Multi-vehicle Bayesian Search for Multiple Lost Targets

• DOI: 10.1109/robot.2005.1570598
• Publication Date: 2006-01-18T23:42:54Z
• Authors: El-Mane Wong; F. Bourgault; T. Furukawa
• Keywords: Recursive Bayesian estimation

This paper presents a Bayesian approach to the problem of searching for multiple lost targets in dynamic environment by team autonomous sensor platforms. The probability density function (PDF) each individual target location is accurately maintained an independent instance general filter. utility search vehicles trajectories given sum `cumulative' detection target. A dual-objective switching also introduced direct towards mode nearest PDF when becomes too low region distinguish between trajectories. Simulation results both clustered and isolated demonstrate effectiveness proposed strategy targets.