Distributed Hierarchical Control for State Estimation With Robotic Sensor Networks

摘要: This paper addresses active state estimation with a team of robotic sensors. The states to be estimated are represented by spatially distributed, uncorrelated, stationary vectors. Given prior belief on the geographic locations states, we cluster in moderately sized groups and propose new hierarchical Dynamic Programming (DP) framework compute optimal sensing policies for each that mitigates computational cost planning combined space. Then, develop decentralized assignment algorithm dynamically allocates clusters robots based pre-computed at cluster. integrated distributed is level but also scales very well large numbers robot We demonstrate efficiency proposed method both simulations real-world experiments using stereoscopic vision