Method and device for predicting inter-city online car-hailing multi-region short-time distribution position

Abstract

The invention provides a method and device for predicting multi-region short-time distribution positions of intercity online-hailed cars, and relates to the technical field of vehicle dispatch, and the method comprises the steps: dividing a plurality of sub-regions according to the obtained road network data of two cities in an operation line of the intercity online-hailed cars, and constructing intercity entrances and exits of the two cities; according to the obtained passenger order placing data of the two cities, respectively constructing initial vehicle shift reporting sets of each sub-region of the two cities; then, a multi-task adaptive Kalman filtering model is independently established for each vehicle; constructing a real vehicle distribution set, a prediction available vehicle distribution set, a pre-urban vehicle set and an on-way vehicle set of the two cities, and updating the pre-urban vehicle set of the two cities after constructing the on-way vehicle set; and after the vehicle leaves the city, the vehicle reports the class again, and the corresponding Kalman filtering model is reset. According to the method, the Kalman filtering model is independently established for each vehicle, so that the characteristics of each vehicle can be better captured, and the accuracy of the multi-region short-time distribution position of the inter-city online car-hailing is improved.

Description

Technical Field

[0001] The present invention relates to the field of vehicle dispatching technology, and in particular to a method and device for predicting the short-term distribution positions of inter-city online car-hailing vehicles in multiple regions. Background Art

[0002] With the popularity of online ride-hailing services, inter-city traffic demand has become highly dynamic and complex, especially during peak hours and holidays, when travel demand in different regions fluctuates greatly. Traditional static scheduling methods cannot meet this changing demand, resulting in increased waiting time for passengers and increased idle driving rates for drivers, affecting service quality and operational efficiency. Inter-city traffic demand is affected by a variety of factors, such as weekday and weekend travel habits, large-scale events, and emergencies, and has a high degree of spatiotemporal variability. Traditional scheduling methods are based on fixed empirical rules and lack the ability to dynamically respond to real-time data, making it difficult to cope with complex demand changes. Therefore, inter-city online ride-hailing services urgently need a location prediction method that can predict and dynamically adjust vehicle distribution in real time to adapt to the rapidly changing demand environment, thereby improving passenger service satisfaction, optimizing driver work efficiency, and improving the overall operational efficiency of online ride-hailing platforms.

[0003] The change of transportation capacity is affected by the dispatching strategy. The prediction based on the data-driven deep learning algorithm is not explanatory, and cannot predict the location and situation of specific vehicles. It cannot update the distribution of transportation capacity in real time as the dispatching strategy changes, and the prediction accuracy is low. The existing patent document CN117593043B "Method, device and equipment for estimating the future short-term time distribution of intercity online vehicles" is not applicable to situations where vehicle speed data is not available, and does not involve the situation of each vehicle in multiple areas in the city.

[0004] In view of this, the applicant filed this application after studying the existing technology. Summary of the invention

[0005] The present invention aims to provide a method and device for predicting the short-term distribution position of inter-city online car-hailing vehicles in multiple regions, so as to solve the problems that the existing model cannot explain the prediction of transportation capacity, cannot predict the specific location and situation of each vehicle, and has low accuracy.

[0006] In order to solve the above technical problems, the present invention is implemented through the following technical solutions:

[0007] A method for predicting the short-term distribution position of inter-city online car-hailing vehicles in multiple regions, comprising:

[0008] S1, based on the real-time road network data of cities A and B in the intercity online ride-hailing routes, divide several sub-areas and construct the intercity entrances and exits of the designated city A. A and the intercity entrance and exit of city B B ;

[0009] S2: Based on the obtained passenger order data of each sub-area of ​​city A and city B, the initial vehicle registration set K of each sub-area of ​​city A and city B is constructed respectively. A and K B ;

[0010] S3, based on the initial vehicle shift set K A and K B , a multi-task adaptive Kalman filter model is established for each vehicle separately; the multi-task adaptive Kalman filter model takes the vehicle's highest detour coefficient, number of passengers and previous state data at the current moment as input, and outputs the next predicted state data and predicted time to leave the city of the vehicle

[0011] S4, at each scheduling decision time t in the established platform scheduling decision time, according to the vehicle shift set K A , construct the real vehicle distribution set of all sub-areas of city A at time t And the predicted available vehicle distribution set of all sub-areas of city A from time t to t+n n is the estimated step length; similarly, according to the vehicle shift set K B , construct the real vehicle distribution set of all sub-areas of city B and the predicted available vehicle distribution set

[0012] S5, taking each scheduling decision time t as a trigger event, traverse the real vehicle distribution set of city A Calculate the detour coefficient of the vehicle at time t, and input the detour coefficient and number of passengers of the city vehicle at time t that is neither empty nor fully loaded into the adaptive Kalman filter model of the vehicle, and output the predicted time for the vehicle to leave the city. According to the predicted departure time Construct a pre-exit vehicle set for multiple areas of city A; similarly, construct a pre-exit vehicle set for multiple areas of city B

[0013] S6, vehicle k from city A passes through the intercity entrance e at time t. A To trigger the event, traverse the real vehicle distribution set of city A Calculate the arrival time set of all passengers for vehicles that have left the city Construct a collection of vehicles on the way to city A According to the vehicles on the way The arrival time and drop-off location of the last passenger in the game are updated to update the predicted available vehicle distribution set of city B. Similarly, vehicle k from city B passes through the intercity entrance e at time t. BTo trigger the event, construct the vehicle set en route to city B And update the predicted available vehicle distribution set of city A

[0014] S7, based on the actual time of leaving the city of the vehicle obtained in real time Update the data of the adaptive Kalman filter model of the corresponding vehicle until the vehicle reports for duty again, and reset the adaptive Kalman filter model of the vehicle.

[0015] The present invention also provides a device for predicting the short-term distribution position of inter-city online car-hailing vehicles in multiple regions, comprising:

[0016] The multi-region dispatch environment construction unit is used to divide several sub-regions according to the real-time road network data of cities A and B in the inter-city online car-hailing routes, and construct the inter-city entrances and exits of the designated city A. A and the intercity entrance and exit of city B B ;

[0017] The multi-region vehicle shift set construction unit is used to construct the initial vehicle shift set K of each sub-region of city A and city B according to the obtained passenger order data of each sub-region of city A and city B. A and K B ;

[0018] The multi-task model building unit is used to build a model based on the initial vehicle report set K A and K B , a multi-task adaptive Kalman filter model is established for each vehicle separately; the multi-task adaptive Kalman filter model takes the vehicle's highest detour coefficient, number of passengers and previous state data at the current moment as input, and outputs the next predicted state data and predicted time to leave the city of the vehicle

[0019] The multi-region vehicle distribution set construction unit is used to obtain the real vehicle distribution set and the predicted available vehicle distribution set. At each scheduling decision time t in the established platform scheduling decision time, according to the vehicle shift set K A , construct the real vehicle distribution set of all sub-areas of city A at time t And the predicted available vehicle distribution set of all sub-areas of city A from time t to t+n n is the estimated step length; similarly, according to the vehicle shift set K B , construct the real vehicle distribution set of all sub-areas of city B and the predicted available vehicle distribution set

[0020] The multi-region pre-outbound vehicle set construction unit is used to obtain the multi-region pre-outbound vehicle set, taking each scheduling decision time t as a trigger event, and traversing the real vehicle distribution set of city A. Calculate the detour coefficient of the vehicle at time t, and input the detour coefficient and number of passengers of the city vehicle at time t that is neither empty nor fully loaded into the adaptive Kalman filter model of the vehicle, and output the predicted time for the vehicle to leave the city. According to the predicted departure time Construct a pre-exit vehicle set for multiple areas of city A; similarly, construct a pre-exit vehicle set for multiple areas of city B

[0021] The en route vehicle set construction unit is used to obtain the en route vehicle set, taking vehicle k from city A passing through the intercity entrance e at time t A To trigger the event, traverse the real vehicle distribution set of city A Calculate the arrival time set of all passengers for vehicles that have left the city Construct a collection of vehicles on the way to city A According to the vehicle collection The arrival time and drop-off location of the last passenger in the game are updated to update the predicted available vehicle distribution set of city B. Similarly, vehicle k from city B passes through the intercity entrance e at time t. B To trigger the event, construct the vehicle set en route to city B And update the predicted available vehicle distribution set of city A

[0022] The model updating unit is used to update the actual time of vehicles leaving the city according to the real-time information. Update the data of the adaptive Kalman filter model of the corresponding vehicle until the vehicle reports for duty again, and reset the adaptive Kalman filter model of the vehicle.

[0023] The present invention also provides a device for predicting the short-term distribution positions of multiple regions of inter-city online ride-hailing vehicles, comprising a processor and a memory, wherein the memory stores a computer program, and the computer program can be executed by the processor to implement a method for predicting the short-term distribution positions of multiple regions of inter-city online ride-hailing vehicles as described above.

[0024] The present invention also provides a computer-readable storage medium, on which computer-readable instructions are stored. When the computer-readable instructions are executed by a processor of a device where the computer-readable storage medium is located, a method for predicting the short-term distribution position of multi-region inter-city online car-hailing vehicles as described above is implemented.

[0025] In summary, compared with the prior art, the present invention has the following beneficial effects:

[0026] (1) The present invention establishes a Kalman filter model for each vehicle separately, and the data between vehicles do not interfere with each other and are updated independently, which can better capture the characteristics of each vehicle.

[0027] (2) The multi-task adaptive Kalman filter model constructed by the present invention not only uses the previous state data as the model input, but also inputs the current number of passengers and the highest detour coefficient of the vehicle into the Kalman filter model to obtain the predicted time to leave the city. The closer the highest detour coefficient and the number of passengers are to the constraint value, the more likely the vehicle is to leave the city directly, making the vehicle distribution prediction explainable and allowing the vehicle to "have a trace".

[0028] (3) The present invention can predict the location and status of specific vehicles in multiple areas without obtaining the vehicle's driving speed, and update vehicle information in real time, which can effectively improve the prediction accuracy, adapt to the rapidly changing demand environment and scheduling strategy, optimize the driver's work efficiency, and improve the overall operational efficiency of the online car-hailing platform.

[0029] (4) The present invention divides a city into several sub-areas, and can obtain the vehicle conditions in a specific area of ​​a city in real time, reduce data transmission, and improve operation efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0031] Figure 1 A schematic diagram of a method for predicting the short-term distribution positions of inter-city online ride-hailing vehicles in multiple regions provided in Example 1 of the present invention.

[0032] Figure 2 A schematic diagram of a model of a multi-task adaptive Kalman filter provided in Embodiment 1 of the present invention.

[0033] Figure 3 A schematic diagram of vehicle trajectory prediction on the way provided in Example 1 of the present invention.

[0034] Figure 4 A real-time prediction process for the short-term distribution locations of inter-city online ride-hailing vehicles in multiple regions provided in Example 1 of the present invention.

[0035] Figure 5 This is a graph showing the prediction results of the number of available vehicles in a certain area of ​​city B provided in Example 1 of the present invention.

[0036] Figure 6This is a diagram showing the prediction results of the number of available seats in a certain area of ​​city B provided in the first embodiment of the present invention.

[0037] Figure 7 A schematic diagram of a device for predicting the short-term distribution positions of inter-city online ride-hailing vehicles in multiple regions provided in Example 2 of the present invention.

[0038] The present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. DETAILED DESCRIPTION

[0039] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention. Therefore, the following detailed description of the embodiments of the present invention provided in the drawings is not intended to limit the scope of the invention claimed for protection, but merely represents the selected embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0040] Embodiment 1

[0041] Embodiment 1 of the present invention provides a method for predicting the short-term distribution positions of inter-city online ride-hailing vehicles in multiple regions, which can be implemented by a prediction device for predicting the short-term distribution positions of inter-city online ride-hailing vehicles in multiple regions (hereinafter referred to as the prediction device), and in particular, executed by one or more processors in the prediction device.

[0042] In this embodiment, the prediction device may be an electronic device equipped with a processor, the processor having a computer program of the prediction method of the short-term distribution location of multiple areas of the inter-city online car-hailing vehicle and the computer program can be executed, such as a computer, workstation, server, etc., which is not specifically limited in the present invention.

[0043] In this embodiment, the Kalman filter model is an efficient recursive filter that can estimate the state of a dynamic system from a series of noisy observation data. The Kalman filter model is based on the state space representation of a linear dynamic system, which assumes that the system state is linear and that both the process noise and the observation noise are Gaussian distributed. The model iteratively estimates the state of the system through two main steps: prediction and update.

[0044] Prediction phase: predict the current state based on the system's previous state and control input; also predict the current state covariance.

[0045] Update phase: Use new measurement data to update the prediction; calculate a gain (Kalman gain) that determines which part of the prediction or measurement is more reliable; update the estimated state and covariance.

[0046] like Figure 1 As shown, a method for predicting the short-term distribution position of inter-city online ride-hailing vehicles in multiple regions includes steps S1 to S7.

[0047] S1, based on the real-time road network data of cities A and B in the intercity online ride-hailing routes, divide several sub-areas and construct the intercity entrances and exits of the designated city A. A and the intercity entrance and exit of city B B .

[0048] The road network data includes the internal road network structure of the two cities, the inter-city entrances and exits, A and e B The expressway connects the two cities. The left and right ends of the expressway are connected to the only intercity entrances and exits of the two cities. Intercity online ride-hailing vehicles leave the city through the intercity entrances and exits, pass through the expressway, and then arrive at the other city through the intercity entrance and exit of the other city.

[0049] Divide city A into |U| regions A U , U={1,…,u,…,|U|}, u is the area number of city A; divide city B into |W| areas B W , W={1,…,w,…,|W|}, w is the area number of city B; |U| and |W| are the number of areas divided by city A and city B.

[0050] S2: Based on the obtained passenger order data of each sub-area of ​​city A and city B, the initial vehicle registration set K of each sub-area of ​​city A and city B is constructed respectively. A and K B .

[0051] According to the number of drivers on the operating routes in each sub-area u of city A, an initial vehicle shift set K consisting of the same number of vehicles as the number of drivers is constructed. A , The vehicle shift report set of sub-area u in city A;

[0052] The vehicle k is reported for duty, which means that the vehicle k can be used for the next dispatch. After the vehicle k delivers the last passenger, it will be added to the vehicle report set in the city.

[0053] The order data of each passenger p can be represented by the tuple Composition, of which o p is the boarding location of passenger p, d p It is the drop-off point, u p is the area code of the boarding location, w p is the area code of the drop-off location, u p and w p Not in the same city, It's the earliest time to get on the bus. It is the latest boarding time;

[0054] Suppose the set of time numbers for platform scheduling decisions is T, T = {0, 1, …, t, …, |T|}, and any time interval of [t-1, t] is equal.

[0055] The vehicle k has at time t: rated capacity The position at time t is l t , vehicle status s t , the vehicle area number u t , the assigned passenger set P t ={p1,…,p i ,,…p q}, that is, having a tuple And vehicle k needs to be in the travel time window of passenger p Pick up passengers; q is the number of passengers.

[0056] Similarly, construct the initial vehicle registration set K of city B B .

[0057] S3, based on the initial vehicle shift set K A and K N , a multi-task adaptive Kalman filter model is established for each vehicle separately; the multi-task adaptive Kalman filter model takes the vehicle's highest detour coefficient, number of passengers and previous state data at the current moment as input, and outputs the next predicted state data and predicted time to leave the city of the vehicle

[0058] When calculating the detour coefficient of vehicle k at the current moment, the detour coefficient of vehicle k after each insertion of a passenger is calculated based on the initial vehicle shift set, and the highest detour coefficient r at the current moment is updated. t ;

[0059] The detour coefficient = the total travel mileage of vehicle k after the new passenger is inserted divided by the total travel mileage of the vehicle before the new passenger is inserted;

[0060] If the calculated detour coefficient is greater than the highest detour coefficient r at the current momentt When the maximum detour coefficient r is updated t ; otherwise remain unchanged.

[0061] For example, when vehicle k receives the first passenger order, the total distance traveled is 10 kilometers, and the maximum detour coefficient is initialized to 0. After receiving the second passenger order at time t, it needs to detour to pick up passenger 2. At this time, the total distance traveled is set to 15 kilometers, and the detour coefficient = 15 / 10 = 1.5, which is greater than the current maximum detour coefficient of 0, and the maximum detour coefficient is updated to 1.5. When the third passenger is picked up, the total distance traveled becomes 21 kilometers, and the detour coefficient at this time is = 21 / 15 = 1.4. Since 1.4<1.5, the maximum detour coefficient remains unchanged.

[0062] In this embodiment, if Figure 2 As shown in the figure, because the scheduling of each vehicle is independent, an adaptive Kalman filter model is established for each vehicle, which can produce different outputs according to different model inputs of the vehicle (such as the current maximum detour coefficient and the number of passengers of the vehicle). First, the Kalman filter model is initialized, and then at each scheduling decision time t, the Kalman filter model is called to predict the next state of the vehicle to reflect the situation of each vehicle, and is updated independently to avoid mutual interference between the models. The specific steps are as follows:

[0063] Step 1: Initialize the state vector and error covariance matrix of the Kalman filter model:

[0064] In this embodiment, the state vector Error covariance matrix State transition matrix Observation Matrix Observation noise covariance matrix Process noise covariance matrix

[0065] Among them, d s is the initial distance of vehicle k from the exit, t o is the time offset, d v is the distance variance, t v is the time variance;

[0066] Step 2: Based on the number of passengers q of vehicle k at time t and the highest detour coefficient r at the current time t , adjust the process noise covariance matrix Q and the state transfer matrix F, the formula is as follows:

[0067]

[0068] in, and is the weight coefficient of the number of passengers in the process noise covariance matrix and the state transfer matrix, and is the weight coefficient of the detour coefficient in the process noise covariance matrix and the state transfer matrix; is the detour coefficient threshold;

[0069] Step 3: Based on the previous state vector, process noise covariance matrix and estimated error covariance matrix, predict the next state vector and error covariance. The formula is as follows:

[0070] x t|t-1 =Fx t-1|t-1 ;

[0071] P ′ t|t-1 =FP ′ t-1|t-1 F * +Q;

[0072] Among them, x t-1|t-1 is the estimated state vector at time t-1, x t|t-1 is the predicted state vector at time t, P ′ t|t-1 is the forecast error covariance matrix at time t, P ′ t-1|t-1 is the estimated error covariance matrix at time t-1, * is the transposed matrix;

[0073] Then, predict the time to leave the city The calculation method of is: the time offset of the predicted state vector at the current time + time t.

[0074] S4, at each scheduling decision time t in the established platform scheduling decision time, according to the vehicle shift set K A , construct the real vehicle distribution set of all sub-areas of city A at time t And the predicted available vehicle distribution set of all sub-areas of city A from time t to t+n n is the estimated step length; similarly, according to the vehicle shift set K B , construct the real vehicle distribution set of all sub-areas of city B and the predicted available vehicle distribution set

[0075] The real vehicle distribution set It refers to all sub-areas in city A that have reported to work at time t and have not passed through the intercity entrance e A And the remaining capacity on the vehicle The vehicle set is expressed as:

[0076]

[0077] in, represents the actual vehicle distribution set of sub-area u of city A at time t;

[0078] The predicted available vehicle distribution set It refers to all sub-areas of city A that report to work during the period [t, t+n] and do not pass through the intercity entrance e A And the remaining capacity on the vehicle The vehicle set is expressed as:

[0079]

[0080] At the beginning of each scheduling decision time t, let Right now To ensure the real-time and accuracy of the forecasting process;

[0081] in, represents the predicted available vehicle distribution set of sub-area u in city A during the period [t, t+n];

[0082] Similarly, construct the real vehicle distribution set of city B and the predicted available vehicle distribution set

[0083] S5, taking each scheduling decision time t as a trigger event, traverse the real vehicle distribution set of city A Calculate the detour coefficient of the vehicle at time t, and input the detour coefficient and number of passengers of the city vehicle at time t that is neither empty nor fully loaded into the adaptive Kalman filter model of the vehicle, and output the predicted time for the vehicle to leave the city. According to the predicted departure time Construct a set of vehicles that are expected to leave the city in multiple areas of city A; similarly, vehicle k from city B passes through the intercity entrance e at time t B To trigger the event, construct a collection of vehicles that are expected to leave the city in multiple areas of city B

[0084] Taking each scheduling decision time t as the trigger event, traverse the real vehicle distribution set of all sub-areas of city A

[0085] If the vehicle in sub-area u of city A The vehicle k in the equation is neither empty nor fully loaded, then the highest detour coefficient r of vehicle k at time t is t , the number of passengers q and the previous state data are input into the adaptive Kalman filter model of vehicle k to obtain the predicted departure time of vehicle k

[0086] If you predict the time to leave the city Then the vehicle k is taken from the real vehicle distribution set of area u in city A and the predicted available vehicle distribution set Remove it from the list and add vehicle k to the set of vehicles to leave the city , construct a collection of vehicles that are expected to leave the city:

[0087]

[0088] in, represents the set of vehicles expected to leave the city in the sub-area u of city A during the period [t, t+n], and the vehicle k is calculated based on the predicted time of leaving the city. Add to middle;

[0089] Similarly, construct the set of vehicles that are expected to leave the city in city B.

[0090] S6, vehicle k from city A passes through the intercity entrance e at time t. A To trigger the event, traverse the real vehicle distribution set of city A Calculate the arrival time set of all passengers for vehicles that have left the city Construct a collection of vehicles on the way to city A According to the vehicles on the way The arrival time and drop-off location of the last passenger in the game are updated to update the predicted available vehicle distribution set of city B. Similarly, vehicle k from city B passes through the intercity entrance e at time t. B To trigger the event, construct the vehicle set en route to city B And update the predicted available vehicle distribution set of city A

[0091] Traverse the real vehicle distribution set of all sub-areas of city A If the vehicle in sub-area u of city A Vehicle k has passed the intercity entrance e A , then vehicle k is moved from and Removed;

[0092] like Figure 3 As shown, calculate the expected arrival time set of all passengers on vehicle k

[0093]

[0094] in, represents the i-th passenger p i Arrival time, is the arrival time of the last passenger, The calculation method is:

[0095] Through the intercity entrance A With passengers i Drop-off location p The distance between the two vehicles is divided by the default speed of the road. For example, highways are usually set to 120 kilometers per hour and urban roads are set to 50 kilometers per hour.

[0096] Construct a collection of vehicles on the way to city A like Then the vehicle k is assigned the arrival time Add to the collection of vehicles on the way to city A like Then the vehicle k is assigned the arrival time of the last passenger and the area number of the drop-off location p After updating the information of vehicle k, add the real vehicle distribution set of city B

[0097] Similarly, construct the set of vehicles on the way to city B And update the predicted available vehicle distribution set of city A

[0098] S7, based on the actual time of leaving the city of the vehicle obtained in real time Update the data of the adaptive Kalman filter model of the corresponding vehicle until the vehicle reports for duty again, and reset the adaptive Kalman filter model of the vehicle.

[0099] In this embodiment, if Figure 4 As shown, initialize t = 0, and obtain the real vehicle distribution at time t in real time; let the time number i of the platform scheduling decision be 0, obtain the detour coefficient and number of passengers of vehicle k at the current time, perform real-time multi-task Kalman filter prediction, and output the real-time prediction of vehicle k's departure time until the real vehicle distribution set of all sub-areas of city A is traversed. Then t=t+1, continue to iterate and calculate the real-time prediction of the vehicle distribution in all areas of city A during the period [t, t+n].

[0100] If the vehicle set of sub-area u in city A Vehicle k has passed the intercity entrance e A , according to the actual departure time of vehicle k The location and intercity entrance / exit when the prediction is triggered A Distance Update the adaptive Kalman filter model of vehicle k. The update process is as follows:

[0101] Step 1: Calculate the observed residual y t , the formula is:

[0102]

[0103] Step 2: Calculate the observed residual covariance S t , the formula is:

[0104] S t =HP ′ t|t-1 H * +R;

[0105] Step 3: Calculate the Kalman gain Z t , the formula is:

[0106]

[0107] Step 4: Update the state vector x t|t , the formula is:

[0108] x t|t =x t|t-1 +Z t y t ;

[0109] Step 5: Update the error covariance matrix P ′ t|t , the formula is:

[0110] P ′ t|t =P ′ t|t-1 -K t HK t|t-1 ;

[0111] Similarly, update the real vehicle distribution set of all sub-areas of city B The adaptive Kalman filter model corresponding to the vehicles that have left the city;

[0112] After vehicle k re-registers, the adaptive Kalman filter model of vehicle k is reset and the learning rate lr is introduced to allow the model to remember the historical training process. The reset process is as follows:

[0113] 1) Count the average out-of-city time t of vehicle k’s historical trips on that day avg and distance d avg ;

[0114] 2) Adjust the state vector x of vehicle k;

[0115]

[0116] 3) Reset the error covariance matrix P of vehicle k ′ , that is, initialize P ′ .

[0117] In another specific embodiment, in order to verify the effectiveness of the model and model solution proposed by the present invention, a practical example will be used to illustrate the application of the present invention.

[0118] The order data of a certain line of a company's intercity online car-hailing service on June 2, 2023 was selected as the research object. After one day of scheduling, the evaluation indicators used for the estimated results of each out-of-city time are the root mean square error (RMSE), mean absolute error (MAE), and symmetric mean absolute percentage error (SMAPE). The lower the values ​​of RMSE, MAE, and SMAPE, the better the estimation accuracy of the model. The formula is as follows:

[0119]

[0120] Among them, y i is the actual number of available vehicles (or actual number of available seats) of the capacity data, is the predicted number of available vehicles (or the predicted number of available seats), ti is the number of divided time slices; the vehicle distribution at adjacent moments is considered unchanged, and the number of available seats in a certain area can be counted by the number of available vehicles in that area; the experimental parameter settings are shown in Table 1.

[0121] Table 1. Experimental parameter settings

[0122]

[0123] The comparison algorithms used are HA (Historical Average), XGBOOST (eXtremeGradient Boosting, an integrated learning algorithm based on the gradient boosting framework), LSTM (Long Short-Term Memory, a long short-term memory neural network model), and T-GCN (Temporal Graph Convolutional Network, a temporal graph convolutional network model for traffic flow prediction).

[0124] Among them, the data set for the comparison algorithm consists of the intercity round-trip online car-hailing data of this route from April 1, 2023 to June 4, 2023. The division ratio of the training set, validation set, and test set is 7:2:1. The prediction results are shown in Tables 2 and 3.

[0125] Table 2. Prediction results of the number of available vehicles in City A<—>City B

[0126]

[0127] Table 3. Prediction results of available seats in city A<—>city B

[0128]

[0129] It can be seen from Table 2 and Table 3 that the prediction results of the number of available vehicles and the number of available seats are better than those of other comparison algorithms and have higher prediction accuracy.

[0130] from Figure 5 and Figure 6 It can be seen that in the prediction results of the sub-regions, the predicted number of available vehicles and the actual number of vehicles, as well as the predicted number of available seats and the actual number of seats are highly similar, which confirms the feasibility of this embodiment in predicting the short-term distribution positions of inter-city online car-hailing vehicles in multiple regions, and verifies that the solution quality of this embodiment is high and can meet the engineering needs of operators.

[0131] In summary, compared with the prior art, the present invention has the following beneficial effects:

[0132] The present invention aims to solve the problem of low prediction accuracy of transportation capacity by existing data-driven models. It builds a separate model for each vehicle to better capture the characteristics of each vehicle. According to the current number of passengers and detour coefficient of the vehicle, it adaptively inputs the Kalman filter model to make the vehicle distribution prediction explainable and make the vehicle "traceable". The present invention is applicable to situations where the vehicle speed is not available and does not involve multiple areas in the city. It can predict the location and status of specific vehicles in multiple areas and update the status in real time. It can effectively improve the prediction accuracy, adapt to the rapidly changing demand environment and scheduling strategies, optimize the work efficiency of drivers, improve the overall operational efficiency of the online car-hailing platform, and provide a high-value scheduling reference for the online car-hailing platform. It can be used in the vehicle relocation scheduling strategy for supply and demand balance in the future.

[0133] Embodiment 2

[0134] like Figure 7 As shown, the second embodiment of the present invention further provides a device for predicting the short-term distribution position of inter-city online car-hailing vehicles in multiple regions, including:

[0135] The multi-region dispatch environment construction unit is used to divide several sub-regions according to the real-time road network data of cities A and B in the inter-city online car-hailing routes, and construct the inter-city entrances and exits of the designated city A. A and the intercity entrance and exit of city B B ;

[0136] The multi-region vehicle shift set construction unit is used to construct the initial vehicle shift set K of each sub-region of city A and city B according to the obtained passenger order data of each sub-region of city A and city B. A and K B ;

[0137] The multi-task model building unit is used to build a model based on the initial vehicle report set K A and K B , a multi-task adaptive Kalman filter model is established for each vehicle separately; the multi-task adaptive Kalman filter model takes the vehicle's highest detour coefficient, number of passengers and previous state data at the current moment as input, and outputs the next predicted state data and predicted time to leave the city of the vehicle

[0138] The multi-region vehicle distribution set construction unit is used to obtain the real vehicle distribution set and the predicted available vehicle distribution set. At each scheduling decision time t in the established platform scheduling decision time, according to the vehicle shift set K A , construct the real vehicle distribution set of all sub-areas of city A at time t And the predicted available vehicle distribution set of all sub-areas of city A from time t to t+n n is the estimated step length; similarly, according to the vehicle shift set K B , construct the real vehicle distribution set of all sub-areas of city B and the predicted available vehicle distribution set

[0139] The multi-region pre-outbound vehicle set construction unit is used to obtain the multi-region pre-outbound vehicle set, taking each scheduling decision time t as a trigger event, and traversing the real vehicle distribution set of city A. Calculate the detour coefficient of the vehicle at time t, and input the detour coefficient and number of passengers of the city vehicle at time t that is neither empty nor fully loaded into the adaptive Kalman filter model of the vehicle, and output the predicted time for the vehicle to leave the city. According to the predicted departure time Construct a pre-exit vehicle set for multiple areas of city A; similarly, construct a pre-exit vehicle set for multiple areas of city B

[0140] The en route vehicle set construction unit is used to obtain the en route vehicle set, taking vehicle k from city A passing through the intercity entrance e at time t A To trigger the event, traverse the real vehicle distribution set of city A Calculate the arrival time set of all passengers for vehicles that have left the city Construct a collection of vehicles on the way to city A According to the vehicle collection The arrival time and drop-off location of the last passenger in the game are updated to update the predicted available vehicle distribution set of city B. Similarly, vehicle k from city B passes through the intercity entrance e at time t. B To trigger the event, construct a set of vehicles en route to city B And update the predicted available vehicle distribution set of city A

[0141] The model updating unit is used to update the actual time of vehicles leaving the city according to the real-time information. Update the data of the adaptive Kalman filter model of the corresponding vehicle until the vehicle reports for duty again, and reset the adaptive Kalman filter model of the vehicle.

[0142] Embodiment 3

[0143] The third embodiment of the present invention also provides a prediction device for the short-term distribution positions of multiple regions of inter-city online ride-hailing vehicles, which includes a memory and a processor, wherein the memory stores a computer program, and the computer program can be executed by the processor to implement the prediction method for the short-term distribution positions of multiple regions of inter-city online ride-hailing vehicles as described above.

[0144] Embodiment 4

[0145] The fourth embodiment of the present invention also provides a computer-readable storage medium, on which computer-readable instructions are stored. When the computer-readable instructions are executed by the processor of the device where the computer-readable storage medium is located, the method for predicting the short-term distribution position of multiple areas of inter-city online car-hailing vehicles as mentioned above is implemented.

[0146] In several embodiments provided in the embodiments of the present invention, it should be understood that the disclosed apparatus and method can also be implemented in other ways. The apparatus and method embodiments described above are merely schematic. For example, the flowcharts in the accompanying drawings show the possible architecture, functions and operations of the apparatus, method and computer program product according to multiple embodiments of the present invention. In this regard, each box in the flowchart or block diagram can represent a module, a program segment or a part of a code, and the module, program segment or a part of the code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flow chart, and the combination of boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or action, or can be implemented with a combination of dedicated hardware and computer instructions.

[0147] In addition, the functional modules in the various embodiments of the present invention may be integrated together to form an independent part, or each module may exist independently, or two or more modules may be integrated to form an independent part.

[0148] If the functions are implemented in the form of software function modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present invention is essentially or partly contributed to the prior art or part of the technical solution can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions to enable a computer device (which can be a personal computer, an electronic device, or a network device, etc.) to perform all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk. It should be noted that, in this article, the terms "include", "comprise" or any other variant thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements that are not explicitly listed, or also includes elements inherent to such a process, method, article or device. Without more constraints, an element defined by the phrase "comprising a..." does not exclude the existence of other identical elements in the process, method, article or apparatus comprising the element.

[0149] The terms used in the embodiments of the present invention are only for the purpose of describing specific embodiments, and are not intended to limit the present invention. The singular forms "a", "said" and "the" used in the embodiments of the present invention and the appended claims are also intended to include plural forms, unless the context clearly indicates other meanings.

[0150] It should be understood that the term "and / or" used in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship.

[0151] The word "if" as used herein may be interpreted as "at the time of" or "when" or "in response to determining" or "in response to detecting", depending on the context. Similarly, the phrases "if it is determined" or "if (stated condition or event) is detected" may be interpreted as "when it is determined" or "in response to determining" or "when detecting (stated condition or event)" or "in response to detecting (stated condition or event)", depending on the context.

[0152] The "first\second" mentioned in the embodiments is only to distinguish similar objects, and does not represent a specific order for the objects. It is understandable that the "first\second" can be interchanged with the specific order or sequence where permitted. It should be understood that the objects distinguished by "first\second" can be interchanged where appropriate, so that the embodiments described herein can be implemented in an order other than those illustrated or described herein.

[0153] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for predicting the short-term distribution position of inter-city online car-hailing vehicles in multiple regions, characterized in that: include: S1, based on the real-time road network data of cities A and B in the intercity online ride-hailing routes, divide several sub-areas and construct the intercity entrances and exits of the designated city A. A and the intercity entrance and exit of city B B ; S2: Based on the obtained passenger order data of each sub-area of ​​city A and city B, the initial vehicle registration set K of each sub-area of ​​city A and city B is constructed respectively. A and K B ; S3, based on the initial vehicle shift set K A and K B , a multi-task adaptive Kalman filter model is established for each vehicle separately; the multi-task adaptive Kalman filter model takes the vehicle's highest detour coefficient, number of passengers and previous state data at the current moment as input, and outputs the next predicted state data and predicted time to leave the city of the vehicle S4, at each scheduling decision time t in the established platform scheduling decision time, according to the vehicle shift set K A , construct the real vehicle distribution set of all sub-areas of city A at time t And the predicted available vehicle distribution set of all sub-areas of city A from time t to t+n n is the estimated step length; similarly, according to the vehicle shift set K B , construct the real vehicle distribution set of all sub-areas of city B and the predicted available vehicle distribution set S5, taking each scheduling decision time t as a trigger event, traverse the real vehicle distribution set of city A Calculate the detour coefficient of the vehicle at time t, and input the detour coefficient and number of passengers of the city vehicle at time t that is neither empty nor fully loaded into the adaptive Kalman filter model of the vehicle, and output the predicted time for the vehicle to leave the city. According to the predicted departure time Construct a pre-exit vehicle set for multiple areas of city A; similarly, construct a pre-exit vehicle set for multiple areas of city B S6, vehicle k from city A passes through the intercity entrance e at time t. A To trigger the event, traverse the real vehicle distribution set of city A Calculate the arrival time set of all passengers for vehicles that have left the city Construct a collection of vehicles on the way to city A According to the vehicle collection on the way The arrival time and drop-off location of the last passenger in the game are updated to update the predicted available vehicle distribution set of city B. Similarly, vehicle k from city B passes through the intercity entrance e at time t. B To trigger the event, construct the vehicle set en route to city B And update the predicted available vehicle distribution set of city A S7, based on the actual time of leaving the city of the vehicle obtained in real time Update the data of the adaptive Kalman filter model of the corresponding vehicle until the vehicle reports for duty again, and reset the adaptive Kalman filter model of the vehicle.

2. According to claim 1, a method for predicting the short-term distribution position of inter-city online car-hailing vehicles in multiple regions is characterized by , in step S1: The road network data includes the internal road network structure of the two cities, the inter-city entrances and exits, A and e B The expressway connects the two cities. The left and right ends of the expressway are connected to the only intercity entrances and exits of the two cities. Intercity online ride-hailing vehicles leave the city through the intercity entrances and exits, pass through the expressway, and then arrive at the other city through the intercity entrance and exit of the other city. Divide city A into |U| regions A U , U={1,…,u,…,|U|}, u is the area code of city A; Divide city B into |W| regions B W , W={1,…,w,…,|W|}, w is the area code of city B.

3. A method for predicting the short-term distribution position of inter-city online car-hailing vehicles in multiple regions according to claim 2, characterized in that , in step S2: According to the number of drivers on the operating routes in each sub-area u of city A, an initial vehicle shift set K consisting of the same number of vehicles as the number of drivers is constructed. A , The vehicle shift report set of sub-area u in city A; The vehicle k is reported for duty, which means that the vehicle k can be used for the next dispatch. After the vehicle k delivers the last passenger, it will be added to the vehicle report set in the city; The order data for each passenger p is represented by a tuple Composition, among which, o p is the boarding location of passenger p, d p It is the drop-off point, u p is the area code of the boarding location, w p is the area code of the drop-off location, u p and w p Not in the same city, It's the earliest time to get on the bus. It is the latest boarding time; The vehicle k has at time t: rated capacity The position at time t is l t , vehicle status s t , the vehicle area number u t , the assigned passenger set P t ={p1,…,p i ,,…p q }, that is, having a tuple And vehicle k needs to be in the travel time window of passenger p Pick up passengers; q is the number of passengers; Similarly, construct the initial vehicle registration set K of city B B .

4. A method for predicting the short-term distribution position of inter-city online car-hailing vehicles in multiple regions according to claim 3, characterized in that When calculating the detour coefficient of vehicle k at the current moment, the detour coefficient of vehicle k after each insertion of a passenger is calculated according to the initial vehicle shift set, and the highest detour coefficient r at the current moment is updated. t ; The detour coefficient = the total travel mileage of vehicle k after the new passenger is inserted divided by the total travel mileage of the vehicle before the new passenger is inserted; If the calculated detour coefficient is greater than the highest detour coefficient r at the current moment t When , the maximum detour coefficient r is updated t ; otherwise remain unchanged.

5. A method for predicting the short-term distribution position of inter-city online car-hailing vehicles in multiple regions according to claim 4, characterized in that ,In step S3, when establishing a multi-task adaptive Kalman filter model for each vehicle, the Kalman filter model is initialized first, and then at each scheduling decision time t, the Kalman filter model is called to predict the next state of the vehicle to reflect the situation of each vehicle and update it independently to avoid mutual interference between the models. The specific steps are: Step 1: Initialize the state vector and error covariance matrix of the Kalman filter model: Define the state vector The error covariance matrix is ​​P ′ , the state transfer matrix is ​​F, the observation matrix is ​​H, and the observation noise covariance matrix is The process noise covariance matrix is ​​Q; Among them, d s is the initial distance of vehicle k from the exit, t o is the time offset, d v is the distance variance, t v is the time variance; Step 2: Based on the number of passengers q of vehicle k at time t and the highest detour coefficient r at the current time t , adjust the process noise covariance matrix Q and the state transfer matrix F, the formula is as follows: in, and is the weight coefficient of the number of passengers in the process noise covariance matrix and the state transfer matrix, and is the weight coefficient of the detour coefficient in the process noise covariance matrix and the state transfer matrix; is the detour coefficient threshold; Step 3: Based on the previous state vector, process noise covariance matrix and estimated error covariance matrix, predict the next state vector and error covariance. The formula is as follows: x t|t-1 =Fx t-1|t-1 ; P ′ t|t-1 =FP ′ t-1|t-1 F * +Q; Among them, x t-1|t-1 is the estimated state vector at time t-1, x t|t-1 is the predicted state vector at time t, P ′ t|t-1 is the forecast error covariance matrix at time t, P ′ t-1|t-1 is the estimated error covariance matrix at time t-1, * is the transposed matrix; Then, predict the time to leave the city is the time offset of the predicted state vector at the current time + time t.

6. A method for predicting the short-term distribution position of inter-city online car-hailing vehicles in multiple regions according to claim 5, characterized in that In step S4, the real vehicle distribution set It refers to all sub-areas in city A that have reported to work at time t and have not passed through the intercity entrance e A And the remaining capacity on the vehicle The vehicle set is expressed as: in, represents the actual vehicle distribution set of sub-area u of city A at time t; The predicted available vehicle distribution set It refers to all sub-areas of city A that report to work during the period [t, t+n] and do not pass through the intercity entrance e A And the remaining capacity on the vehicle The vehicle set is expressed as: At the beginning of each scheduling decision time t, let Right now To ensure the real-time and accuracy of the forecasting process; in, represents the predicted available vehicle distribution set of sub-area u in city A during the period [t, t+n]; Similarly, construct the real vehicle distribution set of city B and the predicted available vehicle distribution set 7. A method for predicting the short-term distribution position of inter-city online ride-hailing vehicles in multiple regions according to claim 6, characterized in that ,Step S5 specifically includes: Taking each scheduling decision time t as the trigger event, traverse the real vehicle distribution set of all sub-areas of city A If the vehicle in sub-area u of city A The vehicle k in the equation is neither empty nor fully loaded, then the highest detour coefficient r of vehicle k at time t is t , the number of passengers q and the previous state data are input into the adaptive Kalman filter model of vehicle k to obtain the predicted departure time of vehicle k If you predict the time to leave the city Then the vehicle k is taken from the real vehicle distribution set of area u in city A and the predicted available vehicle distribution set Remove it from the list and add vehicle k to the set of vehicles to leave the city , construct a collection of vehicles that are expected to leave the city: in, represents the set of vehicles expected to leave the city in the sub-area u of city A during the period [t, t+n], and the vehicle k is calculated based on the predicted time of leaving the city. Add to middle; Similarly, construct the set of vehicles that are expected to leave the city in city B.

8. A method for predicting the short-term distribution position of inter-city online ride-hailing vehicles in multiple regions according to claim 7, characterized in that , step S6 specifically includes: Traverse the real vehicle distribution set of all sub-areas of city A If the vehicle in sub-area u of city A Vehicle k has passed the intercity entrance e A , then vehicle k is moved from and Removed; Calculate the arrival time set of all passengers on vehicle k in, represents the i-th passenger p i Arrival time, is the arrival time of the last passenger, The calculation method is: Through the intercity entrance A With passengers i Drop-off location p The distance between the two roads is divided by the default speed of the road; Construct a collection of vehicles on the way to city A like Then the vehicle k is assigned the arrival time Add to the collection of vehicles on the way to city A like Then the vehicle k is assigned the arrival time of the last passenger and the area number of the drop-off location p After updating the information of vehicle k, add the real vehicle distribution set of city B Similarly, construct the set of vehicles on the way to city B And update the predicted available vehicle distribution set of city A 9. A method for predicting the short-term distribution position of inter-city online car-hailing vehicles in multiple regions according to claim 8, characterized in that , step S7 specifically includes: Traverse the real vehicle distribution set of all sub-areas of city A If the vehicle set of sub-area u in city A Vehicle k has passed the intercity entrance e A , according to the actual departure time of vehicle k The location and intercity entrance / exit when the prediction is triggered A Distance Update the adaptive Kalman filter model of vehicle k. The update process is as follows: Step 1: Calculate the observed residual y t , the formula is: Step 2: Calculate the observed residual covariance S t , the formula is: S t =HP ′ t|t-1 H * +R; Step 3: Calculate the Kalman gain Z t , the formula is: Step 4: Update the state vector x t|t , the formula is: x t|t =x t|t-1 +Z t y t ; Step 5: Update the error covariance matrix P ′ t|t , the formula is: P ′ t|t =P ′ t|t-1 -K t HK t|t-1 ; Similarly, update the real vehicle distribution set of all sub-areas of city B The adaptive Kalman filter model corresponding to the vehicles that have left the city; After vehicle k re-registers, the adaptive Kalman filter model of vehicle k is reset and the learning rate lr is introduced to allow the model to remember the historical training process. The reset process is as follows: 1) Count the average out-of-city time t of vehicle k’s historical trips on that day avg and distance d avg ; 2) Adjust the state vector x of vehicle k; Among them, lr is the learning rate; 3) Reset the error covariance matrix P of vehicle k ′ , that is, initialize P ′ .

10. A device for predicting the short-term distribution position of inter-city online car-hailing vehicles in multiple regions, characterized in that: include: The multi-region dispatch environment construction unit is used to divide several sub-regions according to the real-time road network data of cities A and B in the inter-city online car-hailing routes, and construct the inter-city entrances and exits of the designated city A. A and the intercity entrance and exit of city B B ; The multi-region vehicle shift set construction unit is used to construct the initial vehicle shift set K of each sub-region of city A and city B according to the obtained passenger order data of each sub-region of city A and city B. A and K B ; The multi-task model building unit is used to build a model based on the initial vehicle report set K A and K B , a multi-task adaptive Kalman filter model is established for each vehicle separately; the multi-task adaptive Kalman filter model takes the vehicle's highest detour coefficient, number of passengers and previous state data at the current moment as input, and outputs the next predicted state data and predicted time to leave the city of the vehicle The multi-region vehicle distribution set construction unit is used to obtain the real vehicle distribution set and the predicted available vehicle distribution set. At each scheduling decision time t in the established platform scheduling decision time, according to the vehicle shift set K A , construct the real vehicle distribution set of all sub-areas of city A at time t And the predicted available vehicle distribution set of all sub-areas of city A from time t to t+n n is the estimated step length; similarly, according to the vehicle shift set K B , construct the real vehicle distribution set of all sub-areas of city B and the predicted available vehicle distribution set The multi-region pre-outbound vehicle set construction unit is used to obtain the multi-region pre-outbound vehicle set, taking each scheduling decision time t as a trigger event, and traversing the real vehicle distribution set of city A. Calculate the detour coefficient of the vehicle at time t, and input the detour coefficient and number of passengers of the city vehicle at time t that is neither empty nor fully loaded into the adaptive Kalman filter model of the vehicle, and output the predicted time for the vehicle to leave the city. According to the predicted departure time Construct a pre-exit vehicle set for multiple areas of city A; similarly, construct a pre-exit vehicle set for multiple areas of city B The en route vehicle set construction unit is used to obtain the en route vehicle set, taking vehicle k from city A passing through the intercity entrance e at time t A To trigger the event, traverse the real vehicle distribution set of city A Calculate the arrival time set of all passengers for vehicles that have left the city Construct a collection of vehicles on the way to city A According to the vehicle collection on the way The arrival time and drop-off location of the last passenger in the game are updated to update the predicted available vehicle distribution set of city B. Similarly, vehicle k from city B passes through the intercity entrance e at time t. B To trigger the event, construct a set of vehicles en route to city B And update the predicted available vehicle distribution set of city A The model updating unit is used to update the actual time of vehicles leaving the city according to the real-time information. Update the data of the adaptive Kalman filter model of the corresponding vehicle until the vehicle reports for duty again, and reset the adaptive Kalman filter model of the vehicle.

Images