A particle filter for vehicle tracking with lane level accuracy under GNSS-denied environments

Vehicle positioning is of significant importance for intelligent transportation applications. However, accurate positioning (e.g., with lane level accuracy) is very difficult to obtain due to lack of measurements with high confidence, especially in an environment without full access to a global navigation satellite system (GNSS). In this paper, a novel information fusion algorithm based on a particle filter (PF) is proposed to achieve lane level tracking accuracy under a GNSS-denied environment. We consider the use of both large scale and small scale signal measurements for positioning. Time-of-arrival (TOA) measurements using the radio frequency signals from known transmitters or roadside units, and acceleration or gyroscope measurements from an inertial measurement unit (IMU) allow us to form a coarse estimate of the vehicle position using an extended Kalman filter (EKF). Subsequently, small scale measurements such as lane-change detection, radar ranging from the guardrails and information from a high resolution digital map are incorporated to refine the position estimates using a PF. A probabilistic model is introduced to characterize the lane changing behaviors and a multi-hypothesis model is formulated for the radar range measurements to robustly weight the particles and refine the tracking results. The performance of the proposed tracking framework is verified by simulations using a four-lane highway.