshold, terminate the search.

[0052] The intersection determination also needs to consider the following special cases: the adjacent vertex vertex vertex adjacent of the current vertex vertex_current is already in the opposite search set, but a shorter path may be obtained by reaching vertex_adjacent through vertex_current. In this case, the distance information of vertex_adjacent needs to be updated; the search speeds of the two directions may be unbalanced. A search depth balance factor balance_factor is set. When the number of search nodes in one direction exceeds the balance_factor of the other direction, the search is forced to switch to the other direction; for highway sections with one-way traffic restrictions, the feasibility of the path needs to be verified when determining the intersection to ensure that the generated path meets the traffic conditions.

[0053] It is worth noting that although the above description outlines the general process of calculating multiple different highway combination paths using Dijkstra's algorithm, the specific implementation details may vary depending on the region and route. For example, consider the planning of a hazardous materials transportation route between two cities approximately 800 kilometers apart: firstly, the route from the starting point A toll station to the ending point Z toll station is divided into three sub-intervals for searching, each sub-interval being approximately 250-300 kilometers. The first round of Dijkstra's algorithm search yielded the shortest path: Toll Station A -> Service Area B -> Interchange C -> Service Area D (first sub-section) -> Interchange E -> Service Area F -> Interchange G (second sub-section) -> Service Area H -> Interchange I -> Toll Station Z (third sub-section), with a total length of approximately 795 kilometers. This path utilizes national highway trunk lines throughout. After increasing the weight of the preferred path by 50%, the second round of search yielded the second-best path: Toll Station A -> Service Area J -> Interchange K (first sub-section) -> Service Area L -> Interchange M -> Service Area N (second sub-section) -> Interchange O -> Service Area P -> Toll Station Z (third sub-section), with a total length of approximately 830 kilometers. This path combines national and provincial highways. The third round of search yielded the following path: Toll Station A -> Service Area Q -> Interchange R (first sub-section) -> Service Area S -> Interchange T (second sub-section) -> Service Area U -> Interchange V -> Z toll station (third sub-section), with a total mileage of approximately 870 kilometers, uses some provincial expressways to replace highway sections; since the total mileage of the fourth possible path exceeds 1.5 times that of the shortest path (approximately 1192 kilometers), the algorithm terminates the search.

[0054] For example, consider the planning of hazardous materials transportation routes between two cities approximately 600 kilometers apart. The bidirectional Dijkstra algorithm simultaneously searches from both the starting point (toll station A) and the ending point (toll station Z): A priority queue, `forward_queue`, is maintained in the starting direction, sorted by distance to starting point A; another priority queue, `backward_queue`, is maintained in the ending direction, sorted by distance to ending point Z. The two directions are searched alternately. The search ends when node M (to the starting point) and node N (to the ending point) are located on the same highway. The path from A to M is then connected to the path from N to Z, resulting in the first complete path: Toll station A -> Service area B -> Interchange C -> Interchange M (forward search endpoint) -> Interchange N (reverse search endpoint) -> Interchange Y -> Toll station Z, with a total distance of approximately 585 kilometers. After increasing the weight of all road segments in this path by 50%, the second round of bidirectional search meets at nodes P and Q in both directions, resulting in the second path: Toll station A -> Service area D -> Interchange E -> Interchange P -> Interchange Q -> The total distance from Interchange X to Toll Station Z is approximately 620 kilometers. The third round of search meets at the R and S nodes in both directions, resulting in a third path: Toll Station A -> Service Area F -> Interchange G -> Interchange R -> Interchange S -> Interchange W -> Toll Station Z, with a total distance of approximately 650 kilometers. Compared to the unidirectional Dijkstra algorithm, the bidirectional search reduces the search space by approximately 40%, and the number of search nodes decreases from approximately 900 to approximately 530, significantly improving the algorithm's execution efficiency.

[0055] S3: Using a multi-constraint matching algorithm, the vehicle load data and the hazardous materials type data are used as constraints to evaluate the feasibility of the candidate paths and select a set of compliant paths that meet the constraints.

[0056] First, a constraint evaluation matrix is ​​constructed. The row vectors of the matrix represent different types of constraints, including hard constraints such as the vehicle's current total weight, hazard UN number, packaging class, and transport time requirements, as well as soft constraints such as road grade preference, travel time, and route length. When conducting route feasibility assessment, each parameter of each candidate route is matched against the conditions in the constraint evaluation matrix. For hard constraints, a Boolean-type judgment method is used, meaning the route must fully meet the constraint requirements to pass the evaluation. For soft constraints, a fuzzy evaluation method is used, calculating the route's comprehensive score by setting weight coefficients. If any segment traversed by a candidate route does not meet vehicle weight restrictions or hazardous materials transport restrictions, the route is directly marked as infeasible. The process involves several steps: When a candidate path meets all hard constraints, its evaluation score under soft constraints is calculated. The evaluation score is calculated using a weighted summation method, with road grade preference weighted at 0.4, travel time weighted at 0.35, and path length weighted at 0.25. All candidate paths that pass the hard constraint evaluation are sorted from highest to lowest soft constraint evaluation score, and the three highest-scoring paths are selected as the compliant path set. If fewer than three candidate paths pass the hard constraint evaluation, all paths that pass are included in the compliant path set. Each path in the compliant path set is marked with detailed constraint satisfaction information, including surplus load capacity in weight-restricted areas and surplus time during hazardous materials transport periods, providing a basis for subsequent real-time scheduling decisions.

[0057] The hard constraint judgment table includes three main categories: vehicle load constraints, hazardous materials type constraints, and time constraints. The vehicle load constraint checks whether the weight limits of all bridges, tunnels, and special road sections along the route are greater than the vehicle's current total weight. If the weight limit of any road section is less than the vehicle's current total weight, the judgment result is false. The hazardous materials type constraint checks whether the UN number of the hazardous material belongs to the list of hazardous materials permitted for passage on each road section. If the list of prohibited or restricted hazardous materials on any road section includes the UN number of the currently transported hazardous material, the judgment result is false. The time constraint checks whether all road sections are within the permitted passage time according to the special time control requirements for hazardous materials transportation. If the expected passage time overlaps with the prohibited passage time of any road section, the judgment result is false. If all three constraint judgment results for a candidate route are true, the route passes the hard constraint judgment and enters the soft constraint evaluation stage. If any constraint judgment result is false, the route will be directly eliminated.

[0058] The evaluation score of this path under soft constraints is shown in the following formula:

[0059]

[0060] in, The path is evaluated and scored. Let be the road grade coefficient for the i-th road segment (1.0 for expressways, 0.8 for national highways, 0.6 for provincial highways, and 0.4 for county roads). Let be the length (in kilometers) of the i-th road segment. This represents the total path length (in kilometers). Ideal travel time (hours). This represents the actual estimated travel time (in hours). The straight-line distance between the start and end points (in kilometers). The congestion impact coefficient is set to 0.2. The average congestion index for the route (obtained via the Traffic Situation API). The baseline congestion index is set at 1.0. This represents the number of road segments contained in the path.

[0061] S4: Perform real-time risk assessment on the paths in the set of compliant paths, calculate the comprehensive risk value, and prioritize the paths based on real-time traffic data. When temporary traffic control is detected in the sensitive area, return to S2 to determine candidate paths.

[0062] First, a sensitive area impact model is constructed, classifying sensitive areas along the route, such as schools, hospitals, and residential areas, into three levels according to their importance. Schools and hospitals are classified as Level 1 sensitive areas with an impact radius of 1000 meters; residential areas with a density greater than 5000 people / square kilometer are classified as Level 2 sensitive areas with an impact radius of 800 meters; and other densely populated areas are classified as Level 3 sensitive areas with an impact radius of 500 meters. When calculating the path risk coefficient, the cumulative mileage of each path within the impact range of each level of sensitive area is first calculated, and weighting coefficients are assigned according to different levels: Level 1 sensitive areas have a weight of 1.5, Level 2 sensitive areas have a weight of 1.2, and Level 3 sensitive areas have a weight of 1.0. The weighted sensitive area mileage is divided by the total path mileage to obtain the basic risk coefficient. For the processing of real-time traffic data, the main consideration is the road segment... The risk compensation mechanism considers factors such as current traffic flow, average vehicle speed, weather conditions, and road construction. When a road segment experiences congestion, the risk coefficient of that segment is increased according to the degree of congestion. For every 0.1 increase in the congestion coefficient, the risk coefficient increases by 5%. If there is road construction on the road segment, the risk coefficient increases by 20%. When encountering rain or snow, the risk coefficient of the affected road segment increases by 15%. In the dynamic risk compensation mechanism, the real-time risk coefficient of each road segment is updated every 5 minutes, and the comprehensive risk value of the entire route is recalculated. The comprehensive risk value is calculated using a segmented accumulation method, that is, the route is divided into several road segments, the risk value of each road segment is calculated separately, and then the sum is obtained. Finally, the set of compliant routes is prioritized according to the comprehensive risk value of the route, with the path having a higher priority. The ranking result is pushed to the vehicle dispatcher in real time.

[0063] First, each path is divided into multiple assessment units of fixed length. For each assessment unit, a basic risk value and a dynamic risk value are calculated. The calculation logic for the basic risk value is to identify the spatial relationship between the assessment unit and sensitive areas at all levels, count the overlapping mileage of sensitive areas at all levels within the assessment unit, multiply the overlapping mileage by the corresponding level's weight coefficient, sum them up, and then divide by the total mileage of the assessment unit to obtain the basic risk value of the unit. The calculation logic for the dynamic risk value is to obtain real-time road condition parameters within the assessment unit, including the ratio of average vehicle speed to design speed, weather conditions, road construction status, etc., match these parameters with a preset risk coefficient mapping table to obtain the risk increment corresponding to each parameter, and sum all risk increments to obtain the dynamic risk value. The basic risk value and the dynamic risk value are weighted according to a preset weight ratio to obtain the unit risk value of the assessment unit. Finally, the unit risk values ​​of all assessment units on the path are summed, and the transition risk value is calculated based on the risk value difference between adjacent assessment units. The sum of the unit risk values ​​and the transition risk value are added together to obtain the comprehensive risk value of the path.

[0064] In summary, the GPS-based vehicle positioning-based real-time route optimization method according to embodiments of the present invention has been clarified. It acquires GPS positioning data, vehicle load data, hazardous material type data, and road restriction data of hazardous materials transport vehicles, and combines this with feasibility assessment and real-time risk assessment of candidate routes to achieve dynamic route optimization and priority ranking. When temporary traffic control is detected in sensitive areas, the route can be quickly replanned. This significantly improves the safety and flexibility of hazardous materials transport route planning, effectively reduces potential risks during transportation, and ensures transportation efficiency while further enhancing transportation safety.

[0065] Here, those skilled in the art will understand that the specific operations of each step in the above-described GPS vehicle positioning-based real-time path optimization method have been referenced above. Figure 1 and Figure 2 The method for real-time path optimization based on GPS vehicle positioning has been described in detail, and therefore, its repeated description will be omitted.

[0066] In summary, the GPS-based vehicle positioning-based real-time route optimization method according to embodiments of the present invention has been clarified. It acquires GPS positioning data, vehicle load data, hazardous material type data, and road restriction data of hazardous materials transport vehicles, and combines this with feasibility assessment and real-time risk assessment of candidate routes to achieve dynamic route optimization and priority ranking. When temporary traffic control is detected in sensitive areas, the route can be quickly replanned. This invention can significantly improve the safety and flexibility of hazardous materials transport route planning, effectively reduce potential risks during transportation, and further enhance transportation safety while ensuring transportation efficiency.

Claims

1. A real-time route optimization method based on GPS vehicle positioning, characterized in that, include: Obtain GPS location data, vehicle load data, hazardous materials type data, and road restriction data for vehicles transporting hazardous materials; Based on the GPS positioning data, multiple candidate routes between the current location of the hazardous materials transport vehicle and its destination are determined; Using the vehicle load data and the hazardous materials type data as constraints, the feasibility of the candidate routes is evaluated, and a set of compliant routes that meet the constraints is selected. Real-time risk assessment is performed on the paths in the set of compliant paths, a comprehensive risk value is calculated, and the paths are prioritized based on real-time traffic data. When temporary traffic control is detected in the sensitive area, candidate paths are re-determined.

2. The real-time route optimization method based on GPS vehicle positioning according to claim 1, characterized in that, The multiple candidate paths are determined using a hierarchical road network search algorithm, which divides the electronic map road network data into a highway layer, a national and provincial highway layer, and an urban road layer, and establishes the connection relationship between road segment nodes in each layer; The main road segments are divided based on the distance between the current geographic coordinates and the target geographic coordinates, and the repetition rate is determined to obtain multiple candidate paths that meet the requirements.

3. The real-time route optimization method based on GPS vehicle positioning according to claim 2, characterized in that, The hierarchical road network search algorithm can be expressed as follows: in, Let p be the search function for path p. Let x_s be the straight-line distance from the starting point x_s to the ending point x_e. Let k be the weight of the k-th segment of the highway layer. This represents the actual length of the k-th segment of the highway layer. The speed limit for the kth segment of national, provincial, and urban roads. This represents the grade coefficient of the k-th segment of national, provincial, and urban roads. The congestion index for the k-th segment of national and provincial highways and urban roads. This represents the total number of road segments on the highway level. This refers to the total number of road sections in the national, provincial, and urban road layers.

4. The real-time route optimization method based on GPS vehicle positioning according to claim 3, characterized in that, When the main road segment is a highway, multiple different highway combination paths are calculated using the Dijkstra algorithm. Based on the distance, it is determined whether to divide the road into sub-intervals for separate search. Then, these sub-interval paths are connected to form a complete path. When the number of search nodes exceeds the preset number, the bidirectional Dijkstra algorithm is used, starting the search from both the starting point and the ending point simultaneously. The search is completed when the vertex sets of the two search directions intersect.

5. The real-time path optimization method based on GPS vehicle positioning according to claim 4, characterized in that, The Dijkstra algorithm includes the following steps: Initialize the distance from the starting point to all vertices except itself to infinity, and the distance from the starting point to itself to zero; Establish a priority queue. When the priority queue is not empty, perform the following operations: Take the first element of the priority queue as the current processing vertex; Obtain all neighboring vertices of the currently processed vertex; Calculate the new distance from the starting point through the currently processed vertex to each of the adjacent vertices; When the new distance is less than the currently recorded distance of the corresponding adjacent vertex, the distance value of the adjacent vertex is updated to the new distance, and the adjacent vertex is added to the priority queue; Record the predecessor vertex corresponding to the updated distance; Once the target vertex has been processed, the shortest path is obtained by reversing the process from the target vertex through the predecessor vertex.

6. The real-time route optimization method based on GPS vehicle positioning according to claim 4, characterized in that, The determination of the intersection of the vertex sets of the two search directions is achieved by maintaining the start and end point access sets. When the same vertex appears in both sets, the shortest total distance through that vertex is calculated and updated, and the search is terminated early when the threshold condition is met.

7. The real-time route optimization method based on GPS vehicle positioning according to claim 1, characterized in that, The constraints include hard constraints and soft constraints; The hard constraints include vehicle load constraints, hazardous materials type constraints, and time constraints. The soft constraints are evaluated using a fuzzy evaluation method, which calculates the overall score of the path by setting weight coefficients.

8. The real-time path optimization method based on GPS vehicle positioning according to claim 7, characterized in that, The soft constraint assessment score is calculated as follows: in, The path is evaluated and scored. Let i be the road grade coefficient of the i-th road segment. Let be the length of the i-th road segment. This is the total path length. For ideal travel time, This is the actual estimated travel time. The straight-line distance between the start and end points. The congestion impact coefficient, The average congestion index for the route. As the benchmark congestion index, This represents the number of road segments contained in the path.

9. The real-time path optimization method based on GPS vehicle positioning according to claim 8, characterized in that, All candidate paths that pass the hard constraint assessment are sorted from highest to lowest according to their soft constraint assessment scores, and the three paths with the highest scores are selected as the set of compliant paths.

10. The real-time path optimization method based on GPS vehicle positioning according to claim 1, characterized in that, The comprehensive risk value is calculated by dividing the path into multiple assessment units, calculating the basic risk value and dynamic risk value of each unit, weighting the two to obtain the unit risk value, summing the risk values ​​of all units, calculating the transition risk value based on the difference in risk values ​​between adjacent assessment units, and adding the sum of the unit risk values ​​to the transition risk value to obtain the comprehensive risk value of the path.

Images