Intelligent port vehicle driving intelligent cooperation method and system

Abstract

The application relates to the technical field of traffic control, in particular to a vehicle driving intelligent cooperation method and system of a smart port. The method obtains an initial path of a transport vehicle based on a port grid map; takes the initial path to be optimized as a target path, and obtains an overlap rate between the target path and the remaining initial paths; obtains an influence coefficient based on the overlap rate, uses the influence coefficient to optimize a cost function in an A-star algorithm to obtain an optimal function, and uses the A-star algorithm based on the optimal function to obtain an optimal transport path. The application has the effects of reducing congestion of port transport vehicles and improving port throughput.

Description

Technical Field

[0001] The present application relates to the field of traffic control technology, and in particular to a method and system for intelligent coordinated vehicle travel in a smart port. Background Art

[0002] With the rapid development of global trade and the growing demand for port logistics, smart ports, as an important direction for the digital transformation of ports, have become a key technology area for improving operational efficiency, reducing costs, and enhancing safety. A smart port refers to a modern port system that optimizes port operating processes by integrating advanced technologies to improve efficiency, safety, and sustainability. A smart port utilizes digital maps, real-time perception systems, and high-performance communication networks to achieve intelligent and collaborative cargo transportation, vehicle scheduling, and resource management. Its core goal is to drive port operations through technology, reduce manual intervention, lower operating costs, increase port throughput, and enhance environmental adaptability. In the process of implementing transport vehicle scheduling, it is necessary to plan the transport routes of transport vehicles. The A-star algorithm is a common path finding and graph traversal algorithm with good performance and accuracy. In the process of path planning using the A-star algorithm, a grid map, a starting point, and an end point are first input. Each grid in the grid map represents a node, and the cost of the node is calculated. The node with the lowest cost is selected from the open list and moved to the closed list. Neighboring nodes are expanded and the cost is updated. After reaching the end point, the path is traced back and the optimal path is output to complete the path planning. In the traditional A-star algorithm, the vehicle From the starting point to the node The cost function is ,in For the target vehicle to go from the starting point to the node the actual cost (such as distance or time), is the target vehicle slave node A heuristic estimate of the cost to the endpoint (e.g., Euclidean distance).

[0003] In actual vehicle transportation, the paths of different transport vehicles interact with each other. However, the traditional A-star path planning algorithm relies on static grid map information. The resulting path cannot adapt to dynamic changes in map information and does not account for the mutual influence of a large number of transport vehicles during movement. This makes smart ports prone to traffic congestion and vehicle conflicts during operation, impacting transportation safety and efficiency and posing certain security risks. Summary of the Invention

[0004] In order to solve the problem of traffic congestion or vehicle conflicts that easily occur when traditional algorithms are applied in smart ports, the present application provides a method and system for intelligent vehicle travel coordination in smart ports.

[0005] In a first aspect, the application provides a vehicle driving intelligent coordination method for a smart port, which adopts the following technical solution:

[0006] A vehicle driving intelligent coordination method for a smart port, comprising the steps of: obtaining an initial path of a transport vehicle based on a port grid map; taking the initial path to be optimized as a target path, and obtaining an overlap rate between the target path and the remaining initial paths; obtaining an influence coefficient based on the overlap rate, using the influence coefficient to optimize a cost function in an A-star algorithm to obtain an optimal function, and using the A-star algorithm based on the optimal function to obtain an optimal transport path.

[0007] The step of obtaining the overlap rate between the target path and the remaining initial paths comprises: for the target path and any initial path, obtaining coordinate sequences of the target path and the initial path, constructing optimal matching point pairs, and calculating a shortest path distance between the two coordinate sequences based on distances between the optimal matching point pairs; taking the number of optimal matching point pairs with a Euclidean distance of 0 in the optimal matching point pairs as an overlap amount, and taking a normalized result of the overlap amount as an initial overlap degree; obtaining an adjustment value based on the shortest path distance, and taking a product of the adjustment value and the initial overlap degree as the overlap rate between the target path and the initial path.

[0008] The beneficial effect is that, in the process of transporting the transport vehicle, the initial paths of all transport vehicles in the port are first obtained (which can be understood as paths obtained by a traditional path planning algorithm or paths planned by a person). For the target path of the target vehicle that needs to be adjusted, the overlap rate between the target path and the remaining initial paths is calculated. The size of the overlap rate reflects how much the target vehicle and the other initial paths overlap in the transportation process. The higher the overlap rate, the higher the risk of conflict and congestion between the target vehicle and other transport vehicles in the process of transportation scheduling based on the target path. The traditional cost function is adjusted based on the overlap rate, thereby obtaining an optimal function. Finally, the optimal transport path is obtained based on the optimal function. The optimal function combines the overlap rate between the target path and the other initial paths, so that the mutual influence between different transport vehicles is considered in the process of obtaining the optimal transport path, thereby reducing the port congestion or vehicle conflict situation and improving the port throughput.

[0009] In the calculation process of the overlap rate, the coordinate sequences between the target path and the initial paths are matched based on the DTW (Dynamic Time Warping) algorithm, and the Euclidean distances between the optimal matching point pairs are used to reflect the overlap rate between the two paths. In the optimal matching point pair, if the Euclidean distance in an optimal matching point pair is 0, it means that the nodes corresponding to the two matching points coincide, and therefore the more the number of feature points with a Euclidean distance of 0 in the two paths, the higher the overlap rate between the two paths.

[0010] Optionally, the sum of the Euclidean distances between each pair of optimal matching points is taken as the path shortest distance between the target path and the initial path.

[0011] The distance between each matching point represents the similarity between the target path and the initial path, and the smaller the path shortest distance, the greater the similarity between the target path and the initial path, and thus the higher the risk of conflict during the operation of the target vehicle. The number of coinciding parts of the collaborative nodes collectively reflects the coincidence rate between the target path and the initial path, thereby improving the robustness and accuracy of the coincidence rate calculation.

[0012] Optionally, the DTW algorithm is used to match the two coordinate sequences to obtain the optimal matching point pair.

[0013] Optionally, the step of obtaining the adjustment value based on the path shortest distance comprises: taking the normalized result of the path shortest distance as an adjustment index, and taking the calculation result of an exponential function with the adjustment index as the base and the inverse of the adjustment index as the index as the adjustment value.

[0014] The smaller the path shortest distance, the more similar the target path is to other initial paths, and thus the higher the coincidence rate between the target path and the initial path. The exponential function makes the path shortest distance inversely proportional to the final coincidence rate. Normalizing the path shortest distance facilitates subsequent calculations.

[0015] Optionally, the ratio of the path shortest distance to the number of optimal matching point pairs is taken as the normalized result of the path shortest distance.

[0016] Optionally, the step of obtaining the influence coefficient based on the coincidence rate comprises: obtaining the coordinate difference between the current target vehicle and the transport vehicle; taking the product of the coordinate difference and the corresponding coincidence rate as a local influence value; taking the sum of the local influence values corresponding to the plurality of initial paths as a global influence value, and taking the normalized result of the global influence value as the influence coefficient.

[0017] The higher the coincidence rate of a target path to an initial path, but the greater the current coordinate difference between the target vehicle and other transport vehicles, indicating that the distance between the two transport vehicles is still relatively far, and the risk of interference in the short term is low. For such cases, the influence of other vehicles on the target vehicle can be appropriately reduced, the influence coefficient is calculated based on the coordinate difference and the coincidence rate, and the cost function is adjusted through the influence coefficient, thereby further improving the effect of optimizing the target path of the target vehicle.

[0018] Optionally, the Euclidean distance between the current coordinates of the target vehicle and the initial vehicle is taken as the coordinate difference.

[0019] ​Optionally, the number of optimal matching point pairs with the Euclidean distance of 0 in the optimal matching point pair and the total number of nodes in the target path are taken as the normalized result of the coincidence amount.

[0020] In a second aspect, the vehicle driving intelligent coordination system of the smart port provided by the present application adopts the following technical solution:

[0021] A vehicle driving intelligent coordination system of a smart port, comprising a processor and a memory, the memory storing computer program instructions which, when executed by the processor, implement the vehicle driving intelligent coordination method of a smart port according to the above.

[0022] The vehicle driving intelligent coordination method of a smart port described above is generated into a computer program and stored in the memory to be loaded and executed by the processor, so that the system is made according to the memory and the processor, and is convenient to use.

[0023] The present application has the following technical effects:

[0024] The present application realizes port vehicle transportation path coordination based on the improved A-star algorithm cost function. First, the DTW algorithm is used to calculate the coincidence rate between the target path of the target vehicle and the initial paths of other running transportation vehicles, and the vehicle path sequence is asymmetrically matched to avoid traffic congestion caused by the high coincidence rate between the target path of the target vehicle and the initial paths of other transportation vehicles, and to affect the efficiency of goods transportation. Then, the distance between the current position coordinates of the target vehicle and other transportation vehicles is combined to dynamically weight the coincidence rate, avoid excessive punishment to the long-distance transportation vehicles, and plan the optimal transportation path for the target vehicle based on the improved A-star algorithm, which can improve the transportation safety and efficiency. BRIEF DESCRIPTION OF DRAWINGS

[0025] Figure 1 is a step flow chart of a vehicle driving intelligent coordination method of a smart port according to an embodiment of the present application.

[0026] Figure 2 is a method flow chart of step S2 of a vehicle driving intelligent coordination method of a smart port according to an embodiment of the present application. DETAILED DESCRIPTION

[0027] The embodiment of the application discloses a vehicle driving intelligent cooperation method of a smart port, obtains initial paths of each transport vehicle, obtains an influence coefficient of a target path of a target vehicle (a vehicle to be optimized) according to the coincidence rate between the target path and other initial paths, adjusts a cost function in a traditional A-star algorithm according to the influence coefficient, thereby obtaining an optimal function, and obtains an optimal transport path of the target vehicle based on the optimal function, thereby completing scheduling of the transport vehicle. In the method, the coincidence rate between the target path and other initial paths is used to adjust the cost function, so that the optimal transport path obtained finally takes into account the mutual influence of the movements of different transport vehicles, thereby further optimizing the path of the transport vehicle scheduling, reducing port congestion, and improving port throughput.

[0028] With reference to Figure 1 The vehicle driving intelligent cooperation method of the smart port comprises steps S1-S3.

[0029] S1: obtaining initial paths of transport vehicles based on a port grid map.

[0030] Firstly, the system obtains a digital map of the port through a drone, satellite image or existing map data. The map includes all roads, loading and unloading points, container stacking areas, buildings and other facilities. In order to facilitate calculation and path planning, the entire port area is divided into uniform grids, and passable and impassable areas are marked in the grids. This step is a conventional technical means in the art, and will not be described here.

[0031] S2: taking an initial path to be optimized as a target path, and obtaining a coincidence rate between the target path and other initial paths.

[0032] With reference to Figure 2 Step S2 comprises steps S21-S23

[0033] S21: for the target path and any initial path, obtaining coordinate sequences of the target path and the initial path, constructing optimal matching point pairs, and calculating a shortest path distance between the two coordinate sequences based on distances between the optimal matching point pairs.

[0034] A transport vehicle to be planned is taken as a target vehicle. In a traditional A-star algorithm, a center point of each grid is taken as a node, and each node corresponds to a coordinate. Start point and end point coordinates are generated in the digital map. A planned route of a vehicle (including the target vehicle) running in the port is obtained, the route is taken as an initial path, and an initial path corresponding to the target vehicle is taken as a target path. The planned route corresponding to the target vehicle is taken as a target route, and the target route is optimized to obtain an optimal transport path of the target vehicle.

[0035] For each transport vehicle, its corresponding planned route can be represented by a coordinate sequence. For example: The coordinate sequence corresponding to the planned route of the transport vehicles can be expressed as , Indicates the The transport vehicle The horizontal coordinate of each node; Indicates the The transport vehicle The vertical coordinate of the node.

[0036] Because the two vehicles' starting and ending points may differ, and the coordinates of the nodes in the overlapping transport route segments are located in different locations in their respective sequences, the DTW (Dynamic Time Warping) algorithm can be used to stretch or compress the two route sequences to obtain multiple optimal matching point pairs. The sum of the Euclidean distances between these multiple optimal matching point pairs is used as the shortest path distance.

[0037] S22: The number of optimal matching point pairs with a Euclidean distance of 0 among the optimal matching point pairs is used as the overlap amount, and the normalized result of the overlap amount is used as the initial overlap degree.

[0038] Specifically, the calculation formula of the initial overlap can be expressed as: Where, represents the initial overlap; Indicates the number of optimal matching points whose Euclidean distance between the two matching points is 0; Indicates the total number of nodes in the target path.

[0039] Where, This value represents the number of optimal matching point pairs whose Euclidean distance between the two matching points is 0. A Euclidean distance of 0 between two matching points indicates that the two nodes overlap. The number of overlapping nodes in the two paths can be used to reflect the overlap rate. The more overlapping nodes there are between the target path and the initial path, the higher the overlap rate between the target and initial paths.

[0040] S23: Obtaining an adjustment value based on the shortest distance of the path, and taking the product of the adjustment value and the initial overlap as the overlap ratio between the target path and the initial path.

[0041] The normalized result of the shortest distance of the path is used as the adjustment index, and the The calculation result of the exponential function with base α and the opposite number of the adjustment index as the exponent is used as the adjustment value.

[0042] Specifically, the calculation formula of the adjustment value can be expressed as: Where, Indicates the adjustment value; represents the shortest path distance between the target path and one of the initial paths; the smaller the shortest path distance is, the higher the similarity between the coordinate sequences corresponding to the target path and the initial path is, and the higher the coincidence rate between the two paths is. represents the number of the optimal matching point pairs represents an exponential function with a natural constant as a base.

[0043] In the formula, represents the shortest path distance between the target path and one of the initial paths; the smaller the shortest path distance is, the higher the similarity between the coordinate sequences corresponding to the target path and the initial path is, and the higher the coincidence rate between the two paths is. is used to normalize the shortest path distance, facilitating subsequent calculation.

[0044] After the adjustment value is obtained, the product of the adjustment value and the initial coincidence degree is taken as the coincidence rate between the target path and the initial path. Here, the adjustment value and the initial coincidence degree are mainly used to jointly reflect the coincidence rate between the target path and the initial path, and in other embodiments, an addition manner can also be used to jointly reflect the coincidence rate between the two paths.

[0045] S3: Obtain an influence coefficient based on the coincidence rate, use the influence coefficient to optimize a cost function in the A-star algorithm to obtain an optimal function, and use the A-star algorithm based on the optimal function to obtain an optimal transportation path.

[0046] In the traditional cost function formula, the path cost of a node is the sum of the actual cost of the target vehicle from the starting point to the node and the heuristic estimated cost from the node to the end point, and direct addition cannot well reflect the influence of path coincidence on path cost, so here the influence coefficient is calculated through the coincidence amount, and the influence coefficient acts on the cost function in the traditional A-star algorithm.

[0047] Specifically, the coordinate difference between the current target vehicle and the transportation vehicle is obtained; the product of the coordinate difference and the corresponding coincidence rate is taken as a local influence value; the sum of the local influence values corresponding to the plurality of initial paths is taken as an overall influence value, and the normalized result of the overall influence value is taken as the influence coefficient.

[0048] Specifically, the optimal function can be represented by the following formula:

[0049] ; in the formula, represents the optimal function; represents a cost function in the traditional A-star algorithm; represents the coincidence rate between the target path and the initial path of the first transportation vehicle; represents the coincidence rate between the target vehicle and the first transportation vehicle. represents the coincidence rate between the target vehicle The Euclidean distance between the current position of the transport vehicle; Indicates the total number of transport vehicles in the port excluding the target vehicle; Indicates An exponential function with base .

[0050] Where, Indicates the target vehicle The target path and The larger the value, the higher the overlap rate between the target path and the initial path, which means the greater the risk of traffic congestion or conflict with other transport vehicles on the target path.

[0051] Indicates the target vehicle Target path and The Euclidean distance between the current positions of the target vehicle and the other transport vehicles—that is, the coordinate difference between the target vehicle and the other transport vehicles—reflects the current positional relationship between the target vehicle and the remaining transport vehicles. In actual transportation, if the target path has a high overlap with the initial path, but the current actual distance is large, this indicates a low short-term risk of conflict between the two, and thus a lower final cost. Conversely, if the distance between the target vehicle and the other transport vehicles is small, the final cost function value should be increased.

[0052] The first part represents the local impact value of the initial path of one of the transport vehicles on the target vehicle. The local impact values ​​of all vehicles except the target vehicle are traversed and accumulated to obtain the overall impact value of all remaining transport vehicles on the target path. Finally, the normalized result of the overall impact value is used as the impact coefficient to facilitate subsequent calculations.

[0053] Based on the optimal function, the target vehicle is calculated using the A-star algorithm. The cost of each node on the transport path is selected, and the node with the smallest cost function value is selected until the end point is found and the target vehicle is obtained. The optimal transport route is determined by minimizing multi-vehicle conflicts while ensuring safe and rapid cargo transportation, thus achieving coordinated transport. Using the optimal function as the cost function in the A-star algorithm to plan the route is a conventional technique in the field and will not be elaborated on here.

[0054] The above are all preferred embodiments of the present application, and are not intended to limit the scope of protection of the present application. Therefore, any equivalent changes made based on the structure, shape, and principle of the present application should be included in the scope of protection of the present application.

Claims

1. A vehicle driving intelligent coordination method for a smart port, characterized by: The method includes the following steps: obtaining an initial path for transport vehicles based on a port grid map; taking the initial path to be optimized as the target path, obtaining the overlap rate between the target path and the remaining initial paths; obtaining an influence coefficient based on the overlap rate, using the influence coefficient to optimize the cost function in the A-star algorithm to obtain an optimal function, and using the A-star algorithm based on the optimal function to obtain the optimal transport path; the expression of the optimal function is: Where, represents the optimal function; Represents the cost function in the traditional A-star algorithm; Indicates the target path With the The overlap rate between the initial paths of the transport vehicles; Indicates the target vehicle With the The Euclidean distance between the current position of the transport vehicle; Indicates the total number of transport vehicles in the port excluding the target vehicle; Indicates An exponential function with base ; The step of obtaining the overlap rate between the target path and the remaining initial paths includes: for the target path and any initial path, obtaining the coordinate sequences of the target path and the initial path, constructing the optimal matching point pair, and calculating the shortest path distance between the two coordinate sequences based on the distance between the optimal matching point pairs; taking the number of optimal matching point pairs with a Euclidean distance of 0 in the optimal matching point pairs as the overlap amount, and taking the normalized result of the overlap amount as the initial overlap degree; obtaining an adjustment value based on the shortest path distance, and taking the product of the adjustment value and the initial overlap degree as the overlap rate between the target path and the initial path; the calculation formula of the adjustment value is: Where, Indicates the adjustment value; represents the shortest path distance between the target path and one of the initial paths; represents the number of optimal matching point pairs, Represents an exponential function with a natural constant as the base.

2. The vehicle driving intelligent coordination method of a smart port according to claim 1 is characterized in that: The sum of the Euclidean distances between each optimal matching point pair is taken as the shortest path distance between the target path and the initial path.

3. The vehicle driving intelligent coordination method of a smart port according to claim 1 is characterized in that: Use the DTW algorithm to match the two coordinate sequences and obtain the optimal matching point pair.

4. The vehicle driving intelligent coordination method of a smart port according to claim 1 is characterized in that: The step of obtaining the adjustment value based on the shortest path distance includes: taking the normalized result of the shortest path distance as the adjustment index; The calculation result of the exponential function with base α and the opposite number of the adjustment index as the exponent is used as the adjustment value.

5. The vehicle driving intelligent coordination method of a smart port according to claim 4 is characterized in that: The ratio of the shortest path distance to the number of optimal matching point pairs is taken as the normalized result of the shortest path distance.

6. The vehicle driving intelligent coordination method of a smart port according to claim 1 is characterized in that: The steps of obtaining the influence coefficient based on the overlap rate include: obtaining the coordinate difference between the current target vehicle and the transport vehicle; taking the product of the coordinate difference and the corresponding overlap rate as the local influence value; taking the sum of the local influence values ​​corresponding to multiple initial paths as the overall influence value, and taking the normalized result of the overall influence value as the influence coefficient.

7. The vehicle driving intelligent coordination method of a smart port according to claim 6 is characterized in that: The Euclidean distance between the current coordinates of the target vehicle and the initial vehicle is taken as the coordinate difference.

8. The vehicle driving intelligent coordination method of a smart port according to claim 1 is characterized in that: The number of optimal matching point pairs with a Euclidean distance of 0 in the optimal matching point pairs and the total number of nodes in the target path are taken as the normalized result of the overlap.

9. A smart port vehicle driving intelligent coordination system, characterized by: include: A processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, a vehicle driving intelligent coordination method for a smart port according to any one of claims 1 to 8 is implemented.

Images