Vehicle speed estimation method based on transition probability matrix and rule constraint optimization

Abstract

The invention discloses a vehicle speed estimation method based on a transition probability matrix and rule constraint optimization, and the method comprises the steps: obtaining GPS data in a vehicle driving process, and carrying out the calculation to obtain corresponding speed data and acceleration data; performing data preprocessing on the GPS data, the speed data and the acceleration data to obtain a vehicle driving data set, and constructing a speed transition probability matrix under a plurality of traffic scenes according to the vehicle driving data set; dividing the current travel of the target vehicle into a plurality of sub-travels according to the traffic scene, and determining a speed transition probability matrix corresponding to each sub-travel; according to the initial speed of the target vehicle, the speed of the target vehicle at the next moment is predicted based on the speed transition probability matrix and a preset constraint rule, and a speed estimation sequence of each sub-travel is obtained; and performing speed smoothing, correction and sequence splicing on the speed estimation sequence to obtain vehicle speed time sequence data. The method improves the accuracy of vehicle speed estimation, and can be applied to the technical field of intelligent traffic.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent transportation, and in particular to a vehicle speed estimation method based on transition probability matrix and rule constraint optimization. BACKGROUND

[0002] At present, the field of transportation focuses on the path of pollution reduction and carbon reduction synergy, and constructs and improves the refined carbon emission statistical accounting system. As the core tool for quantifying automobile pollutant emissions and carbon emissions, the emission inventory is changing from regional road network macro-micro statistics to single vehicle micro-analysis, and its refinement level depends on the representation ability of individual vehicle driving data. Under the condition of strict specification of collection, storage and use of personal travel data, existing traffic detection technology is difficult to obtain the complete travel trajectory of individual vehicles in the whole domain and full quantity. The path planning algorithm only outputs the static travel path scheme, and cannot obtain the vehicle motion information. Therefore, the vehicle speed estimation method is needed to predict the speed of the vehicle on the travel path to obtain the complete travel trajectory of the vehicle.

[0003] The existing vehicle speed estimation method can be divided into mathematical interpolation method, mathematical modeling method and method based on time series prediction model. The mathematical interpolation method is not suitable for the scene with high data sparsity, and cannot effectively handle the speed estimation task in complex traffic environment. The mathematical modeling method has high calculation cost and is easy to fall into local optimum, and is often used in closed highway scenes. The time series prediction model has high requirements for data quantity and computing resources, has the problems of uninterpretability and error accumulation, and is difficult to solve the long distance speed estimation problem.

[0004] Therefore, it is urgent to propose a vehicle speed estimation method which can accurately restore the instantaneous motion state of individual vehicles. SUMMARY

[0005] The present application aims to at least partly solve one of the problems of the prior art.

[0006] To this end, one object of the present application is to provide a vehicle speed estimation method based on transition probability matrix and rule constraint optimization, which improves the accuracy of vehicle speed estimation and is suitable for vehicle speed estimation of different distances in various traffic scenes.

[0007] Another object of the present application is to provide a vehicle speed estimation device based on transition probability matrix and rule constraint optimization.

[0008] In order to achieve the above technical purpose, the technical scheme adopted by the embodiments of the present application comprises: On the one hand, the embodiments of the present application provide a vehicle speed estimation method based on transition probability matrix and rule constraint optimization, comprising the following steps: The system acquires GPS data during vehicle operation and calculates the corresponding speed and acceleration data based on the GPS data. The GPS data, the speed data, and the acceleration data are preprocessed to obtain a vehicle driving dataset, and speed transition probability matrices for multiple traffic scenarios are constructed based on the vehicle driving dataset. The current journey of the target vehicle is divided into multiple sub-journeys according to the traffic scenario, and the speed transition probability matrix corresponding to each sub-journey is determined. Based on the initial speed of the target vehicle, the speed of the target vehicle at the next moment is predicted based on the speed transition probability matrix and preset constraint rules, thereby obtaining the speed estimation sequence of each sub-journey; The speed estimation sequence is smoothed, corrected, and spliced ​​to obtain the vehicle speed time series data corresponding to the current journey of the target vehicle.

[0009] Furthermore, in one embodiment of the present invention, the GPS data includes vehicle ID, vehicle latitude and longitude coordinates, and the time of data collection, and the speed data and the acceleration data are calculated by the following formula:

[0010]

[0011]

[0012] in, express Time's up The distance the vehicle traveled at any given time. Represents the Earth's radius. express The vehicle's latitude and longitude coordinates at that moment. express The vehicle's latitude and longitude coordinates at that moment. express Vehicle speed at any given moment express Vehicle acceleration at any moment express Time's up The time interval between moments.

[0013] Furthermore, in one embodiment of the present invention, the step of preprocessing the GPS data, the speed data, and the acceleration data to obtain a vehicle driving dataset specifically includes: Performing outlier detection on the GPS data, the speed data and the acceleration data, and removing the GPS data with abnormal latitude and longitude, correcting the speed data with abnormal speed and the acceleration data with abnormal acceleration; According to the speed data, performing vehicle motion state recognition, and dividing the GPS data, the speed data and the acceleration data into multiple travel records according to the recognition result; Obtaining an OSM city road network map, matching the GPS data with the nearest road segment based on a nearest neighbor matching method, and removing the GPS data, the speed data and the acceleration data with a distance between the vehicle latitude and longitude coordinates and the matched road segment greater than a preset threshold; According to the vehicle ID, the GPS data, the speed data and the acceleration data corresponding to the travel records are summarized to obtain the vehicle driving data set.

[0014] Further, in an embodiment of the present application, the traffic scenarios are divided according to the travel period, the travel area and the road type, wherein the travel period includes peak hours, off-peak hours and flat peak hours, the travel area includes main urban areas and non-main urban areas, and the road type includes urban expressways, urban expressway ramps, highways, highway ramps, main roads, main road ramps, secondary roads, secondary road ramps, branch roads and residential road.

[0015] Further, in an embodiment of the present application, the speed transition probability matrix under multiple traffic scenarios is constructed according to the vehicle driving data set, which specifically includes: According to the vehicle driving data set, the vehicle speed value range is determined, and the vehicle speed value range is discretized into multiple speed intervals according to a preset speed interval interval; Obtaining a travel record from the vehicle driving data set, and determining the travel period, the travel area and the road type according to the GPS data corresponding to the travel record, and further determining the traffic scenario type corresponding to the travel record; According to the traffic scenario type, the vehicle driving data set is divided into multiple traffic scenario travel record sets; According to the travel record set, multiple adjacent time speed pairs in the corresponding traffic scenario are determined, and the speed transition probability of the vehicle speed from a first speed interval to a second speed interval at adjacent time is calculated according to the speed pair; According to the speed transition probability, the speed transition probability matrix in the corresponding traffic scenario is determined.

[0016] Further, in an embodiment of the present application, the current journey of the target vehicle is divided into a plurality of sub-journeys according to traffic scenes, and a speed transition probability matrix corresponding to each of the sub-journeys is determined, which specifically comprises: obtaining GPS navigation information corresponding to the current journey of the target vehicle, determining a plurality of traffic scenes passed by the target vehicle according to the GPS navigation information, and dividing the current journey into a plurality of sub-journeys according to the plurality of traffic scenes passed by the target vehicle; determining the speed transition probability matrix corresponding to each of the sub-journeys according to the traffic scene corresponding to each of the sub-journeys.

[0017] Further, in an embodiment of the present application, the speed of the target vehicle at the next moment is predicted based on the speed transition probability matrix and a preset constraint rule according to the initial speed of the target vehicle, to obtain a speed estimation sequence of each of the sub-journeys, which specifically comprises: determining the travel distance of each of the sub-journeys, and initializing a cumulative estimated distance; predicting the speed of the target vehicle at the next moment based on the speed transition probability matrix corresponding to the current sub-journey and a preset high-speed preference constraint rule, low-speed jump constraint rule, and tail deceleration constraint rule according to the initial speed, and updating the cumulative estimated distance until the cumulative estimated distance equals the travel distance of the current sub-journey, to obtain the speed estimation sequence of the current sub-journey; resetting the cumulative estimated distance, taking the ending speed of the previous sub-journey as a new initial speed, and predicting the speed estimation sequence of the next sub-journey based on the speed transition probability matrix corresponding to the next sub-journey and the preset high-speed preference constraint rule, low-speed jump constraint rule, and tail deceleration constraint rule; The high-speed preference constraint rule is that if the road type of the current sub-journey is a highway / urban expressway, or the travel period of the current sub-journey is a low peak period, or the travel area of the current sub-journey is a non-main urban area, the probability weight of the high-speed interval in the corresponding speed transition probability matrix is increased. The low-speed jump constraint rule is that if the vehicle speed is in a low-speed state for a preset time length, a jump speed is determined according to the road type of the current sub-journey, and the vehicle speed is corrected to the jump speed. The tail deceleration constraint rule is that if the current sub-journey is the last sub-journey of the current journey, and the cumulative estimated distance reaches a preset proportion of the travel distance of the current sub-journey, the current speed of the target vehicle is uniformly decelerated according to a preset acceleration until the vehicle speed drops to 0.

[0018] Further, in one embodiment of the present application, the speed smoothing, correction and sequence splicing of the speed estimation sequence are performed to obtain the vehicle speed time series data corresponding to the current trip of the target vehicle, which specifically includes: performing smoothing processing on the speed estimation sequence of each sub-trip by an exponential smoothing method; performing scaling correction on the speed estimation sequence according to the distance error of each sub-trip; performing splicing processing on the speed estimation sequence according to the sequence of each sub-trip to obtain the vehicle speed time series data.

[0019] On the other hand, an embodiment of the present application provides a vehicle speed estimation device based on transition probability matrix and rule constraint optimization, which comprises: a data acquisition module configured to acquire GPS data in a vehicle driving process, and calculate corresponding speed data and acceleration data according to the GPS data; a speed transition probability matrix construction module configured to perform data preprocessing on the GPS data, the speed data and the acceleration data to obtain a vehicle driving data set, and construct a speed transition probability matrix under multiple traffic scenarios according to the vehicle driving data set; a sub-trip division module configured to divide a current trip of a target vehicle into multiple sub-trips according to traffic scenarios, and determine the speed transition probability matrix corresponding to each sub-trip; a speed estimation module configured to predict the speed of the target vehicle at the next time based on the speed transition probability matrix and a preset constraint rule according to the initial speed of the target vehicle, to obtain a speed estimation sequence of each sub-trip; an estimation sequence processing module configured to perform speed smoothing, correction and sequence splicing on the speed estimation sequence to obtain vehicle speed time series data corresponding to the current trip of the target vehicle.

[0020] On the other hand, an embodiment of the present application provides an electronic device, which comprises: at least one processor; at least one memory configured to store at least one program; when the at least one program is executed by the at least one processor, the at least one processor implements the above-mentioned vehicle speed estimation method based on transition probability matrix and rule constraint optimization.

[0021] On the other hand, an embodiment of the present application further provides a computer readable storage medium, which stores a computer program executable by a processor, and the computer program implements the above-mentioned vehicle speed estimation method based on transition probability matrix and rule constraint optimization when executed by the processor.

[0022] In another aspect, the embodiments of the present application also provide a computer program product comprising a computer program which, when executed by a processor, implements the vehicle speed estimation method based on transition probability matrix and rule constraint optimization described above.

[0023] The advantages and beneficial effects of the present application will be partially given in the following description, partially will become obvious from the following description, or will be understood by the practice of the present application: The embodiments of the present application obtain GPS data in the vehicle driving process, calculate corresponding speed data and acceleration data according to the GPS data, perform data preprocessing on the GPS data, the speed data and the acceleration data, obtain a vehicle driving data set, construct speed transition probability matrices in multiple traffic scenarios according to the vehicle driving data set, divide a current trip of a target vehicle into multiple sub-trips according to the traffic scenarios, determine speed transition probability matrices corresponding to the sub-trips, predict a speed of the target vehicle at a next time based on the speed transition probability matrices and a preset constraint rule according to a starting speed of the target vehicle, obtain a speed estimation sequence of each sub-trip, and perform speed smoothing, correction and sequence splicing on the speed estimation sequence to obtain vehicle speed time series data corresponding to the current trip of the target vehicle. The embodiments of the present application extract speed change probability features in different traffic scenarios to form corresponding speed transition probability matrices, perform vehicle speed estimation at each time based on the speed transition probability matrices, improve the accuracy of vehicle speed estimation, overcome problems such as long-distance error accumulation, speed fluctuation and low-speed state retention in speed estimation, and have strong robustness in different trip distances and are suitable for vehicle speed estimation in different distances in multiple traffic scenarios. BRIEF DESCRIPTION OF DRAWINGS

[0024] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following introduces the drawings needed to be used in the embodiments of the present application. It should be understood that the drawings introduced in the following are only for the convenience of clearly describing part of the embodiments in the technical solutions of the present application, and other drawings can be obtained by those skilled in the art without creative labor on the premise that the drawings are not attached.

[0025] Figure 1 A step flow chart of a vehicle speed estimation method based on transition probability matrix and rule constraint optimization provided by the embodiments of the present application is shown in the following figure: Figure 2 A principle schematic diagram of a vehicle speed estimation method provided by the embodiments of the present application is shown in the following figure: Figure 3 A vehicle GPS data spatial distribution schematic diagram provided by the embodiments of the present application is shown in the following figure: Figure 4A schematic diagram of a speed transition probability matrix of a main urban area provided by an embodiment of the present application is shown in FIG. 1. Figure 5 A schematic diagram of a speed transition probability matrix of a non-main urban area provided by an embodiment of the present application is shown in FIG. 2. Figure 6 A comparison diagram of a speed estimation sequence of a sub-trip and a true value provided by an embodiment of the present application is shown in FIG. 3. Figure 7 A comparison diagram of speed estimation results of different speed estimation methods provided by an embodiment of the present application is shown in FIG. 4. Figure 8 A performance comparison diagram of different speed estimation methods on different trip distances provided by an embodiment of the present application is shown in FIG. 5. Figure 9 A structural block diagram of a vehicle speed estimation device based on transition probability matrix and rule constraint optimization provided by an embodiment of the present application is shown in FIG. 6. Figure 10 A structural block diagram of an electronic device provided by an embodiment of the present application is shown in FIG. 7. DETAILED DESCRIPTION

[0026] In order to make the objects, technical solutions and advantages of the present application clearer, the present application will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and do not limit the present application. When the following description refers to the accompanying drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The implementation described in the following exemplary embodiments does not represent all the implementations consistent with the embodiments of the present application, but is only an example of devices and methods consistent with some aspects of the embodiments of the present application as described in the appended claims.

[0027] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the present application belongs. The terms used herein are only for the purpose of describing the embodiments of the present application and are not intended to limit the present application.

[0028] The vehicle speed estimation method based on transition probability matrix and rule constraint optimization provided by the embodiments of the present application can be applied to a terminal, can also be applied to a server, and can further be software running in the terminal or the server. In some embodiments, the terminal can be a smart phone, a tablet computer, a notebook computer, a desktop computer, a smart speaker, a smart watch, a vehicle-mounted terminal, and the like, but is not limited thereto. The server end can be configured as a standalone physical server, can also be configured as a server cluster or a distributed system composed of multiple physical servers, can further be configured as a cloud server providing basic cloud computing services such as cloud service, cloud database, cloud computing, cloud function, cloud storage, network service, cloud communication, middleware service, domain name service, security service, CDN, and big data and artificial intelligence platform, and the server can also be a node server in a blockchain network. The software can be an application that implements the vehicle speed estimation method based on transition probability matrix and rule constraint optimization, and the like, but is not limited to the above forms.

[0029] The present application can be used in many general or special computer system environments or configurations. For example: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, distributed computing environments including any of the above systems or devices, and the like. The present application can be described in the general context of computer-executable instructions executed by a computer, such as a program module. Generally, program modules include routines, programs, objects, components, data structures, and the like that perform specific tasks or implement specific abstract data types. The present application can also be practiced in a distributed computing environment, in which tasks are performed by remote processing devices connected by a communication network. In a distributed computing environment, program modules can be located in local and remote computer storage media, including storage devices.

[0030] It should be noted that in each specific embodiment of the present application, when relevant processing is required according to user information, user behavior data, user history data, and user location information, and the like related to the identity or characteristics of the user, the user's permission or consent will be obtained first, and the collection, use, and processing of these data will comply with relevant laws, regulations, and standards. In addition, when the embodiments of the present application need to obtain sensitive personal information of the user, the separate permission or separate consent of the user will be obtained through a pop-up window or a jump to a confirmation page, and after obtaining the separate permission or separate consent of the user, the necessary user-related data for enabling the embodiments of the present application to normally run will be obtained.

[0031] Reference Figure 1The embodiment of the present application provides a vehicle speed estimation method based on transition probability matrix and rule constraint optimization, and specifically comprises the following steps: S101, GPS data in a vehicle driving process is acquired, and corresponding speed data and acceleration data are calculated according to the GPS data; S102, data preprocessing is performed on the GPS data, the speed data and the acceleration data, a vehicle driving data set is obtained, and a speed transition probability matrix under multiple traffic scenes is constructed according to the vehicle driving data set; S103, a current journey of a target vehicle is divided into multiple sub-journeys according to traffic scenes, and speed transition probability matrices corresponding to the sub-journeys are determined; S104, a speed of the target vehicle at a next moment is predicted based on the speed transition probability matrix and a preset constraint rule according to a starting speed of the target vehicle, and a speed estimation sequence of each sub-journey is obtained; S105, speed smoothing, correction and sequence splicing are performed on the speed estimation sequence, and vehicle speed time series data corresponding to the current journey of the target vehicle is obtained.

[0032] Specifically, the embodiment of the present application extracts speed change probability features under different traffic scenes to form corresponding speed transition probability matrices, estimates the vehicle speed at each moment based on the speed transition probability matrices, improves the accuracy of vehicle speed estimation, overcomes problems such as long-distance error accumulation, speed fluctuation and low-speed state retention in speed estimation, has strong robustness under different travel distances, and is suitable for vehicle speed estimation of different distance journeys in multiple traffic scenes.

[0033] As Figure 2 shown is a principle schematic diagram of the vehicle speed estimation method provided by the embodiment of the present application, which shows a speed estimation process of a vehicle from a starting point (speed ) to an ending point (speed ), and the vehicle speed estimation method of the embodiment of the present application will be introduced completely in combination with Figure 2 .

[0034] Further, as an optional implementation manner, the GPS data comprises a vehicle ID, vehicle longitude and latitude coordinates and a collection moment, and the speed data and the acceleration data are calculated through the following formula:

[0035]

[0036]

[0037] wherein, represents a moment to vehicle driving distance at the time, denotes the radius of the earth, denotes vehicle latitude and longitude coordinates at the time, denotes vehicle latitude and longitude coordinates at the time, denotes vehicle speed at the time, denotes vehicle acceleration at the time, denotes time interval from time interval.

[0038] Specifically, high-resolution GPS data during vehicle driving is obtained, and the data fields include vehicle number, collection date, collection time, longitude and latitude; latitude and longitude coordinates of two adjacent data records and According to the Haversine formula, the vehicle driving distance is calculated, and the calculation formula is as follows:

[0039] In the formula, r is the radius of the earth, and the default value is 6378km; The speed and acceleration corresponding to each data record are calculated, and the calculation formula is as follows:

[0040]

[0041] In the formula, T is the fixed time interval of GPS data collection.

[0042] Further, as an optional implementation, the GPS data, speed data and acceleration data are preprocessed to obtain a vehicle driving data set, which specifically includes: S1021, detecting outliers of the GPS data, speed data and acceleration data, and removing the GPS data with latitude and longitude outliers, and correcting the speed data with speed outliers and the acceleration data with acceleration outliers; S1022, identifying the vehicle motion state according to the speed data, and dividing the GPS data, speed data and acceleration data into multiple travel records according to the identification result; S1023. Obtain the OSM city road network map, match GPS data with the nearest road segment based on the nearest neighbor matching method, and remove GPS data, speed data and acceleration data where the distance between the vehicle's latitude and longitude coordinates and the matched road segment is greater than a preset threshold. S1024. Summarize the GPS data, speed data, and acceleration data corresponding to the travel records based on the vehicle ID to obtain the vehicle driving dataset.

[0043] Specifically, GPS data, velocity data, and acceleration data undergo data preprocessing, including outlier handling, motion state identification, and map matching; for latitude and longitude anomalies, if the longitude or latitude field has a value of -99, the record is removed; for velocity and acceleration anomalies, if... and ,when At that time, ;if or ,but If a vehicle's speed is below 5 km / h for 5 consecutive minutes, the vehicle's movement status is identified as a stopped state to distinguish between two consecutive trips. A trip number field is added to the data record. An OSM city road network map is obtained. Since the nearest neighbor matching method is fast in large-scale map matching scenarios but easily affected by noise, the nearest neighbor matching method is used to match GPS data with the nearest road segment, and data records where the distance between the GPS point and the matched road segment is greater than 50m are removed. Based on the vehicle number, the preprocessed data results are divided into training and testing sets.

[0044] Since vehicle-mounted sensor GPS signals may be affected by factors such as building obstruction, weather changes, and multipath effects, resulting in position drift or data loss, it is necessary to process outliers in vehicle GPS data to avoid interference from erroneous data in subsequent speed estimation. This includes mapping GPS points to the road network through map matching to correct positioning deviations, and identifying vehicle parking status to exclude vehicle parking data that is irrelevant to speed estimation.

[0045] As an optional implementation method, traffic scenarios are further divided according to travel time, travel area, and road type. Travel time includes peak hours, off-peak hours, and off-peak hours; travel area includes main urban area and non-main urban area; and road type includes urban expressway, urban expressway ramp, highway, highway ramp, arterial road, arterial road ramp, secondary arterial road, secondary arterial road ramp, branch road, and residential road.

[0046] Specifically, under the comprehensive action of road design, traffic control, user behavior and environmental factors, the speed variation characteristics of vehicles in different traffic scenarios are significantly different. Markov chain is a random process with Markov property, and the core feature is the memoryless, that is, the probability of the current state changing to the next state is only related to the current state, and is irrelevant to the previous state.

[0047] The speed variation can be regarded as a random process with Markov property, and the speed of the vehicle at the next time only depends on the speed at the current time, so the speed variation characteristics of vehicles in different traffic scenarios can be represented by the speed transition probability matrix of the vehicle.

[0048] In the embodiments of the present application, traffic scenarios are divided into 60 kinds according to travel period, travel area and road type, wherein the travel period includes peak period, off-peak period and flat peak period, the travel area includes main urban area and non-main urban area, and the road type includes 10 types, which are urban expressway, urban expressway ramp, expressway, expressway ramp, main road, main road ramp, secondary road, secondary road ramp, branch road and residential road.

[0049] Further as an optional implementation, the speed transition probability matrix in multiple traffic scenarios is constructed according to the vehicle driving data set, which specifically includes: S1025, determining the vehicle speed value range according to the vehicle driving data set, and discretizing the vehicle speed value range into multiple speed intervals according to the preset speed interval interval; S1026, obtaining the travel record from the vehicle driving data set, and determining the travel period, the travel area and the road type according to the GPS data corresponding to the travel record, and further determining the traffic scenario type corresponding to the travel record; S1027, dividing the vehicle driving data set into multiple traffic scenario travel record sets according to the traffic scenario type; S1028, determining the speed binary tuple of multiple adjacent time points in the corresponding traffic scenario according to the travel record set, and calculating the speed transition probability of the vehicle speed from the first speed interval to the second speed interval at adjacent time points according to the speed binary tuple; S1029, determining the speed transition probability matrix in the corresponding traffic scenario according to the speed transition probability.

[0050] Specifically, according to the characteristics of the training set data, the speed value range is set as [0, V], the speed interval value is 1 km / h, and the speed range is discretized into V speed intervals; the data records of each trip are extracted from the training set according to the data field trip number, in order to ensure the continuity of the speed of adjacent records, the time difference between adjacent data records is calculated, and the data records with a time difference exceeding the sampling interval time T are filtered; the trip period is determined according to the collection time of the GPS data, the trip area is determined according to the latitude and longitude, the road type is determined according to the matching road section of the GPS point, and the training set data is classified according to the traffic scene; under each traffic scene, when the current speed belongs to the interval, the probability that the speed at the next moment belongs to the interval can be calculated according to the following formula:

[0051] , is the probability of the vehicle transferring from the speed interval to the speed interval , and represents the sample number of the vehicle transferring from the speed interval to the speed interval . The speed transfer probability matrix under each traffic scene is generated, and the formula is as follows:

[0052] In the formula, P is the speed transfer probability matrix.

[0053] Further, as an optional implementation, the current trip of the target vehicle is divided into multiple sub-trips according to the traffic scene, and the speed transfer probability matrix corresponding to each sub-trip is determined, which specifically includes: S1031, acquiring the GPS navigation information corresponding to the current trip of the target vehicle, determining the multiple traffic scenes passed by the target vehicle according to the GPS navigation information, and dividing the current trip into multiple sub-trips according to the multiple traffic scenes passed by the target vehicle; S1032, determining the corresponding speed transfer probability matrix according to the traffic scene corresponding to each sub-trip.

[0054] Further, as an optional implementation, the speed of the target vehicle at the next moment is predicted based on the speed transfer probability matrix and the preset constraint rule according to the initial speed of the target vehicle, and the speed estimation sequence of each sub-trip is obtained, which specifically includes: S1041, determining the trip distance of each sub-trip and initializing the cumulative estimated distance; ​​​S1042, according to the initial speed, based on the speed transfer probability matrix corresponding to the current sub-trip and the preset high-speed preference constraint rule, low-speed jump constraint rule, tail deceleration constraint rule, the speed of the target vehicle at the next moment is predicted, and the cumulative estimated distance is updated until the cumulative estimated distance is equal to the travel distance of the current sub-trip, and the speed estimation sequence of the current sub-trip is obtained; S1043, reset the cumulative estimated distance, take the end speed of the last sub-trip as the new initial speed, and predict the speed estimation sequence of the next sub-trip based on the speed transfer probability matrix corresponding to the next sub-trip and the preset high-speed preference constraint rule, low-speed jump constraint rule, and tail deceleration constraint rule; The high-speed preference constraint rule is: if the road type of the current sub-trip is highway / urban expressway, or the travel time period of the current sub-trip is low peak period, or the travel area of the current sub-trip is non-main urban area, the probability weight of the high-speed interval in the corresponding speed transfer probability matrix is increased. The low-speed jump constraint rule is: if the vehicle speed is in the low-speed state for a preset time length, the corresponding jump speed is determined according to the road type of the current sub-trip, and the vehicle speed is corrected to the jump speed. The tail deceleration constraint rule is: if the current sub-trip is the last sub-trip of the current trip, and the cumulative estimated distance reaches a preset proportion of the travel distance of the current sub-trip, the current speed of the target vehicle is uniformly decelerated according to the preset acceleration until the vehicle speed drops to 0.

[0055] Specifically, the vehicle single trip is divided into several sub-trips according to traffic scene classification, and the travel distance corresponding to each sub-trip is calculated according to the latitude and longitude of the first and last data records ; initialize the cumulative estimated distance ; when estimating the speed of each sub-trip respectively, the high-speed preference mechanism, low-speed jump mechanism and tail deceleration mechanism are added as constraint rules.

[0056] High-speed preference mechanism: if the road type of the current traffic scene is highway, urban expressway, or the travel time period is in the low peak period, or the travel area is in the non-main urban area, the probability weight of the high-speed interval (>60km / h) in the corresponding speed transfer probability matrix is increased. Low-speed jump mechanism: if the vehicle speed is in the low-speed state for 1 minute, set the jump speed; the low-speed limit of highway and urban expressway is 40 km / h, and the jump speed ; the low-speed limit of other road types is 10 km / h, and the jump speed ; Tail deceleration mechanism: when the last sub-trip is estimated speed sequence and the cumulative estimated distance , the vehicle decelerates at-1.5 m / s2 accelerates uniformly to 0 km / h, the vehicle speed at the next moment .

[0057] Since the vehicle travels at a high speed in the highway, urban expressway, off-peak period or non-main urban area, the high-speed interval accounts for a large proportion of the vehicle speed, therefore, by increasing the probability weight of the high-speed interval in the speed transition probability matrix, the high-speed preference of the vehicle in such traffic scenarios is strengthened, and the fitting degree of the speed estimation sequence to the actual driving condition of the vehicle is improved; by setting a low-speed escape mechanism, the vehicle speed is prevented from being in the low-speed interval for a long time, and the error of the speed estimation sequence is reduced; by setting a tail deceleration mechanism, the speed estimation sequence is ensured to decrease to 0 km / h at the end of the trip, and the accuracy of the speed estimation sequence is improved.

[0058] Based on the above-described constraint rules, the speed at the next moment is estimated according to the current speed and the speed transition probability matrix of the current traffic scenario, and the cumulative estimated distance is updated synchronously until , The update formula is as follows:

[0059] The ending speed of the previous sub-trip will be used as the starting speed of the next sub-trip, and the speed estimation of the next sub-trip will be performed, and so on, so that the speed estimation sequence of each sub-trip is finally obtained .

[0060] Further as an optional implementation, the speed estimation sequence is subjected to speed smoothing, correction and sequence splicing to obtain the vehicle speed time series data corresponding to the current trip of the target vehicle, which specifically includes: S1051, smoothing the speed estimation sequence of each sub-trip by an exponential smoothing method; S1052, scaling and correcting the speed estimation sequence according to the distance error of each sub-trip; S1053, splicing the speed estimation sequence according to the sequence of each sub-trip to obtain the vehicle speed time series data.

[0061] Specifically, in order to improve the stability of the sequence, the exponential smoothing method is used to smooth the speed estimation sequence of each sub-trip, and the formula is as follows:

[0062] In the formula, is a preset weighting coefficient; In order to improve the accuracy of the speed estimation sequence, the speed estimation sequence is scaled according to the distance error of each sub-trip, and the formula is as follows:

[0063]

[0064] In the formula, is a distance error coefficient; is a speed estimation sequence of the i th sub-trip, is a length of the speed estimation sequence of the sub-trip; connecting all the speed estimation sequences of the sub-trips, finally obtaining vehicle speed time series data corresponding to a single trip of the vehicle.

[0065] The method steps of the embodiment of the application are described above. The vehicle speed estimation method of the application is further described below in combination with a specific embodiment.

[0066] The data source of the embodiment of the application is 20 taxi trip GPS data of a province from June 24, 2024 to August 25, 2024, the sampling interval time is 2s, a total of 38440201 original GPS data records, and the data fields include vehicle number “DEVICE_ID”, collection date “DATE”, collection time “TIME”, longitude “LONGITUDE” and latitude “LATITUDE” and the like.

[0067] Figure 3 The vehicle GPS data spatial distribution diagram example in the embodiment of the application is shown, and it can be seen that the example vehicle mainly travels in the main urban area of the city.

[0068] Firstly, the vehicle travel distance, speed and acceleration are calculated according to the GPS data, and the GPS data, speed data and acceleration data are preprocessed, including outlier processing, motion state recognition and map matching.

[0069] In the embodiment of the application, the GPS data of 19 vehicles is taken as a training set, the total number of original data is 36043513, and the effective data accounts for 88.00%; the GPS data of the remaining 1 vehicle is taken as a test set, the total number of original data is 2396688, and the effective data accounts for 90.58%.

[0070] The speed transition probability matrix of different traffic scenes is calculated. In the embodiment of the application, according to the characteristics of the training set data, the speed value range is [0, 150]; the data records of each trip of the vehicle are extracted from the training set, the time difference of adjacent data records is calculated, and the data records with a time difference exceeding the sampling interval time of 2s are filtered; when calculating the speed transition probability matrix, the weighting coefficient α is 0.5.

[0071] As Figure 4 The main urban area speed transition probability matrix (after logarithmic transformation) in the embodiment of the application is shown in the schematic diagram, which is (a) low peak period, (b) peak period and (c) flat peak period, as Figure 5Fig. 1 shows a schematic diagram of the non-main urban speed transition probability matrix (after logarithmic transformation) in the embodiment of the present application, which is respectively (a) low peak, (b) peak and (c) flat peak, and the darker the color, the greater the transition probability.

[0072] According to the traffic scene classification, the single trip of the vehicle in the training set is divided into several sub-trips, and the travel distance corresponding to each sub-trip is calculated according to the latitude and longitude of the first and last data records of the sub-trip; under the constraints of the high-speed preference mechanism, the low-speed jump-off mechanism and the tail deceleration mechanism, the speed of each sub-trip is estimated.

[0073] The exponential smoothing method is used to smooth the vehicle speed estimation sequence of step S104, the weighted coefficient a is 0.5; the speed estimation sequence is scaled according to the distance error of each sub-trip. The speed estimation sequence of all sub-trips is connected, and the speed estimation sequence corresponding to the single trip of the vehicle is obtained.

[0074] As shown in Fig. 2, a comparison diagram of the sub-trip speed estimation sequence and the true value in the embodiment of the present application is shown, it can be found that in the embodiment of the present application, after the exponential smoothing and distance error adjustment, the noise of the speed estimation sequence is reduced and more stable, and the distance error is smaller. Figure 6

[0075] In order to evaluate the performance of the vehicle speed estimation method, the absolute error (MAE), mean square error (MSE) and mean absolute percentage error (MAPE) are selected as evaluation indexes in the embodiment of the present application, and the calculation formula is as follows:

[0076]

[0077]

[0078] In the formula, v is the estimated speed, and v is the true speed. According to the foregoing speed estimation steps, the speed estimation of the test set is performed based on the speed probability transition matrix constructed in the embodiment of the present application. Compared with other vehicle speed estimation methods, the performance of the present application on the test set is observed.

[0079] As shown in Fig. 3, a speed estimation sequence example of different speed estimation methods provided in the embodiment of the present application is shown, which is respectively (a) Proposed (proposed in the present application), (b) SA&SO, (c) MCMC&SO and (d) MCMC. It can be found that the speed estimation sequence of the present application has good speed continuity and speed change smoothness. Figure 7

[0080] ​​​​The performance evaluation results of different speed estimation methods are shown in Table 1.

[0081] Table 1

[0082] From the results in Table 1, it can be found that the MAE, MSE and MAPE indicators of the present application are better than those of other baseline methods, which indicates that the present application can improve the accuracy of speed estimation.

[0083] Further, the total distance error and its percentage error are calculated to compare the performance of each method at different travel distances. The calculation formula is as follows:

[0084]

[0085] In the formula, is the estimated total travel distance, i.e., the sum of the sub-trip distances; D is the actual total travel distance.

[0086] As Figure 8 shown in the performance comparison diagram of different speed estimation methods at different travel distances in the embodiment of the present application, including (a) MAPE comparison and (b) average distance error comparison, it can be found that even if the travel distance increases, the MAPE, total distance error and its percentage error of the present application are all lower and relatively stable, which indicates that the present application can overcome the problem of error accumulation and has good robustness at different travel distances, suitable for long-distance speed estimation scenarios.

[0087] It can be realized that the embodiment of the present application extracts the speed change probability features in different traffic scenarios to form the corresponding speed transition probability matrix, and estimates the vehicle speed at each time based on the speed transition probability matrix, which improves the accuracy of vehicle speed estimation. Through rule constraint optimization and speed smoothing correction, the problems of long-distance error accumulation, speed fluctuation and low-speed state retention in speed estimation are overcome, and the robustness is good at different travel distances, suitable for vehicle speed estimation of different distance trips in various traffic scenarios. Compared with the prior art, the present application performs well in different traffic scenarios and different travel distances.

[0088] Referring to Figure 9 , the embodiment of the present application provides a vehicle speed estimation device based on transition probability matrix and rule constraint optimization, comprising: a data acquisition module for acquiring GPS data during vehicle travel, and calculating corresponding speed data and acceleration data according to the GPS data; The speed transition probability matrix construction module is configured to perform data preprocessing on the GPS data, the speed data and the acceleration data, obtain a vehicle driving data set, and construct a speed transition probability matrix in multiple traffic scenarios according to the vehicle driving data set. The sub-trip division module is configured to divide a current trip of the target vehicle into multiple sub-trips according to the traffic scenario, and determine a speed transition probability matrix corresponding to each sub-trip. The speed estimation module is configured to predict a speed of the target vehicle at a next time according to a starting speed of the target vehicle, based on the speed transition probability matrix and a preset constraint rule, and obtain a speed estimation sequence of each sub-trip. The estimation sequence processing module is configured to perform speed smoothing, correction and sequence splicing on the speed estimation sequence, and obtain vehicle speed time series data corresponding to the current trip of the target vehicle.

[0089] It can be understood that the contents in the above method embodiments are applicable to the device embodiments, the device embodiments specifically implement the functions of the above method embodiments, and achieve the same beneficial effects as the above method embodiments.

[0090] With reference to Figure 10 The device embodiment of the present application provides an electronic device, which comprises: at least one processor; at least one memory configured to store at least one program; When the at least one program is executed by the at least one processor, the at least one processor implements the above vehicle speed estimation method based on transition probability matrix and rule constraint optimization.

[0091] It can be understood that the contents in the above method embodiments are applicable to the device embodiments, the device embodiments specifically implement the functions of the above method embodiments, and achieve the same beneficial effects as the above method embodiments.

[0092] The device embodiment of the present application further provides a computer readable storage medium, which stores a computer program executable by a processor, and the computer program implements the above vehicle speed estimation method based on transition probability matrix and rule constraint optimization when executed by the processor.

[0093] The computer readable storage medium of the device embodiment of the present application can execute the vehicle speed estimation method based on transition probability matrix and rule constraint optimization provided by the method embodiment of the present application, execute any combination of the steps of the method embodiment, and has the corresponding functions and beneficial effects of the method.

[0094] The embodiment of the present application further provides a computer program product comprising a computer program which, when executed by a processor, implements the vehicle speed estimation method based on a transition probability matrix and rule constraints optimization.

[0095] It can be understood that the contents in the method embodiments are applicable to the program product embodiments, the program product embodiments specifically implement the functions of the method embodiments, and achieve the same beneficial effects as the method embodiments.

[0096] The memory, as a non-transitory computer readable storage medium, can be used to store non-transitory software programs and non-transitory computer executable programs. In addition, the memory can include a high-speed random access memory, and can also include a non-transitory memory, such as at least one magnetic disk storage device, a flash memory device, or other non-transitory solid-state memory device. In some embodiments, the memory can optionally include a memory remotely arranged relative to the processor, and these remote memories can be connected to the processor through a network. Examples of the above network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0097] The embodiments described in the embodiments of the present application are used to more clearly illustrate the technical solutions of the embodiments of the present application, and do not constitute a limitation on the technical solutions provided by the embodiments of the present application. Those skilled in the art can know that, with the evolution of technology and the appearance of new application scenarios, the technical solutions provided by the embodiments of the present application are also applicable to similar technical problems.

[0098] The terms "first", "second", "third", "fourth" and the like (if any) in the specification of the present application and the above drawings are used to distinguish similar objects, and do not necessarily have to be used to describe a specific order or sequence. It should be understood that the data thus used can be interchanged under appropriate circumstances, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, system, product or device including a series of steps or units does not have to be limited to those steps or units clearly listed, but can include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0099] In alternative embodiments, the functions / operations in the flow diagrams can occur in sequences other than those depicted. For example, two operations shown in succession can in fact be executed substantially concurrently or the operations can sometimes be executed in the reverse order depending upon the functionality / operations involved. Such variations are contemplated to be within the scope of the present application. Embodiments presented and described in the flow diagrams are examples only and are used to provide an overall understanding of the method of the application. The disclosed methods are not limited to the operations and logical flows presented in this application. Alternative embodiments are contemplated in which the order of various operations is changed and in which sub-operations of a larger operation are performed in parallel rather than sequentially.

[0100] Moreover, while the application has been described in the context of functional modules, it is to be understood that one or more of the functions and / or features described above can be integrated within a single physical device and / or software module, or one or more functions and / or features can be implemented in separate physical devices or software modules. It will also be appreciated that detailed discussion of the actual implementation of each module is not necessary for an understanding of the application. Rather, the properties, functions and internal relationships of the various functional modules disclosed in the devices herein are considered to be of a manner that would be within the skill of an engineer given the benefit of this disclosure. Thus, a person of ordinary skill in the art using routine skill would be able to implement the application as set forth in the claims without undue experimentation, given the benefit of this disclosure. It is also to be understood that the specific concepts disclosed are intended to be illustrative only and not limiting of the scope of the application as set forth in the claims and equivalents thereof.

[0101] If the above functions are implemented in the form of software function units and sold or used as independent products, they can be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the present application essentially or the part that contributes to the prior art or the part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described above in the various embodiments of the present application. The aforementioned storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk or an optical disk, and various media that can store program codes.

[0102] The logic and / or steps represented in the flow diagrams or otherwise described herein, for example, can be embodied in non-transitory computer- readable medium using any combination of hardware, software, and / or firmware. The logic and / or steps can be implemented using any of various computer- readable media for storing or transmitting this computer-readable instructions, such as magnetic storage media (e.g., hard disks), optical storage media (e.g., CD-ROMs, DVDs), nonvolatile memory storage media (e.g., ROMs, EPROMs, EEPROMs), and / or flash memory devices.

[0103] More specific examples (a non-exhaustive list) of the computer-readable medium include the following: an electrical connection (electronic) having one or more wires, a portable computer diskette (magnetic), a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, and a portable compact disc read-only memory (CDROM). Additionally, the computer-readable medium can also be paper or another suitable medium upon which the program is printed, as the program can be electronically captured, for example, via optical scanning of the paper or other medium, then compiled, interpreted, or otherwise processed in a suitable manner, if necessary, and then stored in a computer memory.

[0104] It should be understood that aspects of the application can be implemented in hardware, software, firmware, or combinations thereof. In the above embodiments, various steps or methods can be implemented in software or firmware that is stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, any of the following technologies, or combinations thereof, can be used: a discrete logic circuit having logic gates for implementing logic functions upon data signals, an application specific integrated circuit having appropriate combinational logic gates, a programmable gate array (PGA), a field programmable gate array (FPGA), and / or the like.

[0105] In the above description of the present application, the description using the terms "one embodiment", "another embodiment" or "some embodiments", and the like, means that the particular feature, structure, material, or characteristic being described in connection with the embodiment or example is included in at least one embodiment or example of the application. The appearances of the above-described terms in various places in the specification are not necessarily referring to the same embodiment or example. Further, where a particular feature, structure, material, or characteristic is described in connection with an embodiment or example, it is submitted that it is within the purview of one of ordinary skill in the art to effect such feature, structure, material or characteristic in connection with another embodiment or example, whether or not it is described in the specification or shown in the drawings.

[0106] While the embodiments of the application have been shown and described, it is to be understood that the embodiments described are merely exemplary and are not to be construed as limiting the scope of the application. The scope of the application is defined by the appended claims and their equivalents.

[0107] The above is the specific description of the preferred embodiment of the present application, but the present application is not limited to the above-mentioned embodiments, and those skilled in the art can make various equivalent modifications or replacements without departing from the spirit of the present application, and these equivalent modifications or replacements are all included in the scope defined by the claims of the present application.

Claims

1. A vehicle speed estimation method based on transition probability matrix and rule constraint optimization, characterized in that, Includes the following steps: The system acquires GPS data during vehicle operation and calculates the corresponding speed and acceleration data based on the GPS data. The GPS data, the speed data, and the acceleration data are preprocessed to obtain a vehicle driving dataset, and speed transition probability matrices for multiple traffic scenarios are constructed based on the vehicle driving dataset. The current journey of the target vehicle is divided into multiple sub-journeys according to the traffic scenario, and the speed transition probability matrix corresponding to each sub-journey is determined. Based on the initial speed of the target vehicle, the speed of the target vehicle at the next moment is predicted based on the speed transition probability matrix and preset constraint rules, thereby obtaining the speed estimation sequence of each sub-journey; The speed estimation sequence is smoothed, corrected, and spliced ​​to obtain the vehicle speed time series data corresponding to the current journey of the target vehicle.

2. The vehicle speed estimation method based on transition probability matrix and rule constraint optimization according to claim 1, characterized in that, The GPS data includes the vehicle ID, vehicle latitude and longitude coordinates, and the time of data collection. The speed data and the acceleration data are calculated using the following formula: in, express Time's up The distance the vehicle traveled at any given time. Represents the Earth's radius. express The vehicle's latitude and longitude coordinates at that moment. express The vehicle's latitude and longitude coordinates at that moment. express Vehicle speed at any given moment express Vehicle acceleration at any moment express Time's up The time interval between moments.

3. The vehicle speed estimation method based on transition probability matrix and rule constraint optimization according to claim 1, characterized in that, The process of preprocessing the GPS data, the speed data, and the acceleration data to obtain a vehicle driving dataset specifically includes: Anomaly detection is performed on the GPS data, the speed data, and the acceleration data. GPS data with abnormal latitude and longitude is removed, and speed data and acceleration data with abnormal acceleration are corrected. The vehicle motion status is identified based on the speed data, and the GPS data, speed data, and acceleration data are divided into multiple travel records based on the identification results. Obtain an OSM city road network map, match the GPS data with the nearest road segment based on the nearest neighbor matching method, and remove the GPS data, speed data, and acceleration data where the distance between the vehicle's latitude and longitude coordinates and the matched road segment is greater than a preset threshold; The GPS data, speed data, and acceleration data corresponding to the travel records are summarized based on the vehicle ID to obtain the vehicle driving dataset.

4. The vehicle speed estimation method based on transition probability matrix and rule constraint optimization according to claim 1, characterized in that: Traffic scenarios are categorized based on travel time, travel area, and road type. Travel time includes peak hours, off-peak hours, and non-peak hours. Travel area includes the main urban area and non-main urban area. Road type includes urban expressways, urban expressway ramps, highways, highway ramps, arterial roads, arterial road ramps, secondary arterial roads, secondary arterial road ramps, branch roads, and residential roads.

5. The vehicle speed estimation method based on transition probability matrix and rule constraint optimization according to claim 1, characterized in that, The step of constructing speed transition probability matrices for multiple traffic scenarios based on the vehicle driving dataset specifically includes: The vehicle speed range is determined based on the vehicle driving dataset, and the vehicle speed range is discretized into multiple speed intervals according to a preset speed interval. Travel records are obtained from the vehicle driving dataset, and the travel time, travel area and road type are determined based on the GPS data corresponding to the travel records, thereby determining the traffic scenario type corresponding to the travel records; The vehicle driving dataset is divided into multiple travel record sets for different traffic scenarios based on the traffic scenario type. Based on the travel record set, determine multiple speed pairs for adjacent moments in the corresponding traffic scenario, and calculate the speed transition probability of the vehicle speed changing from the first speed interval to the second speed interval at adjacent moments based on the speed pairs. The speed transition probability matrix for the corresponding traffic scenario is determined based on the speed transition probability.

6. The vehicle speed estimation method based on transition probability matrix and rule constraint optimization according to claim 1, characterized in that, The step of dividing the current journey of the target vehicle into multiple sub-journeys based on the traffic scenario and determining the speed transition probability matrix corresponding to each sub-journey specifically includes: Obtain GPS navigation information corresponding to the current journey of the target vehicle, determine multiple traffic scenarios through which the target vehicle passes based on the GPS navigation information, and divide the current journey into multiple sub-journeys based on the multiple traffic scenarios through which the target vehicle passes. The corresponding speed transition probability matrix is ​​determined based on the traffic scenario corresponding to each sub-trip.

7. The vehicle speed estimation method based on transition probability matrix and rule constraint optimization according to claim 1, characterized in that, The step of predicting the speed of the target vehicle at the next moment based on the initial speed of the target vehicle, the speed transition probability matrix, and preset constraint rules, to obtain the speed estimation sequence for each of the sub-journeys, specifically includes: Determine the travel distance for each sub-trip and initialize the cumulative estimated distance; Based on the starting speed, the speed of the target vehicle at the next moment is predicted based on the speed transition probability matrix corresponding to the current sub-trip and the preset high-speed preference constraint rule, low-speed jump constraint rule, and tail deceleration constraint rule, and the cumulative estimated distance is updated until the cumulative estimated distance is equal to the travel distance of the current sub-trip, thus obtaining the speed estimation sequence of the current sub-trip; The cumulative estimated distance is reset, and the end speed of the previous sub-trip is used as the new starting speed. Based on the speed transition probability matrix corresponding to the next sub-trip and the preset high-speed preference constraint rule, low-speed jump constraint rule, and tail deceleration constraint rule, the speed estimation sequence of the next sub-trip is predicted. The high-speed preference constraint rule is as follows: if the road type of the current sub-trip is a highway / urban expressway, or the travel time of the current sub-trip is during off-peak hours, or the travel area of ​​the current sub-trip is outside the main urban area, the probability weight of the high-speed interval in the corresponding speed transition probability matrix is ​​increased. The low-speed detachment constraint rule is as follows: if the vehicle speed is in a low-speed state for a preset duration, the corresponding detachment speed is determined according to the road type of the current sub-trip, and the vehicle speed is corrected to the detachment speed. The tail deceleration constraint rule is as follows: if the current sub-trip is the last sub-trip of the current trip, and the cumulative estimated distance reaches a preset proportion of the travel distance of the current sub-trip, the current speed of the target vehicle is uniformly decelerated according to a preset acceleration until the vehicle speed drops to 0.

8. A vehicle speed estimation method based on transition probability matrix and rule constraint optimization according to any one of claims 1 to 7, characterized in that, The process of smoothing, correcting, and concatenating the speed estimation sequence to obtain the vehicle speed time-series data corresponding to the current journey of the target vehicle specifically includes: The speed estimation sequence of each sub-journey is smoothed using exponential smoothing. The speed estimation sequence is scaled and corrected based on the distance error of each sub-journey; The speed estimation sequence is spliced ​​together according to the order of each sub-journey to obtain the vehicle speed time series data.

9. A vehicle speed estimation device based on transition probability matrix and rule constraint optimization, characterized in that, include: The data acquisition module is used to acquire GPS data during vehicle operation and calculate the corresponding speed and acceleration data based on the GPS data. The speed transition probability matrix construction module is used to preprocess the GPS data, the speed data, and the acceleration data to obtain a vehicle driving dataset, and to construct speed transition probability matrices for multiple traffic scenarios based on the vehicle driving dataset. The sub-trip segmentation module is used to divide the current trip of the target vehicle into multiple sub-trips according to the traffic scenario, and determine the speed transition probability matrix corresponding to each sub-trip; The speed estimation module is used to predict the speed of the target vehicle at the next moment based on the initial speed of the target vehicle, the speed transition probability matrix, and preset constraint rules, so as to obtain the speed estimation sequence of each sub-journey. The estimation sequence processing module is used to smooth, correct, and concatenate the speed estimation sequence to obtain the vehicle speed time series data corresponding to the current journey of the target vehicle.

10. An electronic device, characterized in that, include: At least one processor; At least one memory for storing at least one program; When the at least one program is executed by the at least one processor, the at least one processor implements a vehicle speed estimation method based on transition probability matrix and rule constraint optimization as described in any one of claims 1 to 8.

Images