Unscented Kalman Filtering for Vehicle Manoeuvre Recognition

The detection of different maneuvers performed by vehicles on highways is an important prerequisite for Collision Avoidance and Collision Mitigation systems. This paper describes an algorithm which is able to distinguish between lane keeping and lane changing maneuvers even in curves. The approach is based on an Interacting Multiple Model (IMM) filter which also takes into account information about the course of the lane. By modeling the driving behavior of a vehicle on highways as a hidden Markov process, the maneuver probabilities can be derived from the IMM filter results. As this problem is highly nonlinear, Unscented Kalman Filtering is used in order to avoid analytical linearization. The system is evaluated using a microscopic traffic simulator.

Language

  • English

Media Info

  • Media Type: CD-ROM
  • Features: Figures; References;
  • Pagination: 10p
  • Monograph Title: ITS Connections: Saving Time. Saving Lives

Subject/Index Terms

Filing Info

  • Accession Number: 01145155
  • Record Type: Publication
  • Files: TRIS
  • Created Date: Nov 25 2009 11:54AM