Cooperative Bayesian Search, Tracking, Localization and Mapping with Belief Fusion

Tomonari Furukawa

Department of Mechanical Engineering

Virginia Polytechnic Institute and State University

E-mail: tomonari at

A Unified Strategy for Complex Missions of Unmanned Vehicles

The talk will present cooperative Bayesian search, tracking, localization and mapping (STLAM) with belief fusion developed by the speaker as a result of his research into cooperative robotics over 10 years, which allows multiple autonomous vehicles to cooperatively search for, track and localize targets while self-localizing and constructing a map of environments in a unified theoretical framework. The proposed cooperative STLAM effectively uses the extended Kalman filter (EKF) as the Gaussian estimator for tracking and localization of observable moving targets and a grid-based method (GM) proposed by the speaker as the non-Gaussian recursive Bayesian estimator for search of unobservable moving targets and also as the so-called global scan-to-map matching simultaneous localization and mapping (SLAM) technique for self-localization and mapping of environments including static targets. The proposed belief fusion allows autonomous vehicles to communicate target beliefs rather than observation likelihoods. The formulations within the unified EKF/GM framework enable seamless estimation and control by cooperative autonomous vehicles without any complication in estimation and control strategies. The belief fusion further achieves both accuracy and efficiency in communication and data fusion. The usage of a graphics processing unit (GPU) for the GM in addition to the central processing unit (CPU) for the EKF and the modeling of computer performance further enable autonomous cooperation in a real-time environment without loss of information. The talk will also cover the successful implementation and demonstration of the proposed work in both real environments and the so-called platform- and the hardware-in-the-loop simulator (PHILS), which allows performance evaluation of cooperative autonomous vehicles in truly real-time environments.


Tomonari Furukawa is Professor of Mechanical Engineering at Virginia Polytechnic Institute and State University, known as Virginia Tech, in USA.  He received the B.Eng. in Mechanical Engineering from Waseda University, Japan, in 1990, the M.Eng. (Research) in Mechatronic Engineering from University of Sydney, Australia, in 1993 and Ph.D in Quantum Engineering and Systems Science from University of Tokyo, Japan, in 1996.  He was an Assistant Professor (1995-1997) and Lecturer (1997-2000) at the University of Tokyo, U2000 Research Fellow (2000-2002) at the Australian Centre for Field Robotics at University of Sydney, and Lecturer (2002-2004) and Senior Lecturer (2004-2008) at University of New South Wales (UNSW), Australia before joining Virginia Tech.

He has extensive research experiences in the field of robotics and computational mechanics.  He is currently a Director of Computational Multiphysics Systems Laboratory, while being also an Acting Director of JOUSTER (Joint Unmanned Systems Test, Experimentation and Research) test site affiliated with Virginia Tech.  He has published over 250 journal and refereed conference papers, served as Editorial Boards for five international journals and an international organizing committee member for over 10 international conferences and won a number of career research awards and paper awards.