| | |
Summary: INTEGRATED DISPLACEMENT TRACKING AND SOUND LOCALIZATION
Parham Aarabi, Qing Hua Wang, Masoud Yeganegi
The Artificial Perception Laboratory, University of Toronto
{parham,wangq,yeganeg}@ecf.utoronto.ca
ABSTRACT
An algorithm for the robust localization of a vehicle using
both displacement tracking and sound localization is pro
posed. The displacement tracking is performed by optical
encoders that enable the turn angle and the movement dis
tance of the vehicle to be estimated. The sound localization
utilizes a speaker mounted on the vehicle and an array of 24
microphones deployed in the environment. The two modal
ities are integrated by modeling the displacement tracking
uncertainty by a Gaussian Mixture Model (GMM) and com
bining it with the probability distribution obtained from the
sound localization system. It is shown that the proposed in
tegrated system results in an average localization error (at
best 11cm) that is better than either modality alone.
1. INTRODUCTION
The precise and robust localization of an autonomous vehi
|
