๐ค AI Summary
This work addresses the challenge of low-cost validation for distributed multi-robot control algorithms by proposing an integrated simulation-to-experiment framework based on the Single Program Multiple Data (SPMD) paradigm. The framework executes multiple instances of the same algorithm in parallel on a single machine, enabling coordination through local state updates and neighbor communication. It supports seamless migration across multiple fidelity levelsโfrom point-mass models and high-fidelity quadrotor simulations to real Crazyflie hardware. Demonstrated on a four-robot position-swapping task, the system exhibits consistent trajectories and scalable computational performance across varying dynamical models, thereby validating both the effectiveness of the underlying non-cooperative game-theoretic algorithm and the practicality and generality of the proposed framework.
๐ Abstract
This paper presents a prototyping framework for distributed control of multi-robot systems, aimed at bridging theory and practical testing of distributed optimization algorithms. Using the Single Program, Multiple Data (SPMD) paradigm, the framework emulates distributed control on a single computer, with each core running the same algorithm using local states and neighbour-to-neighbour communication. We demonstrate the framework on a four-quadrotor position-swapping task using a non-cooperative game-theoretic distributed algorithm. Computational time and trajectory data are compared across the supported dynamics levels: a point-mass model, a high-fidelity quadrotor model, and an experimental hardware testbed using Crazyflie quadcopters. The results show that the framework provides a low-cost and accessible approach for validating distributed algorithms.