Optimizing Mobile Sensor Placement for Distributed Ojbect Tracking
Mobile sensor networks have a variety of applications in security and surveillance. One of the most important considerations for such a mobile sensor network is the problem of how to distribute the sensors to maximize coverage of the space to be monitored and detection of targets. Here we compare several strategies for positioning of mobile sensors, including: stationary sensor placement, sensors that rotate in place, mobile sensors moving between predefined waypoints, and an active search. The active search is a strategy where the mobile sensors repeatedly move to the edge of the recently viewed space in order to view areas which have either not yet been observed, or which have not been observed in a certain amount of time. Data from previous target observations is used to learn a model of target motion, which can be used to predict target movements and areas that targets may be likely to appear. Tests are performed in simulation in order to evaluate the effectiveness of the different sensor positioning strategies and sensor models.