OPTIMAL SECURITY PATROLLING BY MULTIPLE VISION-BASED AUTONOMOUS VEHICLES WITH OMNI-MONITORING FROM THE CEILING

Abstract

A multiple vision-based vehicle system for security patrolling in an indoor environment is proposed. Multiple autonomous vehicles controllable by wireless communication and equipped with cameras, as well as top-view omni-cameras with fish-eye lenses fixed on the ceiling are used as a test bed. First, a method is proposed for collecting navigation environment information, including vehicle turning points and object monitoring points, so that the vehicles can navigate to visit monitored objects without collisions with walls. Next, a point correspondence technique integrated with an image interpolation method is proposed for camera calibration by which the top-view cameras can be utilized to learn the information that enables vehicles to perform the security monitoring task, to locate the vehicles, and to monitor vehicle activities in the navigation phase. An optimal randomized and load-balanced path planning method is proposed as well, which requires shorter time to accomplish object monitoring in one session and provides higher degrees of patrolling security. Good experimental results show feasibility of the proposed methods for the application of multiple-vehicle security patrolling.