Dynamic Coverage Control in a Time-Varying Environment Using Bayesian Prediction.

Area of Science:

• Robotics
• Control Systems
• Statistical Inference

Background:

• Coverage control is crucial for applications like environmental monitoring and search and rescue.
• Accurate agent deployment relies on understanding the target area's density function, which is often unknown.
• Existing methods struggle with dynamic environments and measurement noise.

Purpose of the Study:

• To develop a robust dynamic coverage control strategy for agents operating in environments with unknown density functions.
• To enhance agent deployment by accurately estimating time-varying density using Bayesian methods.
• To ensure the stability and near-optimality of the coverage network.

Main Methods:

• Utilized Bayesian prediction approaches for estimating the unknown density function.
• Developed a novel coverage-control-customized algorithm to acquire Bayesian parameters.
• Implemented a discrete control scheme to manage agents with a stochastic cost function derived from Bayesian estimation.
• Analyzed the mean-square stability of the proposed coverage system.

Main Results:

• Successfully estimated time-varying density functions, considering measurement noise.
• Achieved near-optimal agent deployment through the discrete control scheme.
• Demonstrated the mean-square stability of the coverage system.
• Validated the proposed methods via numerical simulations.

Conclusions:

• The proposed Bayesian-based spatial estimation and discrete control scheme effectively address the dynamic coverage control problem.
• The approach provides a stable and near-optimal solution for agent deployment in environments with unknown and time-varying densities.
• Numerical simulations confirm the practical efficacy of the developed algorithms.