175 results about "Travel behavior" patented technology

An automatic driving system for vehicles safely switches a vehicle from an automatic drive mode to a manual drive mode. An automatic driving controller of the automatic driving system includes a travel behavior determining unit that determines a travel behavior of a vehicle during automatic driving based on an operational state of a steering system, an accelerating system, and / or a braking system thereof; a driver's state determining unit that determines a state of a driver based on a face image of the driver captured by a face image camera, biometric information of the driver obtained from body pressure of the driver and variation thereof detected by body pressure sensors of a driver's seat, and signals from pressure-sensitive sensors arranged at a steering wheel and the like; and a drive mode switching control unit that controls in stages switching to manual driving based on the determination results.

The invention discloses a system and method for calculating driving risks and assisting automobile insurance pricing based on multi-source data. The system includes an intelligent mobile internet terminal and a remote server connected with each other; a plurality of data sensing acquisition units and applications are installed in the intelligent mobile terminal; and the remote server is provided with a risk model algorithm system and includes a scene driving risk analysis sub module, a distracted driving model recognition sub module, a user travel behavior analysis sub module and a driving behavior evaluation and grading system sub module. According to the system and method of the invention, data acquired by the intelligent mobile internet terminal and road traffic information acquired by the remote server are analyzed, and the scores of the sub modules are calculated, and automobile insurance pricing is carried out according to the scores. With the system and method of the invention adopted, based on multi-source data acquisition and fusion, the driving behaviors of users are analyzed, and the travel habits, driving habits and driving risks of the drivers can be effectively calibrated, and therefore, theoretical basis and technical support can be provided for driving behavior-based automobile insurance pricing models.

The invention discloses a bus arrival prediction method based on real-time traffic status information, belonging to an intelligent traffic technology. The bus arrival prediction method comprises the following steps of: preprocessing bus route data; matching bus GPS (global positioning system) points collected in real time onto a bus route according to the preprocessed bus route data; judging travelling behaviors of buses on the route; and carrying out bus arrival prediction based on the real-time traffic status information according to a judgment result. With the adoption of the bus arrival prediction method provided by the invention, on the basis of GPS rapid route matching of a sampling point, the complexity of matching a GPS positioning point and the route is lowered, the arrival distance processing speed is improved, and a simple and effective manner is provided for real-time matching of large-scale GPS data. With the adoption of a bus arrival prediction model based on the real-time traffic status information, dynamic bus prediction is carried out according to the traffic conditions changed in real time; and the bus arrival prediction method has better instantaneity and accuracy.

The invention provides a mobile phone positioning data-based traveling origin-destination (OD) matrix acquisition method, which is characterized by comprising the following steps of: determining a traffic zone division scheme of a certain road net area; acquiring all the mobile phone positioning data of all traveling groups in a certain time range in a specified road net range in the road net area; performing displacement state judgment on each position point of each traveler; obtaining origins and destinations of each traveler in the travel of each time in the certain time range; and mappingthe origins and destinations onto corresponding traffic zones and performing grouping counting on all identified travel behaviors to form original traveling OD matrix data. The method has the advantage that: the traveling OD matrix data in a large city road net range can be automatically acquired with low cost and high frequency fully based on conventional mobile communication network resources by utilizing positioning information in a conventional mobile phone communication network.

The embodiment of the invention discloses a vehicle using traveling behavior prediction method and device, a server and a storage medium. The prediction method comprises the steps that scene characteristic monitoring is performed on a user to be predicted so as to acquire the scene characteristics of an instant application scene in which the user to be predicted is located; and vehicle using traveling prediction is performed according to the user portrait characteristics of the user to be predicted and the scene characteristics. With application of the method, accurate prediction of the customized vehicle using traveling behavior of the user can be realized.

The invention discloses a method for forecasting the passenger volume of urban rail transit under emergencies. The method comprises steps of S1: acquiring urban rail transit physical network information, train running information, passenger volume data and emergency information to determine the passenger volume distribution and affected passenger volume data of urban rail transit; S2: constructing a set of alternative travel plans for passengers under emergencies, updating the affected passenger volume data after the passengers select each alternative travel plan; and S3: calculating the passenger volume of all stations of urban rail transit based on unaffected passenger volume data. The method fully considers the impact of emergencies on the operation of urban rail transit and considers the characteristics of passengers' travel behaviors under emergencies to improve the accuracy of forecasting passenger volume distribution of urban rail transit under emergencies. The method provides a theoretical basis to formulate an effective emergency plan and operation management measures, can effectively divert passengers stranded due to emergencies to protect passenger travel safety, and can improve the service level of the subway operation department.

A method and system for predict that stop of passenger bus trip and disembarkation are provide, Based on the passenger historical AFC data, the travel behavior indexes are calculated, the travel characteristics of the card-swiping passengers are analyzed, and the bus travel modes are extracted, then the passenger disembarkation stations are predicted according to the real-time boarding station information, so as to provide customized and intelligent bus travel services for travelers. The passenger bus stop forecasting method and the passenger bus stop forecasting system can provide more targeted and intelligent station forecasting service for passengers, and remarkably improve the user experience of the information service system.

The present invention discloses a user heterogeneous time value and traffic jam expense budget-based traffic distribution method. According to the present invention, by adding the two traveler economic parameters of the time value and the traffic jam expense budget in a traffic distribution model, the more objective and reasonable road network flow is outputted. According to the present invention, an actual road network is simplified into an abstract traffic network, and a path selection and traffic distribution combined model considering the heterogeneous time value and the traffic jam expense budget is constructed. The time value is a parameter describing the boundary expense, namely, is the fund cost that the users are willing to save and pay for each unit time, and the traffic jam expense budget is the cash quantity that the travelers can dominate the traffic jam expense. Based on considering the traveler heterogeneity, the user travel target can change from the conventional shortest driving time to the minimum comprehensive cost after considering the traffic jam expense budget, namely, the minimum summation of the driving time and the traffic jam expense, so that, obviously, the travel behaviors of the users and the road network flow can be described more reasonably and objectively, and the reference is provided for the path planning and the traffic jam charging.

The invention belongs to the technical field of analysis over travel behaviors of residents, and particularly relates to a position-trajectory-based travel state splitting method, which comprises the following steps of performing GPS (Global Positioning System) positioning to obtain latitude and longitude coordinates of a position trajectory; calculating the distances of trajectory points, instantaneous speed and instantaneous acceleration; performing data cleaning and abnormal data processing; matching a manually identified travel manner; selecting different time intervals T, and splitting a travel state; performing characteristic extraction and sequential storage; performing Euclidean-distance-based similarity measurement on adjacent segments by taking characteristics of the travel segments as characteristic vectors, determining a proper time interval, and determining splitting points to realize the splitting of the travel state according to distance similarity measurement. According to the method, key points of the travel state can be determined more accurately, and the travel state can further be better split; compared with a conventional fixed time period and key point travel state splitting method, the method is more objective and reliable.

The invention relates to an urban rail traffic passenger flow prediction method, apparatus, a storage medium and a computer device. The method herein includes the following steps: real-time acquiring passengers' station entrance and exit data; on the basis of station entrance data and historical travel behaviors, predicting passengers' station exit places; based on the passengers' station entrance and exit data, computing passengers' travel routes by using the OD reverse algorithm, for passengers whose travel routes are not computed by using the OD reverse algorithm, using a multi-probability model to analyze passengers' station entrance and exit data so as to compute travel routes; computing passengers' computed travel routes in combination with passengers' station entrance places and exit places and real-time position information of trains to obtain passengers' real-time positions; on the basis of the real-time acquired passengers' station exit data, correcting passengers' real-time positions; and on the basis of passengers' corrected real-time positions, outputting predicted net passenger flow distribution information. According to the invention, the method herein can predict an individual passenger and makes a breakthrough to traditional modes which count statistics on entire passengers based on probability.

The invention discloses a method for establishing a public transit OD matrix, which is characterized by being based on a cloud computing platform. The method comprises the steps of screening out original public transit data, which conforms to predetermined OD matrix establishment requirements, from public transit data; carrying out preprocessing on the original public transit data in a concurrent mode according to data processing rules; extracting get-on information of each passenger according to the processed data, and analyzing a card swiping record of each passenger; if a bi-directional card swiping record exists, acquiring passenger get-off information of by using a method of estimating a passenger get-off station by analyzing passenger travel behavior rules to acquire the passenger get-off information; if the bi-directional card swiping record does not exist, acquiring the number of transferred people between stations by using a method of estimating the number of get-off people at a station by predicating the station heat degree; and establishing an OD matrix according to the get-on information and the get-off information of the passengers, the number of transferred people between stations and an OD matrix generation rule. The method disclosed by the invention can establish the OD matrix for multiple lines in real time. The invention further discloses a system for establishing the public transit OD matrix.

The invention discloses a mobile phone user travel mode recognition method based on a C4.5 decision tree. The mobile phone user travel pattern recognition method comprises: collecting mobile phone signaling data of a volunteer, and processing the mobile phone signaling data into a travel track sequence with reference to time information and position information of the data; excavating a relationship of a travel characteristic attribute and a corresponding travel traffic mode from the travel track data of the volunteer; dividing the collected mobile phone signaling data of the volunteer into two parts according to a supervised machine learning method, which are respectively a training data set and a test data set; extracting a variety of characteristic attributes of travel behaviors from mobile phone data corresponding to the travel process of the volunteer according to the communication principle and the traffic theory; training a travel mode recognition model in combination with a real travel mode condition fed back by the volunteer via an artificial intelligence algorithm, C4.5 algorithm with supervision and learning ability to obtain a high-precision mobile phone user travel mode recognition model based on the mobile phone signaling data.

This present invention is a spatial data processing system and method that allows the automatic, rapid, scalable analysis and transformation of large amounts of travel behavior data (e.g., tracking data points) into individual “points-of-interest” and discrete trips stored in a spatial database. Each trip has a point-of-interest as a starting and ending location, and contains multiple positions (e.g. latitude and longitudes) which define the travel path of the user / device during that time period.

A travel anomaly detection method based on a hidden Markov model includes the following steps that firstly, a Beidou or GPS positioning device is utilized to acquire monitoring data such as longitude, latitude and dates, and the data are sent to a travel behavior monitoring platform through a GPS module; secondly, coordinate data are divided into workday data and holiday data, and the workday data and the holiday data are clustered through a k-medoids algorithm to acquire two clustering mark point sets respectively; thirdly, the workday coordinate data and the holiday coordinate data are used as two sets of observed quantities, an HMM model is trained through a Baum-Welch algorithm, and then a workday travel rule model and a holiday travel rule model are respectively acquired; fourthly, travel behavior anomalies are detected through the travel rule models. The travel anomaly detection method is high in accuracy and good in reliability.

The invention provides a passenger flow guidance method of urban rail transit based on multi-agent simulation. The method includes: constructing a passenger flow dynamic guidance optimization bilevelprogramming model considering the scope and content of guidance information; taking the sum of the total travel cost and guidance cost of the network as the objective function, the upper optimizationmodel generates the passenger flow guidance scheme by genetic algorithm. The lower level simulation model is used to simulate the passenger 's travel behavior in the current passenger flow guidance scheme, the passenger flow and generalized cost of each section in the simulation period are inputted into the upper optimization model, The upper optimization model substitutes the passenger flow inducement scheme, the passenger flow and the generalized cost into the objective function, and obtains the fitness function under the current solution. Through multiple times of iterations, the optimal dynamic passenger flow inducement scheme with the highest fitness function is obtained. The invention can realize fine guidance of one station one scheme, assist passengers to make reasonable route decision, and alleviate local passenger flow congestion of urban rail transit network.

The present invention relates to the field of vehicle management, and particularly relates to a vehicle management system and method. The vehicle management method comprises the steps that a finger print collecting unit collects finger print information; a right management unit in a remote server performs right configuration on the collected finger print information, so as to generate right information corresponding to the finger print information; the finger print information and corresponding right information are sent to a target vehicle; a finger print identification unit identifies a current finger print input into a vehicle; and a control unit enables a vehicle function according to a right corresponding to the current finger print. While a person hiring the vehicle travels, individual configuration of the target vehicle is protected. Traveling information is collected and stored externally, so that traveling behaviors of the person hiring the vehicle are supervised effectively; or the finger print information and corresponding right information can be rewritten in a cloud server, so that the operation is convenient.

The invention discloses a travel information management method, system and device and a computer readable storage medium, and is applied to a block chain node device. The method comprises the following steps of: obtaining scenic spot information and a travel order, wherein the scenic spot information comprises reception processes of scenic spots, and the travel order comprises tourist informationand a target scenic spot of the travel; issuing an intelligent contract to a block chain and obtaining a scheduling corresponding to the travel order according to the intelligent contract, wherein theintelligent contract specifically describes a travel transaction process; when the block chain node device judges that the travel is ended according to the intelligent contract, acquiring evaluationinformation, of one party, about the other two parties of the tourists, the scenic spots and a travel agency, so that the supervision department can supervise travel behaviors of the parties through real evaluation information on the block chain, so as to constraint the travel behaviors, guide the positive development of the travel industry and improve the travel behavior supervision efficiency.

A method based on a combined component path to recognize similar travelers is characterized in that tempo-spatial GPS (global positioning system) trajectories are used to construct a heterogeneous travel network model; the definition of a travel component path is provided; a component path set is selected from the heterogeneous travel network model; user similarity calculation is achieved via a travel similarity measure algorithm; travel behavior similarity between two users is predicted. The method herein has higher recall ratio and precision ratio than single-component-based path methods; the method has good visual intelligible semantic property, and is helpful for accurately discovering potential users and similar groups to provide them with more precise and differential services.

The invention belongs to the user behavior analysis technology field and especially relates to a user behavior classification method and a system thereof. The user behavior classification method comprises the following steps of step a, according to user behavior data, constructing a user-behavior multidimensional description table, and according to the user-behavior multidimensional description table, carrying out classification annotation on a user; step b, according to the multidimensional description table and user classification, constructing a classification model, and according to the classification model, constructing a user behavior knowledge base and a rule base; and step c, collecting new user data, and through the user behavior knowledge base and the rule base, carrying out automatic classification on a new user. In the invention, through constructing a travel behavior knowledge base and the rule base of the user, based on the travel behavior knowledge base and the rule base, the new user can be automatically classified and rapid construction of a new user portrait is realized; and aiming at a distinctive user behavior type, a correlation product is recommended and effective and customized recommending of the product can be realized so that advertisement incomes of an enterprise are increased.

The invention discloses a novel heuristic algorithm for an emergency evacuation plan of an urban traffic network. The novel heuristic algorithm is implemented as follows: (1), an urban emergency evacuation network is established; (2), an optimization model for the emergency evacuation network is established; (3), an emergency evacuation planning problem is optimized by using a Lagrangian relaxation-tabu searching algorithm; and (4), a Lagrangian relaxation problem is solved; to be specific, the operation starts with an evacuation network feasible solution, searching is carried out to update an iteration network state until a stop criterion is met, and a final result is outputted. According to the invnetion, on the basis of combination of the Lagrangian relaxation technique and the tabu searching, the Lagrangian relaxation- is applied to emergency network evacuation planning based on dynamic traffic distribution; and compared with the existing emergency network evacuation plan, the novel heuristic algorithm enabling the travelling behavior of the traveler to be described well becomes efficient and universal and is suitable for a large-scale traffic evacuation network.

A method and apparatus is disclosed herein for estimating travel behavior in a mobile network. In one embodiment, the method comprises receiving event data indicative of user equipment location, pre-processing received event data to produce pre-processed data, performing straight line interpolation on pre-processed data of one or more individuals in the population to estimate intermediate positions of a trajectory of each of the one or more individuals from a first position to a second position, and counting a number of individuals in population at a given time and at a given area.

The invention discloses a vehicle trajectory prediction method and system based on video vehicle passing data. The method comprises the following steps: collecting vehicle passing data, and extractingeffective field information in the vehicle passing data; constructing a basic road network; sorting the vehicle passing data according to the sampling time to form a total travel chain of each vehicle in the sampling time period; extracting two adjacent sample sequences in turn, calculating the travel time of the vehicle passing through two adjacent intersections according to the sample sequences, judging that the two adjacent intersections are travel chain segmentation points, and segmenting the total travel chain into a series of sub-travel chains according to the travel chain segmentationpoints; calculating the steering state probability of each vehicle at each intersection according to the sub-travel chain to obtain a steering probability matrix; determining that the intersection corresponding to the maximum steering probability in the steering probability matrix is the intersection to be reached by the vehicle driving trajectory prediction. The invention avoids the shortcomingsof relying on continuous positioning data, so as to realize the travel behavior characteristic analysis of any vehicle.

Disclosed herein is a failure detecting method for a steering angle sensor that detects a turning angle, turning direction and central position of a steering wheel of a vehicle. It is determined whether the steering angle sensor has failed using traveling behavior of the vehicle, and therefore, the failure of the steering angle sensor is effectively detected even when the steering angle center or the turning angle is incorrectly detected. Also, it is determined whether the steering angle sensor has failed using characteristic differences among pulse signals of the sensor, and therefore, three pulse signals output from the steering angle sensor are effectively detected although any of the pulse signals has failed, whereby the vehicle is stably traveled.

The present invention provides a vehicle state monitoring method and device, wherein the vehicle state monitoring method includes: collecting the vital sign data of the driver, and / or obtaining information on the current driving behavior state of the vehicle; and / or the driving behavior state meets a preset driving state condition, then a notification message about the current vehicle is sent. The vehicle state monitoring method and device of the embodiments of the present invention realize accurate and timely sending of vehicle-related notification messages when the driver is driving in an unsafe state during driving, thereby improving driving safety and stability.

The invention provides a vehicle travel trajectory identification method based on bayonet data, which adopts a spectral clustering method to realize automatic separation of travel chain from the time-space characteristic of bayonet data, thereby identifying the starting point, the end point and the passing point of a single travel trajectory. The method particualy includes generating a complete travel chain of the vehicle on the day; establishing a similarity matrix S; calculating an adjacency matrix W and a degree matrix D; constructing a Laplace matrix L and further determining a characteristic matrix F; using the DBSCASN clustering algorithm to perform cluster analysis, and obtaining the clustering results; determining the cluster partition result of trajectory points to generate sub-trajectories. This method takes into account both the time and space attributes of vehicles, and is more comprehensive than the method based on travel time to partition the trajectory. No preset threshold is needed, and the vehicle trajectory recognition is completely driven by the measured data, which improves the accuracy of trajectory recognition. The analysis results can be used in vehicle travel behavior recognition, travel characteristics analysis, traffic flow OD estimation and so on.

The invention discloses vehicle green travel navigation system design based on a smart phone, and particularly relates to a green navigation system and method with which in any time and space range, travel pollutant emission evaluation and prediction of a vehicle can be achieved. The system comprises a travel data collection part and a green travel navigation part. Through popularity and portability of the smart phone, in cooperation with a VSP pollutant emission model of the vehicle, travel information of a large number of users is obtained conveniently with low expense so as to construct a historical database, and the information is converted into visual and visible pollutant emission data through the green travel navigation part. Because the influence on the city pollutant emission level caused by the travel of the owner of the vehicle is visually known, the purpose that the traffic travel behaviors are improved, and the total pollutant emission amount of the motor vehicles is reduced can be achieved.

The invention relates to the technical field of travel purpose identification, in particular to a travel purpose identification method and device based on mobile phone signaling data. The travel purpose identification method comprises the following steps: extracting OD data, and dividing population types and occupational and residential places; obtaining and associating POI and base station positions with the located blocks respectively to form a position corresponding relation; preliminarily identifying the travel purpose based on the corresponding relationship between the location block of the travel destination base station and the POI type and the occupational and residential place corresponding to the plot, extracting characteristic parameters, clustering the remaining samples based on a K-means clustering algorithm, and finally obtaining the travel purpose; and correcting the travel destination result according to the population type. According to the embodiment, based on the mobile phone signaling data and the POI points, resident travel purposes are divided through a machine learning algorithm, and meanwhile, the result is corrected in combination with travel behavior characteristics of crowds, so that dependence and requirements on priori experience knowledge are reduced, and subjectivity brought by a single rule discrimination method can be avoided.