Honghui Dong

Some practical vehicle speed estimation methods by a single traffic magnetic sensor

This paper proposes three practical vehicle speed estimation methods by a single multifunction magnetic sensor. Compared with traditional methods, this algorithm is simple and convenient to be realized. The multifunction magnetic sensor is described and introduced in this work. Next, a vehicle detection algorithm with a linear time complexity is put forward. Through setting two windows and scanning the vehicle waveform, we obtain the points of vehicle approaching and vehicle leaving, which are the bases for vehicle count, headway time, time occupancy, stopping time, detection of vehicle stopping and presence. The detailed detection methods of vehicle stopping and presence are described. We next present some speed estimation methods in detail. According to three speed estimation methods of Vehicle Length based (VLB), Time Difference based (TDB) and Mean Value based (MVB) to get different reference speeds when a vehicle passes over the sensor, we then analyze the applicability of the three methods. At last, we test the speed estimation methods by adoption of field data. Through the comparison with the real speed of 45 vehicles, it shows that the mean absolute errors (MAE) of VLB, TDB and MVB methods are respectively 4.12km/h, 5.90km/h and 4.05km/h and the mean speed errors of the three methods are all less than 1km/h. These errors are suitable for traffic engineering.

Road Traffic Flow Prediction with a Time-Oriented ARIMA Model

The prediction of the traffic flow can give the people important traveling information. In this paper, the traffic flow prediction problem is studied. An ARIMA model is proposed for the traffic flow prediction. The ARIMA model is trained according to the different period traffic data. Based on the different period data training, the ARIMA model is refined more accuracy. The experiments show that the ARIMA model trained by the time-oriented data can reach a better result than the non time-oriented data trained model.

Traffic Semantic Analysis Based on Mobile Phone Base Station Data

As traffic sensors gradually increase, traffic managers obtain more and more detection data, and the data volume has jumped to the Big Data magnitude. Recently, the cell detail record data are used as an emerging traffic detection data source. Using mobile phone as a probe, its detection data is able to well reflect users travel behavior. Meanwhile, cell phone base stations can be treated as fixed sensor and used to detect people flows in the base station area, reflect the distribution of traffic source, and provide supports for the division of the commuting traffic zone. In this article, the traffic semantic framework is proposed. We analyze the data of cell detail record data in Beijing, and extract four features of base stations: real-time user stock, inflow, outflow and increments, to tag the traffic semantic attribute of the base stations.

A vehicle re-identification algorithm based on multi-sensor correlation

Magnetic sensors can be applied in vehicle recognition. Most of the existing vehicle recognition algorithms use one sensor node to measure a vehicle‖s signature. However, vehicle speed variation and environmental disturbances usually cause errors during such a process. In this paper we propose a method using multiple sensor nodes to accomplish vehicle recognition. Based on the matching result of one vehicle‖s signature obtained by different nodes, this method determines vehicle status and corrects signature segmentation. The co-relationship between signatures is also obtained, and the time offset is corrected by such a co-relationship. The corrected signatures are fused via maximum likelihood estimation, so as to obtain more accurate vehicle signatures. Examples show that the proposed algorithm can provide input parameters with higher accuracy. It improves the average accuracy of vehicle recognition from 94.0% to 96.1%, and especially the bus recognition accuracy from 77.6% to 92.8%.

Real-time road traffic state prediction based on ARIMA and Kalman filter

The realization of road traffic prediction not only provides real-time and effective information for travelers, but also helps them select the optimal route to reduce travel time. Road traffic prediction offers traffic guidance for travelers and relieves traffic jams. In this paper, a real-time road traffic state prediction based on autoregressive integrated moving average (ARIMA) and the Kalman filter is proposed. First, an ARIMA model of road traffic data in a time series is built on the basis of historical road traffic data. Second, this ARIMA model is combined with the Kalman filter to construct a road traffic state prediction algorithm, which can acquire the state, measurement, and updating equations of the Kalman filter. Third, the optimal parameters of the algorithm are discussed on the basis of historical road traffic data. Finally, four road segments in Beijing are adopted for case studies. Experimental results show that the real-time road traffic state prediction based on ARIMA and the Kalman filter is feasible and can achieve high accuracy.

Long-term forecasting oriented to urban expressway traffic situation

Long-term traffic forecasting has become a basic and critical work in the research on road traffic congestion. It plays an important role in alleviating road traffic congestion and improving traffic management quality. According to the problem that long-term traffic forecasting is short of systematic and effective methods, a long-term traffic situation forecasting model is proposed in this article based on functional nonparametric regression. In the functional nonparametric regression framework, autocorrelation analysis (ACF) is introduced to analyze the autocorrelation coefficient of traffic flow for selecting the state vector, and the functional principal component analysis is also used as distance function for computing proximities between different traffic flow time series. The experiments based on the traffic flow data in Beijing expressway prove that the functional nonparametric regression model outperforms forecast methods in accuracy and effectiveness.

Study on spacing optimization for traffic flow detector

This paper discusses the influence factors of detector spacing optimization. Through the corresponding mathematical conversion, it gets the model parameters which influence the result of detector spacing optimization. IDF (Information Degree Function) is proposed to describe the spatial distribution characteristics of traffic information. Then, this paper gives the calibration method of the model parameters. After a general study on all parameters, this paper proposes the MIVM (Maximal Integrated Value Model). And the SPA (Shortest Path Algorithm) is used to solve the problem. Through the example of the Second Ring Road, in Beijing, the model parameters are calibrated on the field data. According to the result of detector spacing optimization, this paper obtains the reasonable density which is fit for Beijing expressway, and provides the basis for the practical application. The MIVM is fit for other cities, too.

Chinese Prosodic Word Prediction Using the Conditional Random Fields

As the basic prosodic unit, the prosodic word play an important role for the naturalness and the intelligibility for the Chinese TTS system. Although many research work have been on this research direction, the precision of the prosodic word prediction is still not satisfying. In this paper, Conditional Random Fields is introduced to model the prosodic predicting process. In this model, more efficient features can be fused together. Compared with the ME model, the CRF model can describe the interacting relations between the neighboring prosodic words. The experiment results show that this Conditional Random Fields Model is competent for the prosodic word prediction task. The f-score of the prosodic word boundary prediction reaches 96.81%.

A Perceptual Experiment Analysis on Level of Service in the Road Traffic

Level of Service (LOS) on road traffic can be considered as the quality which people will feel from the road conditions and traffic state, and can be described as one kind of perception of the drivers and passengers on the traffic flow. Therefore, the road user is the best subject to evaluate LOS and can give a criteria of LOS. Based on this idea, a perceptual experiment is designed in this paper. Different videos are made for different traffic conditions. And different kinds of people score and give the level of service for the traffic conditions by the videos. This experiment result shows that the perceptions of different people are similar. At least 70% of people can reach the same evaluation result. It is also discovered that the speed, flow, and occupancy have a special relation with LOS. And the speed-flow graph can be divided into different regions by the different levels of service. In other words, LOS can be determined by the speed and flow. This result can provide an objective measure of LOS, which can be used as a criteria to evaluate LOS.

Parameter Calibration Method of Microscopic Traffic Flow Simulation Models based on Orthogonal Genetic Algorithm.

Traffic microscopic traffic simulation models have become extensively used in both transportation operations and management analyses, which are very useful in reflecting the dynamic nature of transportation system in a stochastic manner. As far as the microscopic traffic flow simulation users are concerned, the one of the major concerns would be the appropriate calibration of the simulation models. In this paper a parameter calibration method of microscopic traffic flow simulation models based on orthogonal genetic algorithm is presented. In order to improve the capacity of locating a possible solution in solution space, the proposed method incorporates the orthogonal experimental design method into the genetic algorithm. The proposed method is applied to an arterial section of Ronghua Road in Beijing. Through comparing with the parameter calibration method based on genetic algorithm, the advantage of the proposed method is shown. Keywords-Microscopic traffic flow simulation model; Parameter calibration; Orthogonal genetic algorithm; VISSIM