207 results about "Floating car data" patented technology

Participating vehicles and egress points communicate with each other according to an infrastructure mode. Participating vehicles communicate with other participating vehicles according to an ad-hoc mode. In an infrastructure mode packet transmitting method for a participating vehicle, beacon service table packets, vehicle service table packets, or packet bursts are created and transmitted. In an infrastructure mode packet receiving method for a participating vehicle, beacon service table packets, vehicle service table packets, packet bursts, or negative acknowledgement packets are received. In an infrastructure mode packet transmitting method for an egress point, an enhanced beacon packet or a negative acknowledgement packet is created and transmitted. In an infrastructure mode packet receiving method for an egress point, beacon service table packets, vehicle service table packets, or packets bursts are received. In an ad-hoc mode packet transmitting method for a participating vehicle, beacon service table packets, vehicle service table packets, packet bursts, or positive acknowledgement packets are created and transmitted. In an ad-hoc mode packet receiving method for a participating vehicle, beacon service table packets, vehicle service table packets, packet bursts, or positive acknowledgement packets are received.

The invention relates to heuristic path presuming method used to process large-scale floating car data. It includes the following steps: building road net data structure includes node, road section, and road chain, road net topological structure which is the connected relation of the each road chain according to road geography information data in city navigation electric map; processing floating car data by heuristic path presuming method; computing traveling path according to recorded time and position for each floating car. The invention uses road net connectivity topological structure and directed line segment heuristics caused by vehicle position point to search the most possible next traveling path, processes project matching for the selected road to effectively increase processing efficiency; in addition, it combines consecutive many GPS location data to presume, increases algorithm accuracy by vehicle path traveling continuity.

The invention relates to a map matching method of floating car data. An optimal matching path is guaranteed and an alternative path set of each global positioning system (GPS) point is minimized at the same time by means of multistandard dynamic planning technology, and therefore the matching efficiency is improved. In addition, aiming at the characteristics of map matching, a traditional shortest path analytical method is improved, and the matching speed is further improved by reducing a mark initialization process and a searching method from multiple points to a single point. Compared with other current methods, the map matching method of the floating car data is better in matching accuracy and matching performance, can achieve real-time map matching processing of low sample frequency floating car data in a city traffic network, and has business popularization and application prospects.

The invention provides an urban trunk road vehicle trajectory reconstruction method based on fixed-point detector and signal timing data fusion. The method comprises the following steps: 1) establishing a basis matrix: determining a road occupation condition matrix Point[n, distance, t] at the t moment according to a vehicle running matrix Car [vehicle, t] at the t time, and then, calculating vehicle most probable running status at the t+t0 moment according to the road occupation condition matrix Point[n, distance, t], and thus a Car [vehicle, t+t0<->] constant iterative process is generated; 2) carrying out vehicle generating; 3) carrying out vehicle behavior decision; and 4) judging whether trajectory reconstruction is finished. The method has the advantages of fusing fixed-point detector and signal timing data, not depending on high-quality floating car data and demanding less on traffic data acquisition environment. The method is suitable for multilane and coming/going vehicle traffic interference cases, is wide in applicability, is complete and comprehensive in reconstructed trajectory, and can realize the functions of environment assessment, signal control coordination, travel time estimation and jam state early warning and the like.

The invention provides a high-practicability, high-efficiency and low-cost on-line queuing length detection method based on floating vehicle data. The method comprises the following steps of: 1, matching the current position of a vehicle with an electronic map according to positioned information of a global positioning system (GPS); 2, counting parking times according to a preset time interval, a counting interval and a parking point; and 3, estimating a queuing length, and adding limits of error for correction.

The invention provides a travel time predicting method based on a network convergence and taking account of a traffic state time-space relation. The method obtains high-precision data by subjecting floating car data to refined processing. In particular, the process for subjecting the floating car data to refined processing ingeniously uses a convolutional neural network and a recurrent neural network which are mainly used in the field of artificial intelligence currently, fuses the networks, fully takes account of the time-space relation of a traffic state, deeply excavates a traffic state evolution rule, and ultimately achieves precise travel time prediction. Compared with various conventional travel time prediction models, the method fuses the time-space relations of the traffic state, fully excavates historical data characteristics, overcomes a difficulty that other methods cannot cope with the high time-varying characteristics and non-linear change characteristics of the traffic flow, does not need priori knowledge and parameter identification, and has good fault tolerance and robustness, high precision, and good stability.

In a traffic information system, the principal component analysis of the floating car data collected in the past is performed for each traffic area. From among the bases representing the traffic data collected on the road-links in the traffic area, the bases which have strong correlation to the road-links on which real-time traffic data were collected are selected. The weighting coefficients for the selected bases are calculated by projecting the real-time traffic data onto the selected bases. The traffic estimation data are calculated by linearly combining the selected bases with the obtained weighting coefficients as the coefficients of the respective bases. The calculated traffic estimation data are used for the interpolation of the road-links on which the real-time traffic data were not collected.

The invention provides a traffic jam point identification method based on the floating car technology, which comprises the following steps: regularly acquiring numbers, positions, speeds and time information of vehicles in a driving process by using a floating car, and transmitting the acquired data to a data centre; carrying out data filtering pre-processing on the speed information in the data by the data centre to acquire effective floating car data, and extracting all data sets with ultra low-speed driving; and automatically identifying the traffic jam points by the data centre through inspecting frequent ultra low-speed driving areas of multiple vehicles to realize dynamic updating of the traffic jam point information. The invention provides dynamic and correct traffic jam point information for various traffic information service systems, therefore, the cost of information acquisition channels is low, and the information is reliably updated in time, thus having significant meaning on the various traffic information service systems.

The invention discloses a method for predicting multi-period travel time distribution based on floating vehicle data. According to the method, on the basis of a traditional KNN algorithm and through the adoption of the learning ability of a Bayesian model, the robustness and accuracy of travel time predicting are improved; through the effective fusion of history data and real-time data, the travel time average and variance prediction of the continuous multiple time ranges are achieved, and the confidence interval of the travel time floating is built. Compared with a traditional KNN model, the travel time distribution information for half an hour can be predicted effectively in various complex city networks, the result is more reliable and precise.

The invention relates to an urban traffic-oriented traffic index calculation method. The calculation method is defined by flow and a vehicle speed, and is based on commonly used microwave data and floating car data in urban traffic; and historical or real-time traffic indexes of a road network can be determined by adopting methods such as data fusion, statistical modeling and mathematical modeling. The urban traffic-oriented traffic index calculation method has the characteristics of high accuracy of calculation result, high adaptivity, low cost, high feasibility, simplicity and convenience of calculation process, low time complexity and high operation speed.

The invention provides a method and a device for matching maps of floating car data and relates to the technical field of communication. A problem of less accuracy in the traditional GPS (Global Positioning System) is solved. The method can comprises the following steps: at least one track segment matching server is used for respectively receiving information of all track points which are transmitted by a track data server and contained in track segments bound in the corresponding track segment matching server; at least one track segment server is used for respectively matching map for the bound track segments according to the received information of the track points and acquiring a matching result of the track segments; and at least one track segment matching server is used for transmitting the matching result of the track segments to a global map matching server, so that the global map matching server is used for matching a global map according to the received matched results of all track segments. And the method and the device for matching maps of floating car data can be used for matching maps.

The invention discloses a traffic flow change trend extraction method based on floating car data and belongs to the technical field of intelligent transportation. The traffic flow change trend extraction method based on the floating car data comprises obtaining historical floating car data of at least three mouths of a road chain and classifying the data according to feature days; performing denoising and smoothing on the historical floating car data; dividing the historical floating car data of every feature days into at least one time frame according to the change trend of a morning peak and an evening peak; performing preliminary clustering on the classified historical floating car data through a K-means clustering method; further clustering the historical floating car data through a controlled interval method or a two-value method according to a coarse granularity expression of traffic information. The traffic flow change trend extraction method based on the floating car data combines the coarse granularity expression of the traffic information, further merges the traffic flow trends on the basis of a K-means clustering method, enables the traffic flow change trend to be more salient, thereby providing more crystal clear references for traffic flow prediction, route planning and induction, route planning and the like.

The invention discloses a floating car data processing method and a floating car data processing device, which relate to the field of traffic road condition information processing, and solve the problem of large error existing in the current average speed calculation methods. The floating car data processing method comprises the following steps: firstly, calculating the average speed of a floating car according to the effective GPS point data recorded by the floating car on a traveling road; then, calculating the average instantaneous speed of the floating car according to the instantaneous speeds included in all effective GPS point data on the traveling road; and finally, when the sum of the average speed and a preset speed is less than the average instantaneous speed, correcting the average speed by using the average instantaneous speed so as to obtain a corrected average speed. The method and the device disclosed by the invention have the advantages that: because the average speed is corrected by using the average instantaneous speed, the influence of special GPS point data on the average speed is reduced, thereby reducing the error of the average speed, and then improving the accuracy of calculation on the road condition information. The method and the device disclosed by the invention are used for dynamically acquiring traffic jam information in real time.

The invention discloses an expressway road condition identification and prediction method based on floating vehicle data, and the method comprises the following steps: carrying out the data preprocessing based on the original GPS track data, based on a speed calculation method for grid division results, calculating a travel vehicle speed by using an Euclidean distance between track points and travel time as an instantaneous speed of a vehicle at an updating point; based on the traffic flow parameters in the speed calculation grid, calculating the traffic flow parameter value in each unit by taking the space grid and the time grid as units; dimensionally reducing traffic state parameters based on principal component analysis, simiplifying the dimension of data and abandoning irrelevant features in the data; based on k-means traffic state clustering anylsis, identifying different traffic states; and constructing characteristics of different time scales, and establishing each traffic state quantity prediction model based on the long short-term memory neural network LSTM. The expressway traffic state can be accurately identified, and the evolution trend of the expressway traffic statecan be predicted.

The invention relates to a road state merging method based on floating car data (FCD) and an earth magnetism detector. The method sequentially comprises the following steps of: using a floating car for detecting floating point data of each road section of a road network, and obtaining sped information of each road section through processing; using the earth magnetism detector for detecting the occupied rate data of each road section detection point of the road network, and obtaining the density information of each road section through processing; and receiving the data transmitted back by thefloating car and the earth magnetism detector in real time by a background server, obtaining a speed-density regression relational expression through processing, merging the road section real-time density and the speed information into a data base of the background server, and updating the speed-density regression relational expression in real time. The floating car is used for detecting the roadnetwork car flow speed, the earth magnetism detector is used for detecting the road network vehicle flow density, the background server database converts the homogeneous traffic parameter data for carrying traffic state merging and extraction according to the speed-density regression relational expression attribute of each road section of the road network, and the base data is provided for a traffic information issuing system.

The invention provides a method and system for filling missing floating car data. The method includes the following steps of: data screening, map matching, gridding, interpolation region and time determination and missing data interpolation. According to the method and system of the invention, data can have integrity through interpolation. The method and system can be used for estimating traffic congestion degree, calculating road section traffic speed and releasing current traffic announcements so as to provide conveniences for the travel of people.

The invention discloses method and device for preprocessing floating car data and relates to the field of intelligent traffic. The invention solves the problem that the prior art can not distinguish whether the data generated when a floating car generates GPS data in a parking state is passive parking data or active parking data . The method comprises the following steps of: filtering error data in the original floating car data; restoring the floating car data after filtering the error data in a unit of a car; and distinguishing the passive parking data and the active parking data from the floating data generated by each car according to the floating car data information. The embodiment of the invention is mainly applied to an intelligent traffic processing system.

The invention provides an optimized road congestion state discovery method which comprises the following steps that 1, a floating car regularly collects information of car serial numbers, positions, speeds, time and car types in a running process, and sends the information to a data center; 2, the data center filters data, obtains a road section where the floating car is located according to matching of the position information with a GIS (Geographic Information System), and stores floating car data in a position, corresponding to the road section, in a driving road section database; 3, a road section information set of road sections is extracted from the driving road section database by the data center, a weighted average of the speeds is computed by combing the car types and the weight of the speeds, and abnormal speed information is obtained; and 4, the times of the abnormal speed information of each road section are counted within a preset time, and the road section is congested if the times reach the preset value. Therefore, identification of a traffic congestion spot is achieved. The road congestion state discovery method is more efficient, reliable and dynamic, and can improve the road section congestion state discovery accuracy.

The present invention discloses a crossing traffic state recognition method based on floating car data. By introducing the duration of the control period of a crossing traffic signal, the factor that the floating car data is affected by a crossing signal lamp is considered when a traffic recognition is carried out; fast match of a detected area is realized by utilizing a GIS grid, and the crossing traffic state is sensed realtimely. Therefore, the crossing traffic state recognition method based on floating car data can realtimely sensor the traffic state, with no delay, and the recognition accuracy is high.

The invention discloses an intermittent flow path section travel time estimation method. The floating car data and coil flow data exist on a target road section at the same time; the road section average travel time is estimated by fusing the floating car data and coil flow on the target road section; the floating car sample size divided by the coil flow is used as an index of the floating car sample coverage rate; and the estimated value of the road section average travel time is determined by taking the index as reference and combining the excavation of the history data in the same time interval. The method disclosed by the invention can effectively fuse the floating car data and the coil flow data, and the accuracy of the estimated value of the road section travel time is higher than the traditional accuracy of the estimated value of the floating car road section travel time and the estimated value of the coil road section travel time, thereby bringing important significance to the intelligent traffic service, the traffic management and the like. According to the invention, the method and technology are simple and easy to implement, the operation conditions are easy to meet, and the method is easy to popularize and apply in large and medium-size cities.

The invention provides a traffic information providing apparatus which provides traffic information based on user request output which has high accuracy higher accuracy compared with that of prior art. The traffic information providing apparatus comprises an information receiving portion, a data storing portion, a map matching portion, a data filtering portion, and a traffic information prediction portion. The information receiving portion is used for receiving floating car data and user request route data. The data storing portion is used for storing the floating car data, the user request route data, and map data. The map matching portion is used for matching the floating car data with the user request route data to determine locating point data corresponding to the floating car data. Based on the locating point data and the user request route data, the data filtering portion is used for filtering the floating car data stored in the data storing portion, to acquire the floating car data corresponding to the locating point data which is completely or partially consistent with user request routes. Based on the acquired floating car data, the traffic information prediction portion is used for computing traffic information of the user request routes. At one moment, for a same section in different user request routes, the computed traffic information of the section is different.

The invention discloses a map matching method based on conditional random fields and low-sampling-frequency floating car data. The map matching method includes the steps that according to the low-sampling-frequency floating car data, on the basis of a road network model, a candidate projection point set possibly matched with GPS observing points and the observing probability of the candidate projection points in the set are firstly calculated, and then a candidate route set of the adjacent GPS observing points and the transmission probability between every two adjacent candidate projection points are calculated; according to the candidate projection points and candidate routes, in a sliding window, based on a condition random field model, the optimal matched projection points of the observing points are selected through a front-and-back-direction recursive algorithm. According to the map matching method, under the condition of the low sampling frequency, the topological structure of a road network and related information between the GPS observing points are both considered so that the calculation complexity can be low, and map matching accuracy is improved.