An Investigation into Optimization of Flapping Gait of Snake Robot

One of the main advantages of snake-like robot over wheeled vehicles is its capability to move with different modes of locomotion, also called gaits. Because of inherent power inefficiency of snake robots, finding parameters that allow efficient motion for different gaits is of great importance. One of these gaits is called "flapping gait". This gait is usually used for obstacle avoidance in the path of motion. In this study, parameters effecting efficient flapping gait locomotion are investigated. We first drive dynamic equations of motion for an n-link two dimensional snake robot. We study the effect of flapping gait locomotion parameters on speed and input power. We then use Genetic Algorithm (GA) to find parameters for optimal motion. We also investigate the effect of numbers of robot links on maximum achievable speed and obtained optimal gait. We will show that the obtained optimal parameters, generate a similar gait for robot with any number of links. Effect of different environmental conditions (friction effects) on optimal parameter are also investigated. Next we consider input power and robot speed as optimization criterion and find pareto-optimal curve. By extracting points on this curve we find relations that are useful for designing an efficient robot moving with different speed.

Keywords:

snake robot , flapping gait , genetic algorithm , dynamic equation , optimal motion