Multi-ordered short-range mover prediction models for tracking and avoidance

Jamahl Overstreet, Farshad Khorrami

Research output: Contribution to journalConference article

Abstract

This paper introduces a framework and methods that can be used to predict the movements of intelligent moving bodies in the presence of perceived static and dynamic environmental stimulus, such as terrain and weather influences. These methods are especially important for Intelligent-Autonomous Mobile (I-AM) Systems, where they can improve upon contemporary methods for tracking and avoidance by allowing I-AM systems to act or react based on enhanced predictions. These methods can also complement Cooperative Behavior Control (CBC) strategies of distributed, multi-agent systems wherein cooperation can be in the form of prediction rather than direct communication. Probability spatial distributions for intelligent moving objects, with respect to First-Order, Second-Order, and Third-Order predictions, have been formulated. This is a novel method since most prediction approaches use Kalman Filters to estimate future states based solely on previously observed states. Most prediction models do not take into consideration mobility characteristics (e.g., Ackermann Steering), nor the probable decision making capabilities of intelligent entities. By adding higher levels of fidelity to prediction models, more accurate and precise object tracking, and obstacle avoidance and/or engagement can be accomplished with already proven techniques.

Original languageEnglish (US)
Article number6425831
Pages (from-to)1770-1775
Number of pages6
JournalProceedings of the IEEE Conference on Decision and Control
DOIs
StatePublished - 2012
Event51st IEEE Conference on Decision and Control, CDC 2012 - Maui, HI, United States
Duration: Dec 10 2012Dec 13 2012

ASJC Scopus subject areas

  • Control and Systems Engineering
  • Modeling and Simulation
  • Control and Optimization

Fingerprint Dive into the research topics of 'Multi-ordered short-range mover prediction models for tracking and avoidance'. Together they form a unique fingerprint.

  • Cite this