Protocole d’allocation des tâches dans une constellation de robots basée sur la méthode hybride des enchères des enchères et K-Means

Abstract

In today’s context, tactical and operational responses to incidents such as natural disasters and fires are often inadequate due to anthropological factors (such as the risk to the lives of rescue teams) and natural factors (such as the accessibility of affected areas). The increasing use of robots aims to ensure the rescue of people and property in disaster zones without risking the lives of rescuers. Auction methods are used to assign rescue tasks, selecting the appropriate robot for each task. We have enhanced the HaucTioN protocol by integrating the concept of parallelism to eliminate bottlenecks in determining winning bids. The improved protocol, named HaucTioN-K due to the use of the K-Means clustering algorithm to form K groups of bids, proved superior to its predecessor in terms of temporal complexity and experimental performance.