A study on optimization of fish robot maximum velocity using the combination of genetic - Hill Climbing Algorithm

Abstract

Underwater robot is a new trend of researched field which is developing quickly in recent years. Some of the first researches on this field are ROV (Remotely Operated Vehicle), AUV (Autonomous Underwater Vehicle). Lately, a new type of underwater robot which is biomechanical robot is mostly concerned. One of the typical types of this one is fish robot. In this paper, firstly a dynamic model of 3-joint (4 links) Carangiform fish robot type is presented. Secondly, the surveys about the influences of input torque functions' parameters such as: phase angle, frequency and amplitude to the velocity of fish robot by simulation method are introduced. Lastly, fish robot's maximum velocity is optimized by using the combination of Genetic Algorithm (GA) and Hill Climbing Algorithm (HCA). GA is used to create the initial optimal parameters set for the input functions of the system. Then, this set will be optimized one more time by using HCA to be sure that the final parameters set are the “near” global optimization result for the system. Finally, some simulation results are presented to prove the effectiveness of the proposed algorithm.