Learning passive policies with virtual energy tanks in robotics

TL;DR

This paper combines passivity-based control with reinforcement learning using virtual energy tanks to create adaptable, energy-aware robotic control policies that maintain stability and improve performance.

Contribution

It introduces a novel framework that integrates RL with energy tank-based passivity control, addressing limitations of each method individually.

Findings

Simulations validate the effectiveness of the combined approach.

The framework enables energy-aware control policies with stability guarantees.

Potential for new research directions in energy-efficient robotics.

Abstract

Within a robotic context, we merge the techniques of passivity-based control (PBC) and reinforcement learning (RL) with the goal of eliminating some of their reciprocal weaknesses, as well as inducing novel promising features in the resulting framework. We frame our contribution in a scenario where PBC is implemented by means of virtual energy tanks, a control technique developed to achieve closed-loop passivity for any arbitrary control input. Albeit the latter result is heavily used, we discuss why its practical application at its current stage remains rather limited, which makes contact with the highly debated claim that passivity-based techniques are associated with a loss of performance. The use of RL allows us to learn a control policy that can be passivized using the energy tank architecture, combining the versatility of learning approaches and the system theoretic properties which can be inferred due to the energy tanks. Simulations show the validity of the approach, as well as novel interesting research directions in energy-aware robotics.