Parameter optimization of fish robot’s smooth gaiting using Hill Climbing - Genetic Algorithm

Abstract

Recently, the underwater robotics research field has been developed quickly in some kinds of robot such as: ROV (remotely operated vehicle), AUV (autonomous underwater vehicle) or UUV (unmanned underwater vehicle) and etc. These types of underwater robot mostly use propeller or thruster to generate the propulsion force in order to create the movement for robot. However, the research about one kind of autonomous robot called fish robot still remain at low level of technology and there are many thing need to be done in this robot type. Fish robot usually uses its bodypsilas oscillation or its finspsila oscillation to create the movement for it self in underwater environment. In this paper, a model of 3-joint (4 links) fish robot type is presented. This fish robotpsilas smooth gait or smooth motion is optimized by using the combination of hill climbing algorithm (HCA) and genetic algorithm (GA). HCA is used to generate the good initial population for GA and then GA is used to find the optimal parameters set that make fish robot has a smooth gait or smooth motion. Finally, some simulation results are presented to prove this application.