A decentralized algorithm for control of autonomous agents coupled by feasibility constraints

TL;DR

This paper introduces a decentralized control algorithm for multiple autonomous agents with feasibility constraints, enabling parallel computation and coordination, demonstrated through cooperative driving simulations.

Contribution

It presents a novel decentralized control method that divides the problem into sub-problems and uses derivative-based approximations for system evolution, enhancing scalability.

Findings

Effective in cooperative driving scenarios

Decouples computations via communication scheme

Parallelizes control computations

Abstract

In this paper a decentralized control algorithm for systems composed of $N$ dynamically decoupled agents, coupled by feasibility constraints, is presented. The control problem is divided into $N$ optimal control sub-problems and a communication scheme is proposed to decouple computations. The derivative of the solution of each sub-problem is used to approximate the evolution of the system allowing the algorithm to decentralize and parallelize computations. The effectiveness of the proposed algorithm is shown through simulations in a cooperative driving scenario.