Distributed multi-camera synchronization for smart-intruder detection

Abstract

This work proposes algorithms to control the trajectory of a team of cameras for video surveillance. We consider a chain of cameras installed in an environment. These cameras are used to detect smart intruders, who are aware of the cameras' configuration at each time instant, and who schedule their motion to avoid detection, if possible. For this problem setup, we first obtain a lower bound on the worst-case and on the average detection time of a smart intruder. We then propose a team trajectory for the cameras, called the equal-waiting trajectory, with minimum worst-case detection time and constant-factor optimal average detection time. Additionally, we design and analyse a distributed algorithm to steer the cameras in finite time towards the equal-waiting trajectory. Finally, the performance and robustness of our distributed algorithm is further analyzed through a simulation study.