Reducing complexness of control by intelligent mechanics in undulant swimming robots

Abstract

This article introduces a biologically inspired modular swimming robot. Due to defi ned interfaces in mass, energy, and information fl ux, the robot's swimming behavior is changeable: an undulant, successive called anguilliform as well as a thunniform swimming mode is adjustable. Unlike the current state of the art, the robot comes with specifi c designed mechanics for the reduction of the complexity of software-sided control. Thereby, the number of actuators required for propulsion is reduced to the minimum number of one. Currently the prototype robot consists of a basic structure generating amongst others the required torque and several effector modules. The locomotion mode is switchable depending on the number of effector modules. Thereby, the latest anguilliform setup contains three effector modules. The current thunniform confi guration features one effector module. The effector modules are mechanically coupled with a manually tunable compliant joint. Optimum values concerning spring stiffness subjected to the location of the joint within the robot are evaluated by simulation.