Acceleration based PSO for Multi-UAV Source-Seeking

TL;DR

This paper introduces a novel acceleration-based PSO algorithm for multi-UAV source seeking, demonstrating improved stability, convergence, and performance over existing methods through theoretical analysis and high-fidelity simulations.

Contribution

The paper proposes APSO, a new PSO variant updating particle acceleration instead of velocity, with proven stability and superior search performance in UAV source seeking tasks.

Findings

APSO outperforms existing PSO algorithms in various scenarios.

Theoretical stability and convergence of APSO are established.

Simulations confirm enhanced robustness and efficiency of APSO.

Abstract

This paper presents a novel algorithm for a swarm of unmanned aerial vehicles (UAVs) to search for an unknown source. The proposed method is inspired by the well-known PSO algorithm and is called acceleration-based particle swarm optimization (APSO) to address the source-seeking problem with no a priori information. Unlike the conventional PSO algorithm, where the particle velocity is updated based on the self-cognition and social-cognition information, here the update is performed on the particle acceleration. A theoretical analysis is provided, showing the stability and convergence of the proposed APSO algorithm. Conditions on the parameters of the resulting third order update equations are obtained using Jurys stability test. High fidelity simulations performed in CoppeliaSim, shows the improved performance of the proposed APSO algorithm for searching an unknown source when compared with the state-of-the-art particle swarm-based source seeking algorithms. From the obtained results, it is observed that the proposed method performs better than the existing methods under scenarios like different inter-UAV communication network topologies, varying number of UAVs in the swarm, different sizes of search region, restricted source movement and in the presence of measurements noise.