A customized passenger route planning method and system

Abstract

The present invention relates to the technical field of customized passenger transport services, and more specifically, to a customized passenger transport route planning method and system, the method comprising: obtaining the initial route planning results and order sites of a passenger vehicle and constructing graph structure data; in response to the appearance of a new high-priority order for a passenger vehicle, re-planning the route with the coordinate point of the current passenger vehicle as the starting point and the coordinate point of the high-priority order as the end point; obtaining the initial probability distribution of each passenger vehicle based on the actual efficiency of the coordinate point of each passenger vehicle to the coordinate point of the high-priority order in the graph structure data; further correcting the initial probability distribution to obtain a corrected probability distribution model, performing a new route planning result, and completing the customized passenger transport service according to the new route planning result. The present invention dynamically responds to high-priority orders and quickly re-plans the route, effectively reducing vehicle idle driving rate and response delay, and reducing detour costs.

Description

Technical Field

[0001] The present invention relates to the technical field of customized passenger transport services, and more particularly to a customized passenger transport route planning method and system. Background Art

[0002] Customized passenger transport is a personalized and flexible mode of transportation, commonly seen in ride-hailing and corporate shuttle services. It aims to meet passengers' specific travel needs, providing convenient and efficient travel options, deploying vehicles on demand, and enabling point-to-point transportation. Compared to traditional passenger transport, it offers greater flexibility and targeted services.

[0003] Route planning is crucial for customized passenger transport services. Reasonable route planning can effectively reduce operating costs, improve vehicle utilization, shorten passenger wait times, and enhance the passenger travel experience. Through precise planning, vehicles can avoid congested roads and follow optimal routes, achieving efficient resource allocation and ensuring smooth and efficient passenger transport operations. This is a key factor in enhancing the competitiveness of customized passenger transport.

[0004] Current customized passenger transport systems mostly use fixed-stop preset routes or global replanning strategies. Fixed-stop preset routes are unable to accommodate new orders along the way, often resulting in vehicle idle rates exceeding 30%. In the event of sudden orders or traffic congestion, the global replanning strategy requires full route recalculation, resulting in long response delays. Frequent route changes can also cause route fluctuations, such as repeated vehicle U-turns, which seriously impact operational efficiency and passenger experience. When high-priority orders are inserted or traffic conditions change suddenly, existing solutions face bottlenecks: simply inserting new stops will cause detour costs to soar, undermining the ride-sharing revenue model; it is difficult to collaboratively optimize time window fulfillment, mileage costs, and order revenue, leading to increased default rates or wasted resources. Summary of the Invention

[0005] To address the current limitations of customized passenger transport systems due to fixed-stop preset routes and global replanning strategies, such as high vehicle idle rates, response delays, route fluctuations, and difficulty in processing high-priority orders, which seriously affect operational efficiency and passenger experience, the present invention provides solutions in the following aspects.

[0006] In a first aspect, a customized passenger route planning method includes: obtaining the initial route planning results and order sites of a passenger vehicle, wherein the order site is the coordinate point of the passenger vehicle, and constructing graph structure data based on the coordinate points; in response to the emergence of a new high-priority order for the passenger vehicle, the coordinate point of the current passenger vehicle is used as the starting point of the new path planning, and the coordinate point of the high-priority order is used as the end point of the new path planning, and the path planning is re-performed; based on the actual efficiency of the coordinate point of each passenger vehicle to the coordinate point of the high-priority order in the graph structure data, the initial probability distribution of each passenger vehicle is obtained; the initial probability distribution is further corrected to obtain a corrected probability distribution model, and a new path planning result is performed, and the passenger vehicle is controlled to perform high-priority order services according to the new path planning result.

[0007] Its effect is: by dynamically responding to new high-priority orders and quickly replanning routes, it effectively reduces vehicle idle rate and response delay, while optimizing resource allocation based on actual efficiency and probability distribution models, reducing detour costs, and collaboratively optimizing time window fulfillment, mileage costs and order revenue, thereby improving operational efficiency and passenger satisfaction, ensuring that key orders are served in a timely manner, and enhancing the responsiveness of passenger vehicles to high-priority orders.

[0008] Preferably, the constructing of graph structure data includes:

[0009] The coordinate points of passenger vehicles are used as nodes in the graph structure. According to the path distance in the initial route planning result, the edges and weights between nodes are used to represent them. The edges and weights of all nodes are used to construct a completely undirected graph structure data, and it is determined whether the starting point and the end point are consistent.

[0010] The result is a completely undirected graph-structured data structure, using passenger vehicle coordinates as nodes and path distances as edges and weights. This allows for a comprehensive and accurate representation of the spatial relationships and path information between passenger vehicles. This provides a precise data foundation for subsequent route planning, making it more scientific and rational, enabling rapid and accurate response to new high-priority orders, enabling efficient and flexible route adjustments, and improving operational efficiency and service quality.

[0011] Preferably, the actual effective rate includes:

[0012] Calculate the projection value of the vector between the coordinate point of each passenger vehicle and the coordinate point of the high-priority order and the vector between the coordinate point of the current passenger vehicle and the coordinate point of the high-priority order, and take the ratio of the projection value to the modulus of the vector between the coordinate point of each passenger vehicle and the coordinate point of the high-priority order as the actual efficiency of the coordinate point of each passenger vehicle.

[0013] The effect is that by calculating the ratio of the vector projection value and the modulus, we can quantitatively assess the actual efficiency of each passenger vehicle's coordinate point to the coordinate point of the high-priority order. The actual efficiency reflects the efficiency and priority of each passenger vehicle in responding to high-priority orders, providing an important weighting factor for subsequent probability distribution calculations. By considering the spatial relationship between vectors, passenger resources can be more rationally allocated, prioritizing those points that are more important or urgent to the overall mission, thereby improving the performance and reliability of the entire system.

[0014] Preferably, the actual effective rate also includes:

[0015] Use the dot product formula to calculate the product of the vector from the coordinate point of the current passenger vehicle to the coordinate point of the high-priority order and the vector from the coordinate point of each passenger vehicle to the coordinate point of the high-priority order;

[0016] Calculate the product of the modules of the vector from the coordinate point of the current passenger vehicle to the coordinate point of the high-priority order and the vector from the coordinate point of each passenger vehicle to the coordinate point of the high-priority order, and use the cosine function of the ratio between the product of the vectors and the product of the modules as the actual efficiency of the coordinate point of each passenger vehicle to the coordinate point of the high-priority order.

[0017] The effect is that by calculating the cosine value between the vector dot product and the product of the module, the similarity between the current path and the path directly leading to high-priority orders can be accurately measured, thereby evaluating the effectiveness of the path. By identifying and prioritizing those paths that are more similar to the direct path, the path planning time can be reduced and the planning efficiency can be improved.

[0018] Preferably, the corrected probability distribution model includes:

[0019] Using the coordinate point of the current passenger vehicle as the coordinate origin, reconstruct the coordinate system, obtain the relative coordinates of the remaining order stations, and calculate the straight-line Euclidean distance from the current passenger vehicle to the remaining order stations, where the remaining order stations are the order stations other than the terminal point;

[0020] Use a negative exponential function to correct the product of the hyperparameter and the straight-line Euclidean distance to obtain a significant weight value. Multiply the significant weight value by the corresponding coordinate point to obtain a weighted coordinate point. Use the weighted coordinate point to fit a three-dimensional Gaussian function.

[0021] The mean of the three-dimensional Gaussian function is used as the actual coordinate of the starting point in the road network space, and the product of the three-dimensional Gaussian function and the Gaussian function corresponding to the initial probability distribution is used as the corrected probability distribution model.

[0022] The effect is: by reconstructing the coordinate system with the coordinate point of the current passenger vehicle as the coordinate origin, the distance to each order site can be calculated more accurately, thereby more accurately evaluating the importance and priority of each point. By using a negative exponential function to correct the distance, a significant weight value can be obtained. This weight reflects the position and importance of each order site relative to the coordinate point of the current passenger vehicle, which helps to dynamically adjust the path planning.

[0023] Preferably, the calculation method of the initial probability distribution includes:

[0024] Taking the current passenger vehicle as the target device, taking the midpoint of the economically optimal path from the target device's order site to the destination as the center, and the variance as the Gaussian function value, construct a Gaussian function and normalize the Gaussian function;

[0025] The peak value of the normalized Gaussian function is adjusted according to the product of the hyperparameter and the straight-line distance from the end point to the end point of the economically optimal path from the order site to the end point of the target device, and the initial probability distribution of the economically optimal path from the order site to the end point of the target device is obtained.

[0026] Preferably, the calculation method of the Gaussian function includes:

[0027] The midpoint of the economically optimal path from the target site to the destination is taken as the mean of the Gaussian function, and the product of the hyperparameter and the actual efficiency of the coordinate point from the target site to the high-priority order is taken as the standard deviation. The hyperparameter is used to adjust the width of the distribution, calculate the standard deviation, and perform normalization.

[0028] Calculate the square of the distance difference between the midpoints of the economically optimal path from the target station to the end point, divide the square of the distance difference by 2 times the standard deviation, and use a negative exponential function for exponential decay;

[0029] The exponential decay result is used as the exponent of the normalized result to obtain the probability of the target station being selected between the coordinate point of the current passenger vehicle and the coordinate point of the high-priority order.

[0030] In a second aspect, a customized passenger route planning system includes: a processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the customized passenger route planning method described above is implemented.

[0031] The present invention has the following effects:

[0032] 1. This invention dynamically responds to new high-priority orders and rapidly reroutes passenger vehicles, using the coordinates of the high-priority orders as the endpoints of the newly planned routes. This reduces the time it takes to reach high-priority orders, thereby improving operational efficiency. By rapidly adjusting routes and prioritizing high-priority orders, this ensures that critical orders receive timely service, effectively enhancing the responsiveness of passenger vehicles to high-priority orders.

[0033] 2. This invention calculates the actual efficiency of each passenger vehicle's coordinate point to the coordinate point of the high-priority order, and uses this to obtain the initial probability distribution of each passenger vehicle. This is then further modified to form a corrected probability distribution model, enabling a more rational allocation of passenger transport resources. This effectively reduces detour costs and vehicle idleness, while simultaneously optimizing time window fulfillment, mileage costs, and order revenue. This ensures efficient utilization of passenger vehicle resources, enhances the passenger travel experience, reduces wait times, and improves passenger satisfaction. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] The above and other objects, features and advantages of the exemplary embodiments of the present invention will become readily understood by reading the following detailed description with reference to the accompanying drawings. In the accompanying drawings, several embodiments of the present invention are shown in an illustrative and non-limiting manner, and the same or corresponding reference numerals represent the same or corresponding parts, wherein:

[0035] Figure 1 This is a method flow chart of steps S1 to S4 in a customized passenger route planning method according to an embodiment of the present invention.

[0036] Figure 2 This is a structural block diagram of a customized passenger route planning system according to an embodiment of the present invention. DETAILED DESCRIPTION

[0037] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work shall fall within the scope of protection of the present invention.

[0038] The specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0039] Reference Figure 1 A customized passenger route planning method includes steps S1 to S4, specifically as follows:

[0040] S1: Obtain the initial route planning results and order sites of the passenger vehicle, where the order sites are the coordinate points of the passenger vehicle, and construct graph structure data based on the coordinate points.

[0041] The coordinate points of passenger vehicles are used as nodes in the graph structure. According to the path distance in the initial route planning result, the edges and weights between nodes are used to represent them. The edges and weights of all nodes are used to construct a completely undirected graph structure data. , and determine whether the starting point and end point are consistent.

[0042] For example, in a customized passenger transport service, a passenger vehicle might depart from a certain station and return to its starting point after completing all orders. The starting point and the end point may or may not be the same. If they are the same, they are represented as one node in the graph; if they are different, they are represented as two nodes. To prevent the path planning algorithm from directly connecting the starting point to the end point (skipping other order stations), the edge weight between the two nodes is set to infinite, ensuring that the algorithm selects a path that passes through all necessary order stations.

[0043] It should be further explained that after constructing a completely undirected graph, when customizing passenger services, a new high-priority order may appear at a single order site, but the passenger vehicle may be on the path between two order sites. If it follows the original planned path to reach the site to be ordered, it may take too long. If a new path is planned to the site to be ordered starting from the current passenger vehicle, the planning of the new path may be time-consuming, which may lead to a long planning time for the new path, thereby affecting the timely service of high-priority orders.

[0044] S2: In response to a new high-priority order for a passenger vehicle, the coordinate point of the current passenger vehicle is used as the starting point of a new path planning, and the coordinate point of the high-priority order is used as the end point of the new path planning, and the path planning is re-performed.

[0045] It is further explained that in order to plan routes for high-priority orders in a timely manner, when a high-priority order is discovered, the initial route planning results are used to optimize the probability distribution when generating the initial solution, so that when a new route is planned using the road network space, the new route planning results can be quickly obtained to complete the customized passenger transport service.

[0046] Get the corresponding coordinate point in the current passenger vehicle real road network , where each order site corresponds to a coordinate point. When planning a new path, Indicates the starting point of the new path planning , where the end point is the coordinate corresponding to the destination of the high priority order ; It can also be understood as taking the current passenger vehicle when a high-priority order is discovered as the starting point and the high-priority order as the end point.

[0047] S3: Based on the actual efficiency of the coordinate points of each passenger vehicle to the coordinate points of the high-priority orders in the graph structure data, an initial probability distribution of each passenger vehicle is obtained.

[0048] It should be noted that obtaining and The vector corresponding to the two coordinate points And the modulus of the vector, where the vector is from point to , the module length is The straight-line Euclidean distance between the two coordinate points.

[0049] Among them, if and The paths corresponding to the order sites between The higher the overlap, the closer it is to the shortest straight line distance, and the greater the probability of selecting a path, and then obtaining the corresponding coordinate points of each order site to the end point vector.

[0050] Calculate the projection value of the vector between the coordinate point of each passenger vehicle and the coordinate point of the high-priority order and the vector between the coordinate point of the current passenger vehicle and the coordinate point of the high-priority order, and take the ratio of the projection value to the modulus of the vector between the coordinate point of each passenger vehicle and the coordinate point of the high-priority order as the actual efficiency of the coordinate point of each passenger vehicle.

[0051] Specifically, the actual efficiency satisfies the following relationship:

[0052] ;

[0053] Where, Indicates the The actual efficiency of the order site to the coordinate point of the high-priority order, Indicates the The projection value of the vector from the coordinate point of the order site to the coordinate point of the high-priority order on the vector between the current passenger vehicle order site and the high-priority order site, Indicates the The vector of the coordinate point corresponding to the order site to the coordinate point of the high-priority order.

[0054] That is to say, the greater the actual efficiency, the closer the new path planning result is to the first The more economically optimal the path is from an order site to the destination, the more efficient the new path planning result will be.

[0055] Further analysis, based on A Gaussian model is constructed with the midpoint of the economically optimal path from each order site to the destination as the central coordinate point. The Gaussian model is selected, which means that the closer to the center point, the greater the probability of selecting this location as the new path, because the center point of the Gaussian model is the peak point, and as the variance of the Gaussian function increases, the radiation range of the surrounding coordinate points being selected becomes larger.

[0056] In addition, another embodiment further includes:

[0057] Use the dot product formula to calculate the product of the vector from the coordinate point of the current passenger vehicle to the coordinate point of the high-priority order and the vector from the coordinate point of each passenger vehicle to the coordinate point of the high-priority order;

[0058] Calculate the product of the modules of the vector from the coordinate point of the current passenger vehicle to the coordinate point of the high-priority order and the vector from the coordinate point of each passenger vehicle to the coordinate point of the high-priority order, and use the cosine function of the ratio between the product of the vectors and the product of the modules as the actual efficiency of the coordinate point of each passenger vehicle to the coordinate point of the high-priority order.

[0059] Specifically, the actual efficiency satisfies the following relationship:

[0060] ;

[0061] Where, Indicates the The actual efficiency of the order site to the coordinate point of the high-priority order, The vector representing the coordinate point of the current passenger vehicle to the coordinate point of the high-priority order, Indicates the The vector from the coordinate point corresponding to the order site to the coordinate point of the high-priority order, Represents the cosine function.

[0062] That is, calculate the The cosine value between the vector from the coordinate point of each order site to the coordinate point of the high-priority order and the vector from the current passenger vehicle order site to the order site of the high-priority order can be used to obtain the similarity of the two vectors. The closer the cosine value is to 1, the more similar the two vectors are, that is, the closer the path is to the optimal path; the actual efficiency reflects the similarity between the path from the coordinate point of each passenger vehicle to the coordinate point of the high-priority order and the shortest straight-line path. The greater the actual efficiency, the greater the degree of overlap between the path and the shortest straight-line path, and the closer the path is to the optimal path.

[0063] It should also be noted that The midpoint of the economically optimal path from the first order station to the destination is used as the central coordinate point to build a Gaussian model. The Gaussian model indicates that the closer to the center point, the greater the probability of selecting this location as the new driving path. This is because the center point of the Gaussian model is the peak point, and as the variance of the Gaussian function increases, the radiation range of the surrounding coordinate points is larger. However, as the variance increases, the peak point will drop more. What we need is that the larger the variance, the larger the peak point. Therefore, we need to keep the peak value of the Gaussian function at 1, and then multiply it by the variance to obtain the first The initial probability distribution corresponding to the economically optimal path from the order site to the destination.

[0064] Obtaining the initial probability distribution includes the following steps:

[0065] Taking the current passenger vehicle as the target device, taking the midpoint of the economically optimal path from the target device's order site to the destination as the center, and the variance as the Gaussian function value, construct a Gaussian function and normalize the Gaussian function;

[0066] The peak value of the normalized Gaussian function is adjusted according to the product of the hyperparameter and the straight-line distance from the end point to the end point of the economically optimal path from the order site to the end point of the target device, and the initial probability distribution of the economically optimal path from the order site to the end point of the target device is obtained.

[0067] Specifically, the initial probability distribution satisfies the following relationship:

[0068] ;

[0069] Where, Indicates the The initial probability distribution of the economically optimal path from the order site to the destination, represents the Gaussian function (normal distribution), Indicates the The midpoint of the economically optimal path from the order site to the destination is the central coordinate point. represents the maximum value of the Gaussian function distribution, 、 They represent hyperparameters, Indicates the The actual efficiency of transporting order sites to the coordinate points of high-priority orders.

[0070] For example, in this embodiment, , , which can be adjusted by implementers according to specific implementation scenarios.

[0071] The midpoint of the economically optimal path from the target site to the destination is taken as the mean of the Gaussian function, and the product of the hyperparameter and the actual efficiency of the coordinate point from the target site to the high-priority order is taken as the standard deviation. The hyperparameter is used to adjust the width of the distribution, calculate the standard deviation, and perform normalization.

[0072] Calculate the square of the distance difference between the midpoints of the economically optimal path from the target station to the end point, divide the square of the distance difference by 2 times the standard deviation, and use a negative exponential function for exponential decay;

[0073] The exponential decay result is used as the exponent of the normalized result to obtain the probability of the target station being selected between the coordinate point of the current passenger vehicle and the coordinate point of the high-priority order.

[0074] Specifically, the Gaussian function distribution satisfies the following relationship:

[0075] ;

[0076] Where, Indicates the distance between the coordinate point of the current passenger vehicle and the coordinate point of the high-priority order. The probability that an order site is selected, represents the hyperparameter, Indicates the The actual efficiency of the order site to the coordinate point of the high-priority order, Indicates the distance between the coordinate point of the current passenger vehicle and the coordinate point of the high-priority order. Order stations, Indicates the The midpoint of the economically optimal path from an order site to the destination is the central coordinate point.

[0077] However, as the variance increases, the probability value of the Gaussian function will decrease, and then the peak of the Gaussian function needs to be enhanced. However, due to the large variance change, there may be non-uniform problems, resulting in a chaotic final probability distribution, and then the peak of the Gaussian function needs to be normalized. After normalization, the Gaussian function is further enhanced according to the size of the variance. When normalizing, the Gaussian function is multiplied by the inverse of the peak value so that the maximum value corresponding to the peak point is 1, and then the peak value of the current Gaussian function is obtained. , get the normalized Gaussian function , and then negatively correlate the variance, using implementation, where The value range is , thus, the larger the variance, the larger the Gaussian probability value.

[0078] By optimizing the probability distribution of the initial path planning results, an initial solution close to the optimal path is quickly generated. By adjusting the probability distribution, the path planning algorithm is more inclined to choose a path close to the optimal path. Using the optimized initial solution, the path planning algorithm quickly generates the optimal path. With this method, high-priority orders can be quickly responded to in a complex road network space, ensuring that passenger vehicles can reach high-priority orders in a timely manner.

[0079] It should be noted that after obtaining the initial distribution probability value corresponding to each data point in the road network space, although the initial path selection for new route planning can be performed based on the initial probability distribution value, since a distribution probability value can be obtained near each order station, there may be passable and close traffic in the road network space. The closest points are just locations that are far away from other order sites, and these locations cannot obtain higher initial probability values ​​through the above method.

[0080] Although it is possible to calculate the distance between each point in the road network space However, this distance does not represent the travel distance. However, if we calculate the distance from any point in the road network space to If the economic optimal distance is exceeded, the computational cost will be too high.

[0081] Then the initial probability value is further corrected as follows:

[0082] S4: Further modify the initial probability distribution to obtain a corrected probability distribution model, and perform new path planning results. According to the new path planning results, passenger vehicles are controlled to execute high-priority order services.

[0083] Corrected probability distribution model, including:

[0084] Using the coordinate point of the current passenger vehicle as the coordinate origin, reconstruct the coordinate system, obtain the relative coordinates of the remaining order stations, and calculate the straight-line Euclidean distance from the current passenger vehicle to the remaining order stations, where the remaining order stations do not include the end point;

[0085] Use a negative exponential function to correct the product of the hyperparameter and the straight-line Euclidean distance to obtain a significant weight value. Multiply the significant weight value by the corresponding coordinate point to obtain a weighted coordinate point. Use the weighted coordinate point to fit a three-dimensional Gaussian function.

[0086] The mean of the three-dimensional Gaussian function is used as the actual coordinate of the starting point in the road network space, and the product of the three-dimensional Gaussian function and the Gaussian function corresponding to the initial probability distribution is used as the corrected probability distribution model.

[0087] In this embodiment, the weight is calculated by using The function performs negative correlation mapping on the obtained straight-line Euclidean distance, and during the negative correlation mapping, assigns a hyperparameter of 2 to the straight-line Euclidean distance as an adjustment coefficient, that is, the weight corresponding to the starting point is 1.

[0088] The swarm intelligence algorithm may be a genetic algorithm, a particle swarm algorithm, an ant colony algorithm or a greedy wolf algorithm.

[0089] Then, new path planning results can be obtained, and passenger vehicles can be controlled to perform high-priority order services, thereby completing customized passenger services.

[0090] The present invention also provides a customized passenger route planning system. Figure 2 As shown, the system includes a processor and a memory, wherein the memory stores computer program instructions. When the computer program instructions are executed by the processor, a customized passenger route planning method according to the first aspect of the present invention is implemented.

[0091] The system also includes other components well known to those skilled in the art, such as a communication bus and a communication interface. The configuration and functions of these components are known in the art and therefore will not be described in detail here.

[0092] In the present invention, the aforementioned memory can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. For example, the computer-readable storage medium can be any suitable magnetic or magneto-optical storage medium, such as Resistive Random Access Memory (RRAM), Dynamic Random Access Memory (DRAM), Static Random Access Memory (SRAM), Enhanced Dynamic Random Access Memory (EDRAM), High-Bandwidth Memory (HBM), Hybrid Memory Cube (HMC), etc., or any other medium that can be used to store the required information and can be accessed by an application, module, or both. Any such computer storage medium can be part of, accessible to, or connectable to a device. Any application or module described in the present invention can be implemented by computer-readable / executable instructions stored or otherwise retained by such a computer-readable medium.

[0093] In the description of this specification, "multiple" and "several" mean at least two, such as two, three or more, etc., unless otherwise clearly defined.

[0094] While several embodiments of the present invention have been shown and described herein, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Numerous modifications, variations, and alternatives will occur to those skilled in the art without departing from the concept and spirit of the present invention. It should be understood that various alternatives to the embodiments of the present invention described herein may be employed in practicing the present invention.

Claims

1. A customized passenger route planning method, characterized in that: include: Obtain the initial route planning results and order stations of passenger vehicles, where the order stations are the coordinate points of passenger vehicles, and construct graph structure data based on the coordinate points; In response to a new high-priority order for a passenger vehicle, the coordinate point of the current passenger vehicle is used as the starting point of the new route planning, and the coordinate point of the high-priority order is used as the end point of the new route planning, and the route planning is re-performed; According to the actual efficiency of the coordinate points of each passenger vehicle to the coordinate points of the high-priority order in the graph structure data, the initial probability distribution of each passenger vehicle is obtained; The actual efficiency includes: calculating the projection value of the vector between the coordinate point of each passenger vehicle and the coordinate point of the high-priority order and the vector between the coordinate point of the current passenger vehicle and the coordinate point of the high-priority order, and taking the ratio between the projection value and the modulus of the vector between the coordinate point of each passenger vehicle and the coordinate point of the high-priority order as the actual efficiency of the coordinate point of each passenger vehicle; Alternatively, a dot product formula is used to calculate the product of the vector from the coordinate point of the current passenger vehicle to the coordinate point of the high-priority order and the vector from the coordinate point of each passenger vehicle to the coordinate point of the high-priority order; the module product of the vector from the coordinate point of the current passenger vehicle to the coordinate point of the high-priority order and the vector from the coordinate point of each passenger vehicle to the coordinate point of the high-priority order is calculated, and the cosine function of the ratio between the product of the vectors and the product of the modules is used as the actual efficiency of the coordinate point of each passenger vehicle to the coordinate point of the high-priority order; The initial probability distribution is further modified to obtain a corrected probability distribution model, and a new path planning result is performed. According to the new path planning result, passenger vehicles are controlled to execute high-priority order services.

2. A customized passenger route planning method according to claim 1, characterized in that: The constructing of graph structure data includes: The coordinate points of passenger vehicles are used as nodes in the graph structure. According to the path distance in the initial route planning result, the edges and weights between nodes are used to represent them. The edges and weights of all nodes are used to construct a completely undirected graph structure data, and it is determined whether the starting point and the end point are consistent.

3. A customized passenger route planning method according to claim 1, characterized in that: The corrected probability distribution model includes: Using the coordinate point of the current passenger vehicle as the coordinate origin, reconstruct the coordinate system, obtain the relative coordinates of the remaining order stations, and calculate the straight-line Euclidean distance from the current passenger vehicle to the remaining order stations, where the remaining order stations are the order stations other than the terminal point; Use a negative exponential function to correct the product of the hyperparameter and the straight-line Euclidean distance to obtain a significant weight value. Multiply the significant weight value by the corresponding coordinate point to obtain a weighted coordinate point. Use the weighted coordinate point to fit a three-dimensional Gaussian function. The mean of the three-dimensional Gaussian function is used as the actual coordinate of the starting point in the road network space, and the product of the three-dimensional Gaussian function and the Gaussian function corresponding to the initial probability distribution is used as the corrected probability distribution model.

4. A customized passenger route planning method according to claim 1, characterized in that: The calculation method of the initial probability distribution includes: Taking the current passenger vehicle as the target device, taking the midpoint of the economically optimal path from the target device's order site to the destination as the center, and the variance as the Gaussian function value, construct a Gaussian function and normalize the Gaussian function; The peak value of the normalized Gaussian function is adjusted according to the product of the hyperparameter and the straight-line distance from the end point to the end point of the economically optimal path from the order site to the end point of the target device, and the initial probability distribution of the economically optimal path from the order site to the end point of the target device is obtained.

5. A customized passenger route planning method according to claim 4, characterized in that: The calculation method of the Gaussian function includes: The midpoint of the economically optimal path from the target site to the destination is taken as the mean of the Gaussian function, and the product of the hyperparameter and the actual efficiency of the coordinate point from the target site to the high-priority order is taken as the standard deviation. The hyperparameter is used to adjust the width of the distribution, calculate the standard deviation, and perform normalization. Calculate the square of the distance difference between the midpoints of the economically optimal path from the target station to the end point, divide the square of the distance difference by 2 times the standard deviation, and use a negative exponential function for exponential decay; The exponential decay result is used as the exponent of the normalized result to obtain the probability of the target station being selected between the coordinate point of the current passenger vehicle and the coordinate point of the high-priority order.

6. A customized passenger route planning system, characterized in that: include: A processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the customized passenger route planning method according to any one of claims 1 to 5 is implemented.

Images