Emerging Traffic Data Collection Practices Under Mixed Traffic Conditions: Challenges and Solutions

Abstract

AbstractConventional traffic car-following and lane-changing models are not valid for mixed traffic conditions, which consist of a variety of vehicle types maintaining a weak or no lane discipline. Therefore, studies related to vehicle dynamics, driver behavior, including the simultaneous lateral and longitudinal interactions are important to model these conditions and to achieve solutions to the ever-increasing issues of traffic congestion, delays, and crashes. This chapter confers on the novel traffic data collection methods for mixed traffic, extraction and development of accurate and comprehensive datasets, and their applications for both microscopic and macroscopic conditions. Data collection using instrumented vehicles, sensors, unmanned aerial vehicles, camera calibration, and other novel techniques, and data development of trajectories, vehicle interactions, driver behavior, travel time and traffic state information would be explored. The concepts and methodologies related to traffic data collection and extraction, discussed in this chapter, will be thus useful for the researchers in comprehending future challenges and developing possible traffic solutions.KeywordsTraffic data collectionTrajectory datasetNaturalistic drivingCamera calibrationSensors