Behavior Prediction Based on Obstacle Motion Patterns in Dynamically Changing Environments

  • Authors:

  • Zhuo Chen;D. C. K. Ngai;N. H. C. Yung

  • Affiliations:

  • -;-;-

  • Venue:
  • WI-IAT '08 Proceedings of the 2008 IEEE/WIC/ACM International Conference on Web Intelligence and Intelligent Agent Technology - Volume 02
  • Year:
  • 2008

Quantified Score

Hi-index 0.00

Visualization

Abstract

This paper proposes a behavior prediction method for navigation application in dynamically changing environments, which predicts obstacle behaviors based on learned Obstacle Motion Patterns (OMP) from observed obstacle motion trajectories. A multi-level prediction model is then proposed that predicts long-term or short-term obstacle behaviors. Simulation results show that it works well in a complex environment and the prediction is consistent with actual behaviors.