Jinjun Tang

Lane-changes prediction based on adaptive fuzzy neural network

Abstract Lane changing maneuver is one of the most important driving behaviors. Unreasonable lane changes can cause serious collisions and consequent traffic delays. High precision prediction of lane changing intent is helpful for improving driving safety. In this study, by fusing information from vehicle sensors, a lane changing predictor based on Adaptive Fuzzy Neural Network (AFFN) is proposed to predict steering angles. The prediction model includes two parts: fuzzy neural network based on Takagi–Sugeno fuzzy inference, in which an improved Least Squares Estimator (LSE) is adopt to optimize parameters; adaptive learning algorithm to update membership functions and rule base. Experiments are conducted in the driving simulator under scenarios with different speed levels of lead vehicle: 60 km/h, 80 km/h and 100 km/h. Prediction results show that the proposed method is able to accurately follow steering angle patterns. Furthermore, comparison of prediction performance with several machine learning methods further verifies the learning ability of the AFNN. Finally, a sensibility analysis indicates heading angles and acceleration of vehicle are also important factors for predicting lane changing behavior.

Travel Time Estimation Using Freeway Point Detector Data Based on Evolving Fuzzy Neural Inference System

Travel time is an important measurement used to evaluate the extent of congestion within road networks. This paper presents a new method to estimate the travel time based on an evolving fuzzy neural inference system. The input variables in the system are traffic flow data (volume, occupancy, and speed) collected from loop detectors located at points both upstream and downstream of a given link, and the output variable is the link travel time. A first order Takagi-Sugeno fuzzy rule set is used to complete the inference. For training the evolving fuzzy neural network (EFNN), two learning processes are proposed: (1) a K-means method is employed to partition input samples into different clusters, and a Gaussian fuzzy membership function is designed for each cluster to measure the membership degree of samples to the cluster centers. As the number of input samples increases, the cluster centers are modified and membership functions are also updated; (2) a weighted recursive least squares estimator is used to optimize the parameters of the linear functions in the Takagi-Sugeno type fuzzy rules. Testing datasets consisting of actual and simulated data are used to test the proposed method. Three common criteria including mean absolute error (MAE), root mean square error (RMSE), and mean absolute relative error (MARE) are utilized to evaluate the estimation performance. Estimation results demonstrate the accuracy and effectiveness of the EFNN method through comparison with existing methods including: multiple linear regression (MLR), instantaneous model (IM), linear model (LM), neural network (NN), and cumulative plots (CP).

Mixture modeling of freeway speed and headway data using multivariate skew-t distributions

ABSTRACT The knowledge of vehicle speed and headway distributions is very useful for developing microscopic traffic simulation models. Traditionally, speed and headway distributions are often not studied jointly and some microscopic traffic simulation models consider vehicle speeds and arrival times as independent inputs to the traffic simulation process. However, the traditional approaches ignore the possible correlation between freeway vehicle speed and headway. Recently, a Farlie–Gumbel–Morgenstern (FGM) approach was used to construct bivariate distributions to describe the characteristics of speed and headway. The FGM approach only allows a weak statistical dependency and lacks the ability to consider the dynamic correlation structure for speed and headway data collected under different traffic conditions. The objective of this study is to explore the applicability of the finite mixtures of multivariate skew-t distributions to capture the heterogeneity in speed and headway data. The proposed bivariate mixture modeling approach is applied to the 24-hour traffic data collected on IH-35 in Austin, Texas. The results of this study show that the bivariate skew-t mixture model can provide an excellent fit to the multimodal speed and headway distribution. Moreover, the mixture modeling approach can naturally accommodate the varying correlation coefficient by assigning different covariance matrixes for each component in the finite mixture model. The findings in this study can also improve car-following models for simulation purposes.

Short-Term Speed Prediction Using Remote Microwave Sensor Data: Machine Learning versus Statistical Model

Recently, a number of short-term speed prediction approaches have been developed, in which most algorithms are based on machine learning and statistical theory. This paper examined the multistep ahead prediction performance of eight different models using the 2-minute travel speed data collected from three Remote Traffic Microwave Sensors located on a southbound segment of 4th ring road in Beijing City. Specifically, we consider five machine learning methods: Back Propagation Neural Network (BPNN), nonlinear autoregressive model with exogenous inputs neural network (NARXNN), support vector machine with radial basis function as kernel function (SVM-RBF), Support Vector Machine with Linear Function (SVM-LIN), and Multilinear Regression (MLR) as candidate. Three statistical models are also selected: Autoregressive Integrated Moving Average (ARIMA), Vector Autoregression (VAR), and Space-Time (ST) model. From the prediction results, we find the following meaningful results: () the prediction accuracy of speed deteriorates as the prediction time steps increase for all models; () the BPNN, NARXNN, and SVM-RBF can clearly outperform two traditional statistical models: ARIMA and VAR; () the prediction performance of ANN is superior to that of SVM and MLR; () as time step increases, the ST model can consistently provide the lowest MAE comparing with ARIMA and VAR.

Hybrid Prediction Approach Based on Weekly Similarities of Traffic Flow for Different Temporal Scales

Traffic flow prediction is considered a key technology of intelligent transportation systems. This paper presents a hybrid model that combines double exponential smoothing (DES) and a support vector machine (SVM) to predict traffic flow patterns on the basis of weekly similarities in traffic flow. First, in the hybrid model, DES is applied to predict the future data, and its smoothing parameters are determined by the Levenberg–Marquardt algorithm. Second, the SVM is employed to estimate the residual series between the prediction results by the DES model and actual measured data. In the SVM model, the cross-correlation rule is used to optimize its parameters. Third, a case study to test the proposed model with the data at different temporal scales is presented. Furthermore, data-smoothing strategies, including difference and ratio schemes based on weekly similarities, are applied as data processes before prediction. The proposed hybrid model along with the processing scheme demonstrates superiority in prediction accuracy compared with autoregressive integrated moving average, DES, and DES-SVM models.

Traffic flow prediction based on hybrid model using double exponential smoothing and support vector machine

This study develops a hybrid model that combines double exponential smoothing (DES) and support vector machine (SVM) to implement a traffic flow predictor. In the hybrid model, DES is used firstly to predict the future data, and the smoothing parameters of the DES are determined by Levenberg-Marquardt algorithm. Then, SVM is employed to fit the residual series between the predicting results of DES model and actual measured data for its powerful no-linear fitting ability. Finally, a practical application is used to testify the proposed model. In the application, data smoothing and wavelet de-noising technology are applied as data pre-treatment before prediction. In addition, the data smoothing contains difference and ratio smoothing strategy. It is demonstrated the superiority of the new hybrid model and the effectiveness of data pre-treatment through the comparison between the prediction results of DES, autoregressive integrated moving average (ARIMA) and DES-SVM model.

Google Earth elevation data extraction and accuracy assessment for transportation applications

Roadway elevation data is critical for a variety of transportation analyses. However, it has been challenging to obtain such data and most roadway GIS databases do not have them. This paper intends to address this need by proposing a method to extract roadway elevation data from Google Earth (GE) for transportation applications. A comprehensive accuracy assessment of the GE-extracted elevation data is conducted for the area of conterminous USA. The GE elevation data was compared with the ground truth data from nationwide GPS benchmarks and roadway monuments from six states in the conterminous USA. This study also compares the GE elevation data with the elevation raster data from the U.S. Geological Survey National Elevation Dataset (USGS NED), which is a widely used data source for extracting roadway elevation. Mean absolute error (MAE) and root mean squared error (RMSE) are used to assess the accuracy and the test results show MAE, RMSE and standard deviation of GE roadway elevation error are 1.32 meters, 2.27 meters and 2.27 meters, respectively. Finally, the proposed extraction method was implemented and validated for the following three scenarios: (1) extracting roadway elevation differentiating by directions, (2) multi-layered roadway recognition in freeway segment and (3) slope segmentation and grade calculation in freeway segment. The methodology validation results indicate that the proposed extraction method can locate the extracting route accurately, recognize multi-layered roadway section, and segment the extracted route by grade automatically. Overall, it is found that the high accuracy elevation data available from GE provide a reliable data source for various transportation applications.

An adaptive map-matching algorithm based on hierarchical fuzzy system from vehicular GPS data

An improved hierarchical fuzzy inference method based on C-measure map-matching algorithm is proposed in this paper, in which the C-measure represents the certainty or probability of the vehicle traveling on the actual road. A strategy is firstly introduced to use historical positioning information to employ curve-curve matching between vehicle trajectories and shapes of candidate roads. It improves matching performance by overcoming the disadvantage of traditional map-matching algorithm only considering current information. An average historical distance is used to measure similarity between vehicle trajectories and road shape. The input of system includes three variables: distance between position point and candidate roads, angle between driving heading and road direction, and average distance. As the number of fuzzy rules will increase exponentially when adding average distance as a variable, a hierarchical fuzzy inference system is then applied to reduce fuzzy rules and improve the calculation efficiency. Additionally, a learning process is updated to support the algorithm. Finally, a case study contains four different routes in Beijing city is used to validate the effectiveness and superiority of the proposed method.

Enhancing Traffic Incident Detection by Using Spatial Point Pattern Analysis on Social Media

Expedient incident detection and understanding are important in traffic management and control. Social media as important information venues have immense value for increasing an awareness of traffic incidents. In this paper, an attempt is made to assess the potential of using harvested social media for traffic incident detection. Twitter in Seattle, Washington, was chosen as a representative sample environment for this work. A hybrid mechanism based on latent Dirichlet allocation and document clustering was proposed to model incident-level semantic information, while spatial point pattern analysis was applied to explore the spatial patterns and to assess the spatial dependence between incident-topic tweets and traffic incidents. A global Monte Carlo K-test indicated that the incident-topic tweets were significantly clustered at different scales up to 600 m. The nearest neighbor clutter removal method was used to separate feature tweet points from clutter; then a density-based algorithm successfully detected the clusters of tweets posted spatially close to traffic incidents. In multivariate spatial point pattern analysis, K-cross functions were investigated with Monte Carlo simulation to characterize and model the spatial dependence, and a positive spatial correlation was inferred between incident-topic tweets and traffic incidents up to 800 m. Finally, the tweet intensity as a function of distance from the nearest traffic incident was estimated, and a log-linear model was summarized. The experiments supported the notion that social media feeds acted as sensors, which allowed enhancing awareness of traffic incidents and their potential disturbances.

On Missing Traffic Data Imputation Based on Fuzzy C-Means Method by Considering Spatial–Temporal Correlation

The lack of some traffic flow data seriously affects the quality of data collection and analysis in the traffic system. Completing the missing data is one of the most important steps in achieving the functions of intelligent transportation systems. In this paper an approach based on fuzzy C-means (FCM) imputes missing traffic volume data in loop detectors. With spatial–temporal correlation between detectors, the conventional vector-based data structure is first transformed into a matrix-based data pattern. Then, the genetic algorithm is applied to optimize the parameters of cluster size and weighting factor in the FCM model. Finally, the actual traffic flow volume collected at different locations is designed as a testing data set, and two indicators including root mean square error and relative accuracy are used to evaluate the imputation performance of the proposed method by comparison with some conventional methods (multiple linear regression, autoregressive integrated moving average model, and average historical method) by missing ratio. The applications in four scenarios demonstrate that the FCM-based imputation method outperforms conventional methods.