A Dynamic Weighted Area Assignment Based on a Particle Filter for Active Cooperative Perception

摘要

This article addresses an Active cooperative perception problem for networked robots systems. Given a team of networked robots, the goal is finding a target using their inherent uncertain sensor data. The article proposes a particle filter to model the probability distribution of the position of the target, which is updated using detection measurements from all robots. Then, an information-theoretic approach based on the RRT* algorithm is used to determine the optimal robots trajectories that maximize the information gain while surveying the map. Finally, a dynamic area weighted allocation approach based on particle distribution and coordination variables is proposed to coordinate the networked robots in order to cooperate efficiently in this active perception problem. Simulated and real experimental results are provided to analyze, evaluate and validate the proposed approach.