This project focuses on finding patrolling paths for a team of multiple autonomous robots that are persistently monitoring a large environment to detect events occurring at locations of the environment. These events can either be random or in cases, can be adversarial and actively try to avoid being detected. Applications of this work include monitoring forests for fire outbreaks, patrolling a nature reserve to observe animal behaviours or to protect wildlife, and patrolling an urban environment to guard against intruders. We design and analyze random patrolling paths based on Markov chains for the robots.
In persistent monitoring scenarios, locations in an environment need to be visited repeatedly by a team of robots. Since the duration of the events to be observed, or the rate of change of the properties to be monitored, can be different for different locations, each location will have a different latency constraint, which specifies the maximum time allowed between consecutive visits to that location. This project focuses on the problem of finding a set of paths that continually visit a set of locations while collectively satisfying the latency constraints on each location.