A Controlled Interactive Multiple Model Filter for Combined Pedestrian Intention Recognition and Path Prediction

2015 ◽  
Author(s):  
Andreas T. Schulz ◽  
Rainer Stiefelhagen
Keyword(s):  
Multiple Model ◽  
Intention Recognition ◽  
Model Filter ◽  
Interactive Multiple Model ◽  
Path Prediction

scholarly journals Interactive Multiple Model Filter for Land-Mobile Satellite Communications at Ka-Band

IEEE Access ◽  
2017 ◽  
Vol 5 ◽  
pp. 15414-15427 ◽  
Author(s):  
Paulo Victor Rodrigues Ferreira ◽  
Randy Paffenroth ◽  
Alexander M. Wyglinski
Keyword(s):  
Satellite Communications ◽  
Multiple Model ◽  
Ka Band ◽  
Model Filter ◽  
Interactive Multiple Model ◽  
Mobile Satellite

Interactive Multiple Model Filter for Tracking a Pursuer With Proportional Navigation Guidance Law

2018 ◽  
Vol 140 (8) ◽  
Author(s):  
Runle Du ◽  
Xinguang Zou ◽  
Di Zhou ◽  
Jiaqi Liu
Keyword(s):  
Relative Velocity ◽  
Accurate Estimation ◽  
Time Varying ◽  
Multiple Model ◽  
Motion Model ◽  
Proportional Navigation ◽  
Guidance Law ◽  
Model Filter ◽  
Interactive Multiple Model ◽  
Proportional Navigation Guidance

This paper addresses a pursuer tracking problem where the pursuer's acceleration is given by a proportional navigation (PN) guidance law with a time-varying navigation ratio which varies with the relative range between the pursuer and its target. Based on a motion model that exactly describes the relative motion and the PN guidance law, a novel filter for tracking such a pursuer is designed using interactive multiple model (IMM) algorithm and unscented Kalman filtering (UKF) technique. This filter is able to accurately estimate the relative range, relative velocity, and the acceleration of pursuer even if the pursuer adopts a PN guidance law with time-varying navigation ratio. The proposed tracking method is evaluated in extensive Monte Carlo simulations. It is shown that accurate estimation results have been obtained, and the model probabilities in the IMM UKF filter are consistent with real situations.

Sign in / Sign up

Export Citation Format

Share Document