Abstract This article presents a new algorithm for searching odour sources across large search spaces with groups of mobile robots. The proposed algorithm is inspired in the particle swarm optimization (PSO) method. In this method, the search space is sampled by dynamic particles that use their knowledge about the previous sampled space and share this knowledge with other neighbour searching particles allowing the emergence of efficient local searching behaviours. In this case, chemical searching cues about the potential existence of upwind odour sources are exchanged. By default, the agents tend to avoid each other, leading to the emergence of exploration behaviours when no chemical cue exists in the neighbourhood. This behaviour improves the global searching performance. The article explains the relevance of searching odour sources with autonomous agents and identifies the main difficulties for solving this problem. A major difficulty is related with the chaotic nature of the odour transport in the atmosphere due to turbulent phenomena. The characteristics of this problem are described in detail and a simulation framework for testing and analysing different odour searching algorithms was constructed. The proposed PSO-based searching algorithm and modified versions of gradient-based searching and biased random walk-based searching strategies were tested in different environmental conditions and the results, showing the effectiveness of the proposed strategy, were analysed and discussed.
Particle swarm-based olfactory guided search
01.01.2006
doi:10.1007/s10514-006-7567-0
Aufsatz (Zeitschrift)
Elektronische Ressource
Englisch
DDC: | 629 |
Particle swarm-based olfactory guided search
British Library Online Contents | 2006
|An improved diversity-guided particle swarm optimisation for numerical optimisation
British Library Online Contents | 2014
|An improved diversity-guided particle swarm optimisation for numerical optimisation
British Library Online Contents | 2014
|Particle Swarm Optimization with Enhanced Autonomous Search Ability of Each Particle
British Library Online Contents | 2008
|Consideration of Particle Swarm Optimization combined with tabu search
Online Contents | 2010
|