Dissipation-accuracy tradeoffs in autonomous control of smart active matter
read the original abstract
The study of motility control by smart agents offers a promising platform for systematically exploring the fundamental physical constraints underlying the functioning of bio-inspired micro-machines operating far from equilibrium. Here, we address the question of the energy cost required for a self-steering active agent to localise itself within a specific region of space or follow a pre-defined trajectory under the influence of fluctuations and external flows. Building on a stochastic thermodynamic formulation of the problem, we derive a generic relationship between dissipation and localisation accuracy, which reveals a fundamental dissipation-accuracy tradeoff constraining the agent's performance. In addition, we illustrate how our framework enables the derivation of optimal steering policies that achieve localisation at minimum energy expenditure.
This paper has not been read by Pith yet.
Forward citations
Cited by 1 Pith paper
-
Theory of collective learning in populations of adaptive agents
An effective reward function emerges that fully governs the evolution of policy distributions across the population, yielding closed equations for mean and variance under Gaussian assumptions.
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.