A Receding Horizon Approach for Simultaneous Active Learning and Control using Gaussian Processes

TL;DR

This paper introduces a receding horizon method combining active learning and control with Gaussian Processes to improve system modeling and real-time decision-making in dynamical systems.

Contribution

It presents a novel optimization framework using differential entropy for active learning within a receding horizon control scheme, solved efficiently via sequential convex programming.

Findings

Enhanced data quality for system modeling.

Real-time applicability demonstrated in autonomous racing simulations.

Significant improvement over traditional methods.

Abstract

This paper proposes a receding horizon active learning and control problem for dynamical systems in which Gaussian Processes (GPs) are utilized to model the system dynamics. The active learning objective in the optimization problem is presented by the exact conditional differential entropy of GP predictions at multiple steps ahead, which is equivalent to the log determinant of the GP posterior covariance matrix. The resulting non-convex and complex optimization problem is solved by the Sequential Convex Programming algorithm that exploits the first-order approximations of non-convex functions. Simulation results of an autonomous racing car example verify that using the proposed method can significantly improve data quality for model learning while solving time is highly promising for real-time applications.