Authors: Farhad Maroofkhani; Amir Ali Forough Nassiraei; Kazuo Ishii
Addresses: Department of Human Intelligence Systems, Kyushu Institute of Technology, 2-4, Hibikino, Wakamatsu, Kitakyushu, 808-0196, Fukuoka, Japan ' Department of Electrical and Computer Engineering Space Robotic Lab, Concordia University, 1515 St. Catherine St. W., EV-5.139 Montreal, QC, H3G 2W1, Canada ' Department of Human Intelligence Systems, Kyushu Institute of Technology, 2-4, Hibikino, Wakamatsu, Kitakyushu, 808-0196, Fukuoka, Japan
Abstract: An autonomous robot's search strategy is the set of rules that it employs while looking for targets in its environment. Biological systems (e.g., foraging animals) provide useful inspirations for designing optimal stochastic search algorithms for autonomous robots. Due to the complexity of interaction between the robot and its environment, optimisation must be performed in high-dimensional parameter space. We analyse the dependence of search efficiency on environmental parameters and robot characteristics using response surface methodology (RSM), a technique originally developed for experimental design. In this study, the efficiency of a strategy focuses on Lévy walk search strategies on two-dimensional landscapes with clumped resource distributions. We show how RSM techniques can be used to identify optimal parameter values and to describe how sensitive is the efficiency to the changes in these values.
Keywords: Lévy walk; autonomous robots; swarm robot; biomimetic; individual motion; design of experiments; DOE.
International Journal of Reasoning-based Intelligent Systems, 2020 Vol.12 No.3, pp.187 - 192
Received: 05 Oct 2018
Accepted: 12 Nov 2018
Published online: 18 Sep 2020 *