Intelligent pricing management platform based on designated driving track recognition

Abstract

The invention provides an intelligent pricing management platform based on designated driving track recognition, and relates to the technical field of taxi pricing management, the intelligent pricing management platform comprises a pricing system, a path planning system and a pricing system, and the path planning system and the pricing system are both integrated on the pricing management platform; the pricing system comprises a pricing rule, a pre-pricing module and an actual pricing module; the path planning system comprises a navigation path planning module, a dynamic path planning module and a specificity determination module; the technical key points are as follows: a path is planned for a vehicle in real time through an abnormal point identification module and a dynamic path planning module, and if the vehicle runs according to an initial path planned by a navigation path, when a road emergency occurs and a diversion measure needs to be taken, the actual driving mileage and the corresponding price of the vehicle are calculated; and the difference between the price and the predicted price is calculated, and the predicted price is superposed to obtain the final cost as the settlement reference basis, so that the customer and the driver can be prevented from bearing more cost unilaterally, and the loss unilaterally is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of vehicle pricing management, and in particular to an intelligent pricing management platform based on designated driver trajectory recognition. Background Art

[0002] Freight rate refers to the price per unit of cargo carried, the freight charged for transporting a certain unit of cargo. Freight rate is the product of freight rate and volume. The transportation industry is an independent sector of material production, but it does not produce tangible products. Its production activities simply cause passengers or cargo to move in space, and its production and consumption processes are identical. This distinguishes transportation prices from those of general industrial and agricultural products. Its characteristics are: ① The price structure includes factors such as distance and weight, and the unit of calculation is ton-kilometer rate or person-kilometer rate (ton-nautical-mile rate or person-nautical-mile rate); ② It only takes the form of a sales price; and ③ Freight rates vary depending on the type of cargo being transported, the selected transportation method, and the transportation distance.

[0003] Taxis are a familiar mode of transportation, playing a vital role in urban mobility. However, in the current taxi industry, driver violations are common. For example, some drivers fail to follow regulations by using the meter when there are passengers in the taxi. As people's needs evolve, private car chauffeurs have become a popular mode of transportation.

[0004] The patent document with announcement number CN105957156B discloses a method and system for managing taxi driving behavior. The method and system detect passenger boarding information through a gravity sensing device. The driver starts the meter and sends the detection information and the start of metering information to the management platform through the intelligent vehicle-mounted device. The management platform determines whether the taxi driver's driving behavior is in violation of regulations based on whether the start of metering information is received from the meter within a preset time or by comparing the time difference between the start of metering information received from the meter and the detection information of the gravity sensing device within a preset time, thereby achieving effective management of the taxi driver's driving behavior and avoiding illegal driving behaviors such as drivers not using the meter.

[0005] However, in the process of implementing the above technical solution, it was found that the above technical solution had the following technical problems:

[0006] The method and system for judging taxi driving behavior realize effective management of the driver's driving behavior by determining whether a person is on the vehicle and whether the driving time is in accordance with the expected time. However, whether it is a taxi or a designated driver's vehicle, it may encounter road emergencies (temporary road closures, traffic congestion, etc.) during driving and need to re-plan the route. At this time, if the original preset time is used to judge the driving behavior, the consumer or the driver will unilaterally bear the corresponding costs, making it difficult to guarantee the rights and interests of both the driver and the consumer at the same time. Summary of the Invention

[0007] In order to overcome the problem that the existing taxi driving behavior methods and systems may encounter road emergencies (temporary road closures, traffic congestion, etc.) during vehicle driving, and the route needs to be replanned, if the original preset time is used to judge the driving behavior at this time, the consumer or the driver will unilaterally bear the corresponding costs, which makes it difficult to simultaneously ensure the rights and interests of both drivers and consumers. The embodiment of the present application provides an intelligent pricing management platform based on designated driver trajectory recognition. By setting a mileage deviation threshold for the initial route planned by the navigation path planning module and the actual driving distance (vehicle driving trajectory), a price deviation threshold is obtained. During the vehicle driving process, if the price difference between the actual price and the expected price exceeds the price deviation threshold after the vehicle arrives at the destination, the mileage price is used for billing, thereby avoiding causing large losses to the driver;

[0008] At the same time, the abnormal point identification module uses the dynamic window method to plan the vehicle's path in real time by using the real-time obstacle avoidance and path tracking method in a dynamic environment. If the vehicle is traveling according to the initial route planned by the navigation path and an emergency road situation occurs and diversion measures must be taken, the actual mileage of the vehicle and the corresponding price are calculated, and the difference between the actual mileage and the estimated price is calculated. The final fee is added to the estimated price to obtain the final fee as a reference for settlement, so as to avoid customers and drivers unilaterally bearing more fees, reduce unilateral losses, and reflect the principles of fairness and justice.

[0009] The technical solution adopted by the embodiment of the present application to solve the technical problem is:

[0010] An intelligent pricing management platform based on designated driver trajectory recognition includes a pricing system, a route planning system, and a pricing system. Both the route planning system and the pricing system are integrated into the pricing management platform.

[0011] The pricing system includes pricing rules, an estimated pricing module, and an actual pricing module, and the path planning system includes a navigation path planning module, a dynamic path planning module, and a specificity identification module;

[0012] The estimated price module determines the cost of operating the vehicle when traveling along the driving route, i.e., the estimated price, in combination with the pricing rules and based on the optimal driving route selected by the navigation route planning module;

[0013] The actual pricing module determines the mileage from the starting point to the end point based on the pricing rules and the actual driving route obtained by the dynamic path planning module, thereby obtaining the cost of vehicle operation, namely the mileage price;

[0014] Among them, the actual error between the driving distance and the navigation distance during driving is set according to the optimal driving path selected by the navigation path planning module, and the price deviation threshold in the estimated price is set.

[0015] In one possible implementation, the actual error between the driving distance and the navigation distance is used to formulate a threshold parameter using positioning data indicators, and the characteristics of the tail data of the random sequence of the monitoring data are analyzed. The threshold parameter is determined by the following expression:

[0016] Conditional distribution function F of excess sequence T (y) is:

[0017] F T (y)=P(xT≤y|x>T)

[0018] The expression of F(x) with respect to F(y) is:

[0019] F(x)=F T (y)[1-T(F)]+F(T)

[0020] Among them, F(x) is the monitoring index x under the proposed significance level α α The basis of x α =F -1 (x, α)

[0021] There is a correlation between the distribution functions of the original measurement sequence, the over-threshold measurement sequence and the excess sequence, and a relationship can be constructed to solve it.

[0022] The PBdH theorem in extreme value theory shows that for a sufficiently large threshold T, the conditional distribution function F of the excess amount y T (y) converges to the generalized Pareto distribution, that is:

[0023]

[0024] Among them, ξ T , σ T are two evaluation parameters in the POT model. According to the PBdH theorem, by setting the threshold T, the original measurement sequence {x i} to construct the excess sequence {y j} and obtain its distribution function F T (y). The distribution function F corresponding to the excess sequence T (y) has a corresponding relationship with the distribution function F(x) corresponding to the original measurement value sequence, so the distribution function F(x) of the corresponding original measurement value sequence can be solved for any set threshold T that meets the conditions.

[0025] In one possible implementation, during the calculation of the price deviation threshold T, the Logistic-Tent hybrid mapping optimization algorithm is used to update the data, and its expression is:

[0026]

[0027] Among them, r∈(0,4];

[0028] The finder position update equation is:

[0029]

[0030] The calculation formula of parameter ω is:

[0031]

[0032] Where, ω0 is a given positive real number; t is the current iteration number; t0 is the given iteration number;

[0033] The sparrow position update formula based on Levy flight is:

[0034]

[0035] Where γ is the step control parameter; Levy(λ) satisfies Levy~u=t -λ 1<λ<3;

[0036] The sparrow position update based on reverse learning is:

[0037]

[0038] The dynamic selection strategy method is: when rand∈(0, 0.5], select the Levy flight strategy to update the sparrow's position. Otherwise, select the reverse learning strategy to update the sparrow's position.

[0039] In one possible implementation, the navigation path planning module abstracts the environment into a graph during the initial path planning process, where nodes represent locations, edges represent walkable connections, and weights represent costs such as distance and time. The graph is represented as follows:

[0040] Adjacency matrix: A = [a ij ];

[0041] Among them, a ij is the edge weight from node i to j (∞ when there is no connection).

[0042] Adjacency list: A linked list structure that stores the adjacent nodes and edge weights of each node.

[0043] Then the shortest path is calculated by Dijkstra algorithm, as follows:

[0044] Initialize the distance array d[s] = 0, and the rest d[i] = ∞; and continuously perform iterative updates during the calculation process: d[v] = min(d[v], d[u] + weight(u, v));

[0045] Among them, u is the currently visited node, v is the adjacent node of u, and weight(u, v) is the edge weight; at the same time, the priority queue (minimum heap) selects the node with the smallest current distance to expand.

[0046] In one possible implementation, the dynamic path planning module utilizes the outlier identification module to implement real-time obstacle avoidance and path tracking (with the starting and ending points unchanged) in a dynamic environment using a dynamic window method during the dynamic path planning process, specifically as follows:

[0047] Sample feasible speed pairs (v, ω) within the allowed linear speed v and ω range, and based on the cost function:

[0048] J=ω dist c dist +ω angle c angle +ω vel c vel

[0049] Calculate the cost of each speed pair and finally select the speed with the minimum cost as the updated path for the vehicle.

[0050] In one possible implementation, the dynamic path planning module defines the vehicle's running behavior using a differential wheel model based on a dynamic window method during the dynamic path planning process to determine obstacles present during the vehicle's travel:

[0051]

[0052] Where (x(t), y(t)) is the position of the car at time t, θ(t) is the heading angle, v is the linear velocity, ω is the angular velocity, and Δt is the sampling time interval;

[0053] Divide the prediction time T into N discrete time steps and calculate the position (x k ,y k )(k=1,2,…,N), and then check whether these positions overlap with obstacles one by one to predict the vehicle collision obstacle. The specific formula is:

[0054]

[0055] Among them, (x obs ,y obs ) is the coordinate of the obstacle center, r car is the vehicle radius, r obsis the obstacle radius (used to expand the obstacle to avoid edge collision), d safe The safe distance maintained by the car; if there is any discrete point (x k ,y k )satisfy It is then determined that the velocity combination (v, ω) will lead to a collision.

[0056] In one possible implementation, when a vehicle is traveling along the initial route planned by the navigation path and no road emergency occurs that necessitates a route change, if the driver changes lanes and a fare is charged based on actual mileage, the estimated price corresponding to the initial route will be applied.

[0057] At the same time, when the vehicle travels according to the initial route planned by the navigation path and the price obtained by charging based on the actual mileage deviates from the expected price, it is determined whether to charge based on the expected price or the mileage price based on whether the price difference between the actual price and the expected price is included in the price deviation threshold.

[0058] In one possible implementation, when a vehicle is traveling along the initial route planned by the navigation path and an emergency road condition occurs and a diversion is necessary, the actual mileage of the vehicle and the corresponding price are calculated, and the difference between the actual mileage and the estimated price is calculated. In accordance with the principle of fairness, the sum of the price difference and the estimated price is used as the final fee, which is used as a reference for settlement.

[0059] The beneficial effects of this application are:

[0060] First, in this solution, a mileage deviation threshold is set for the initial route planned by the navigation path planning module and the actual driving distance (vehicle driving trajectory), thereby obtaining a price deviation threshold. During the vehicle's driving process, if the price difference between the actual price and the expected price after the vehicle arrives at the destination exceeds the price deviation threshold, the mileage price will be used for charging, thus avoiding large losses for the driver;

[0061] Second, in this solution, the outlier identification module uses the dynamic window method to plan the vehicle's path in real time using the real-time obstacle avoidance and path tracking method in a dynamic environment. If the vehicle is traveling along the initial route planned by the navigation path and an emergency occurs on the road and diversion measures must be taken, the actual mileage of the vehicle and the corresponding price are calculated, and the difference between the actual mileage and the estimated price is calculated. The final fee is obtained by adding the estimated price as a reference for settlement, avoiding the customer and the driver from unilaterally bearing more fees, reducing unilateral losses, and reflecting the principle of fairness and justice. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] Figure 1 This is a structural block diagram of the intelligent pricing management platform based on designated driver trajectory identification of the present invention;

[0063] Figure 2 The figure is a schematic diagram of the pricing process of the intelligent pricing management platform based on designated driver trajectory identification of the present invention. DETAILED DESCRIPTION

[0064] The technical solution in the embodiments of the present application is to solve the problems of the above-mentioned background technology, and the overall idea is as follows:

[0065] Example 1:

[0066] This embodiment introduces the specific structure of the intelligent pricing management platform based on designated driver trajectory recognition. Figure 1 and Figure 2 As shown, it includes a pricing system, a path planning system and a pricing system. The path planning system and the pricing system are integrated into the pricing management platform. The pricing system includes pricing rules, an estimated pricing module and an actual pricing module. The path planning system includes a navigation path planning module, a dynamic path planning module and a specificity recognition module.

[0067] Among them, the estimated pricing module combines the pricing rules and the optimal driving route selected by the navigation path planning module to determine the cost of vehicle operation when traveling along the driving route, that is, the estimated price; the actual pricing module combines the pricing rules and the actual driving route obtained by the dynamic path planning module to determine the mileage from the starting point to the end point, thereby obtaining the cost of vehicle operation, that is, the mileage price

[0068] Secondly, during the driving process according to the optimal driving route selected by the navigation route planning module, the price deviation threshold in the estimated price is set according to the actual error between the driving distance and the navigation distance (the threshold parameter is formulated using the positioning data indicator and the characteristics of the tail data of the random sequence of the monitoring data are analyzed). The threshold parameter determination method is determined by the following expression:

[0069] Conditional distribution function F of excess sequence T (y) is:

[0070] F T (y)=P(xT≤y|x>T)

[0071] The expression of F(x) with respect to F(y) is:

[0072] F(x)=F T (y)[1-F(T)]+F(T)

[0073] Among them, F(x) is the monitoring index x under the proposed significance level α α The basis of x α =F -1 (x, α)

[0074] There is a correlation between the distribution functions of the original measurement sequence, the over-threshold measurement sequence and the excess sequence, and a relationship can be constructed to solve it.

[0075] The PBdH theorem in extreme value theory shows that for a sufficiently large threshold T, the conditional distribution function F of the excess amount y T (y) converges to the generalized Pareto distribution, that is:

[0076]

[0077] Among them, ξ T , σ T are two evaluation parameters in the POT model. According to the PBdH theorem, by setting the threshold T, the original measurement sequence {x i} to construct the excess sequence {y j} and obtain its distribution function F T (y). The distribution function F corresponding to the excess sequence T (y) has a corresponding relationship with the distribution function F(x) corresponding to the original measurement value sequence, so the distribution function F(x) of the corresponding original measurement value sequence can be solved for any set threshold T that meets the conditions;

[0078] Therefore, when the distribution function F(x) of the original measurement sequence is obtained, the monitoring indicator x under the proposed significance level α can be determined. α In summary, we get the threshold T and monitoring index x α The inherent correlation between them can be used to formulate reasonable monitoring indicators;

[0079] In one possible implementation, during the calculation of the price deviation threshold T, the Logistic-Tent hybrid mapping optimization algorithm is used to update the data, and its expression is:

[0080]

[0081] Among them, r∈(0,4];

[0082] The finder position update equation is:

[0083]

[0084] The calculation formula of parameter ω is:

[0085]

[0086] Where, ω0 is a given positive real number; t is the current iteration number; t0 is the given iteration number;

[0087] The sparrow position update formula based on Levy flight is:

[0088]

[0089] Where γ is the step control parameter; Levy(λ) satisfies Levy~u=t -λ 1<λ<3;

[0090] The sparrow position update based on reverse learning is:

[0091]

[0092] The dynamic selection strategy method is: when rand∈(0, 0.5], select the Levy flight strategy to update the sparrow's position. Otherwise, select the reverse learning strategy to update the sparrow's position.

[0093] Based on the two methods mentioned above, a dynamic selection strategy is used to update the sparrow's position, alternating between Levy flight and reverse learning with a certain probability. In the Levy flight strategy, a step factor is used to expand the search range and escape the local optimal dilemma. At the same time, the reverse learning strategy increases the diversity of solutions and improves the algorithm's search optimization performance.

[0094] When a vehicle is traveling along the initial route planned by the navigation route and no road emergency occurs that requires a route change, if the driver changes lanes and is charged based on actual mileage, the estimated price corresponding to the initial route will be applied.

[0095] At the same time, when the vehicle is traveling along the initial route planned by the navigation path, and the price obtained by charging based on the actual mileage deviates from the estimated price, whether the price difference between the actual price and the estimated price is within the price deviation threshold is determined to determine whether to charge based on the estimated price or the mileage price;

[0096] If the price difference between the actual price and the expected price is within the price deviation threshold, the estimated price can be used for charging. If the price difference between the actual price and the expected price exceeds the price deviation threshold, it can be inferred that there is a deviation between the initial route planned by the navigation path planning module and the actual driving distance, and the mileage price will be used for charging.

[0097] The above design obtains a price deviation threshold by setting a mileage deviation threshold for the initial route planned by the navigation path planning module and the actual driving distance (vehicle driving trajectory). During the vehicle's driving process, if the price difference between the actual price and the expected price exceeds the price deviation threshold after the vehicle arrives at the destination, it can be inferred that there is a deviation between the initial route and the actual driving distance. The mileage price will be used for billing, which is as consistent as possible with the cost corresponding to the actual driving situation to avoid causing large losses to the driver.

[0098] Example 2:

[0099] Based on Example 1, this example introduces the application of the navigation path planning module. During the initial path planning process, the navigation path planning module abstracts the environment into a graph, where nodes represent locations, edges represent walkable connections, and weights represent costs such as distance and time. The graph is represented as follows:

[0100] Adjacency matrix: A = [a ij ];

[0101] Among them, a ij is the edge weight from node i to j (∞ when there is no connection).

[0102] Adjacency list: A linked list structure that stores the adjacent nodes and edge weights of each node.

[0103] Then the shortest path is calculated by Dijkstra algorithm, as follows:

[0104] Initialize the distance array d[s] = 0, and the rest d[i] = ∞; and continuously perform iterative updates during the calculation process: d[v] = min(d[v], d[u] + weightt(u, v));

[0105] Where u is the currently visited node, v is the adjacent node of u, and weight(u, v) is the edge weight. At the same time, the priority queue (minimum heap) selects the node with the smallest current distance to expand. This is the workflow of the navigation path planning module to plan the initial driving route.

[0106] The dynamic path planning module uses the outlier identification module to adopt the dynamic window method to achieve real-time obstacle avoidance and path tracking in a dynamic environment (with the starting and ending points unchanged) during the dynamic path planning process. The specific implementation is as follows:

[0107] Sample feasible speed pairs (v, ω) within the allowed linear speed v and ω range, and based on the cost function:

[0108] J=ω dist c dist +ω angle c angle +ω vel c vel

[0109] Calculate the cost of each speed pair and finally select the speed with the minimum cost as the updated path for the vehicle.

[0110] In one possible implementation, the dynamic path planning module defines the vehicle's operating behavior using a differential wheel model based on a dynamic window method during the dynamic path planning process to identify obstacles that may occur during the vehicle's travel (new routes may be formed during obstacle avoidance, such as temporary road closures or severe traffic jams caused by traffic accidents):

[0111]

[0112] Where (x(t), y(t)) is the position of the car at time t, θ(t) is the heading angle, v is the linear velocity, ω is the angular velocity, and Δt is the sampling time interval;

[0113] Divide the prediction time T into N discrete time steps and calculate the position (x k ,y k )(k=1,2,…,N), and then check whether these positions overlap with obstacles one by one to predict the vehicle collision obstacle. The specific formula is:

[0114]

[0115] Among them, (x obs ,y obs ) is the coordinate of the obstacle center, r car is the vehicle radius, r obs is the obstacle radius (used to expand the obstacle to avoid edge collision), d safe The safe distance maintained by the car; if there is any discrete point (x k ,y k )satisfy It is determined that the speed combination (v, ω) will lead to a collision, and a prediction plan is provided for the vehicle to avoid obstacles and change the route in advance.

[0116] If a vehicle is traveling along the initial route planned by the navigation system and an unexpected road condition occurs where the vehicle can pass by waiting (manifested by the vehicle ahead moving slowly), the vehicle will continue along the initial route planned by the navigation system and the fee will be settled based on the estimated price obtained based on the distance of the initial route.

[0117] If a vehicle is traveling along the initial route planned by the navigation path and an emergency occurs on the road and a rerouting measure must be taken, the actual mileage of the vehicle and the corresponding price will be calculated, and the difference between the actual mileage and the estimated price will be calculated. In accordance with the principle of fairness, the sum of the price difference and the estimated price will be used as the final fee, which will be used as a reference for settlement. This provides important data support for discussions between the driver and the customer, and prevents the customer and the driver from unilaterally bearing higher costs.

[0118] The above design plans the vehicle's path in real time through the outlier identification module in conjunction with the dynamic path planning module using the dynamic window method for real-time obstacle avoidance and path tracking in a dynamic environment. If the vehicle is traveling according to the initial route planned by the navigation path and an emergency road situation occurs and diversion measures must be taken, the actual mileage of the vehicle and the corresponding price are calculated, and the difference between the actual mileage and the estimated price is calculated. The final fee is added to the estimated price to obtain the final fee as a reference for settlement, avoiding the customer and the driver from unilaterally bearing more fees, reducing unilateral losses, and reflecting the principle of fairness and justice.

[0119] It is worth noting that the above solution is aimed at calculating vehicle driving costs based on actual mileage, and does not include a method in which the driver and the customer negotiate to determine the cost in the form of a flat price.

[0120] Finally, it should be noted that the above embodiments are merely examples for the purpose of illustrating the present invention and are not intended to limit the embodiments. Those skilled in the art will readily appreciate that other variations or modifications based on the above description are possible. It is not necessary and impossible to provide an exhaustive list of all embodiments. However, obvious variations or modifications arising therefrom remain within the scope of protection of the present invention.

Claims

1. An intelligent pricing management platform based on designated driver trajectory recognition, characterized by: include: Pricing system; The route planning system and the pricing system are integrated on the pricing management platform; The pricing system includes pricing rules, an estimated pricing module, and an actual pricing module, and the path planning system includes a navigation path planning module, a dynamic path planning module, and a specificity identification module; The estimated price module determines the cost of operating the vehicle when traveling along the driving route, i.e., the estimated price, in combination with the pricing rules and based on the optimal driving route selected by the navigation route planning module; The actual pricing module determines the mileage from the starting point to the end point based on the pricing rules and the actual driving route obtained by the dynamic path planning module, thereby obtaining the cost of vehicle operation, namely the mileage price; Among them, the actual error between the driving distance and the navigation distance during driving is set according to the optimal driving path selected by the navigation path planning module, and the price deviation threshold in the estimated price is set.

2. The intelligent pricing management platform based on designated driver trajectory recognition according to claim 1 is characterized by: The actual error between the driving distance and the navigation distance is determined by using the positioning data indicator to formulate the threshold parameter. The characteristics of the tail data of the random sequence of the monitoring data are analyzed, and the threshold parameter is determined by the following expression: Conditional distribution function F of excess sequence T (y) is: F T (y)=P(x-T≤y|x>T) The expression of F(x) with respect to F(y) is: F(x)=F T (y)[1-F(T)]+F(T) Among them, F(x) is the monitoring index x under the proposed significance level α α The basis of x α =F -1 (x, α) There is a correlation between the distribution functions of the original measurement sequence, the over-threshold measurement sequence and the excess sequence, and a relationship can be constructed to solve it. The PBdH theorem in extreme value theory shows that for a sufficiently large threshold T, the conditional distribution function F of the excess amount y T (y) converges to the generalized Pareto distribution, that is: Among them, ξ T , σ T are two evaluation parameters in the POT model. According to the PBdH theorem, by setting the threshold T, the original measurement sequence {x i } to construct the excess sequence {y j } and obtain its distribution function F T (y). The distribution function F corresponding to the excess sequence T (y) has a corresponding relationship with the distribution function F(x) corresponding to the original measurement value sequence, so the distribution function F(x) of the corresponding original measurement value sequence can be solved for any set threshold T that meets the conditions.

3. The intelligent pricing management platform based on designated driver trajectory recognition according to claim 2 is characterized by: In the process of calculating the price deviation threshold T, the Logistic-Tent hybrid mapping optimization algorithm is used to update the data, and its expression is: Among them, r∈(0,4]; The finder position update equation is: The calculation formula of parameter ω is: Where, ω0 is a given positive real number; t is the current iteration number; t0 is the given iteration number; The sparrow position update formula based on Levy flight is: Where γ is the step control parameter; Levy(λ) satisfies Levy~u=t -λ 1<λ<3; The sparrow position update based on reverse learning is: The dynamic selection strategy method is: when rand∈(0, 0.5], select the Levy flight strategy to update the sparrow's position. Otherwise, select the reverse learning strategy to update the sparrow's position.

4. The intelligent pricing management platform based on designated driver trajectory recognition according to claim 1 is characterized by: During the initial path planning process, the navigation path planning module abstracts the environment into a graph, where nodes represent locations, edges represent walkable connections, and weights represent costs such as distance and time. The graph is represented as follows: Adjacency matrix: A = [a ij ]; Among them, a ij is the edge weight from node i to j (∞ when there is no connection). Adjacency list: A linked list structure that stores the adjacent nodes and edge weights of each node. Then the shortest path is calculated by Dijkstra algorithm, as follows: Initialize the distance array d[s] = 0, and the rest d[i] = ∞; and continuously perform iterative updates during the calculation process: d[v] = min(d[v], d[u] + weightt(u, v)); Among them, u is the currently visited node, v is the adjacent node of u, and weight(u, v) is the edge weight; at the same time, the priority queue (minimum heap) selects the node with the smallest current distance to expand.

5. The intelligent pricing management platform based on designated driver trajectory recognition according to claim 4 is characterized by: The dynamic path planning module uses the outlier identification module to adopt the dynamic window method to achieve real-time obstacle avoidance and path tracking in a dynamic environment (with the starting and ending points unchanged) during the dynamic path planning process. Specifically, Sample feasible speed pairs (v, ω) within the allowed linear speed v and ω range, and based on the cost function: J=ω dist ·c dist +oh angle ·c angle +oh vel ·c vel Calculate the cost of each speed pair and finally select the speed with the minimum cost as the updated path for the vehicle.

6. The intelligent pricing management platform based on designated driver trajectory recognition according to claim 5 is characterized by: The dynamic path planning module defines the vehicle's running behavior using a differential wheel model based on a dynamic window method during the dynamic path planning process to determine obstacles present during the vehicle's travel. Where (x(t), y(t)) is the position of the car at time t, θ(t) is the heading angle, v is the linear velocity, ω is the angular velocity, and Δt is the sampling time interval; Divide the prediction time T into N discrete time steps and calculate the position (x k ,y k )(k=1,2,…,N), and then check whether these positions overlap with obstacles one by one to predict the vehicle collision obstacle. The specific formula is: Among them, (x obs ,y obs ) is the coordinate of the obstacle center, r car is the vehicle radius, r obs is the obstacle radius (used to expand the obstacle to avoid edge collision), d safe The safe distance maintained by the car; if there is any discrete point (x k ,y k )satisfy It is then determined that the velocity combination (v, ω) will lead to a collision.

7. The intelligent pricing management platform based on designated driver trajectory recognition according to claim 6 is characterized by: When the vehicle is traveling along the initial route planned by the navigation path and no road emergency occurs that requires a reroute, if the driver changes lanes and is charged based on actual mileage, the estimated price corresponding to the initial route will be applied; At the same time, when the vehicle travels according to the initial route planned by the navigation path and the price obtained by charging based on the actual mileage deviates from the expected price, it is determined whether to charge based on the expected price or the mileage price based on whether the price difference between the actual price and the expected price is included in the price deviation threshold.

8. The intelligent pricing management platform based on designated driver trajectory recognition according to claim 7 is characterized by: When a vehicle is traveling along the initial route planned by the navigation path and an emergency road condition occurs and a rerouting measure must be taken, the actual mileage of the vehicle and the corresponding price are calculated, and the difference between the actual mileage and the estimated price is calculated. In accordance with the principle of fairness, the sum of the price difference and the estimated price is used as the final fee and as a reference for settlement.

Images