Property Investigation of Chemical Plume Tracing Algorithm in an Insect using Bio-machine Hybrid System Daisuke KURABAYASHI 1 , Yosuke TAKAHASHI 1 , Ryo MINEGISHI 1 , Elisa TOSELLO 2 , Enrico PAGELLO 2 , and Ryohei KANZAKI 3 1 Tokyo Institute of Technology, Tokyo, 152-8552, JAPAN, dkura@irs.ctrl.titech.ac.jp, WWW home page: http://www.irs.ctrl.titech.ac.jp 2 IAS-Lab, Dept. of Information Engieering, University of Padua, 35131 Padua, Italy, 3 Research Center for Advanced Science and Technology, The University of Tokyo, Tokyo, 153-8904, Japan Abstract. In this study, we investigated an aspect of the chemical plume tracing behavior of an insect by using a bio-machine hybrid sys- tem. We implemented an experimental system by which an insect brain was connected to a robot body. We observed neural responses to external disturbances and transitions at changes in the motor gain of the robot body. Based on the results of the experiments, we identified a simple control model of the angular velocity of the behavior. We subsequently investigated the effect of the rotational velocity by using information entropy in computer simulations. Keywords: chemical plume tracing, bio-machine hybrid system, silk- worm moth 1 Introduction In this study, we investigated an aspect of the behavior of an insect to locate a chemical source. In air, chemicals do not make a smooth gradient, but form several plumes. Locating a chemical source by following plumes is known as the chemical plume tracing (CPT) problem [1][2]. CPT is potentially important for artificial systems because it can be applied to the location of the source of pollution, finding people trapped under debris after large earthquakes, etc. Since the dynamics of chemicals in the atmosphere is quite complex, we would like to investigate adaptive behaviors to solve CPT exhibited by adaptive algorithms in an animal. Animals effectively use CPT for foraging, mating, localizing, etc. In contrast to artificial robots, most animals, including an insect, have the ability to solve CPT problems. In this study, we investigate the CPT performance of insects by using an insect brain machine-interface system, called a bio-machine hybrid system. We then consider the effect of motion parameters by using computer simulations and information entropy.