Genetic-algorithm-based global design optimization of tree-type robotic systems involving exponential coordinates

Abstract

This paper proposes the simultaneous design optimization of the geometry (joint position and directions) and the topology (joint distribution and connection) of tree-type robotic systems based on an exponential coordinate system expression. Tree-type systems represent a versatile system expression of mechanical systems comprising multiple serial link chains branching from the root. Previously, we derived general closed-form formulas of the kinematics and dynamics of tree-type systems in exponential coordinates by introducing a connectivity matrix called the chain matrix. In this study, these results are first extended to floating base and closed-chain systems to enlarge the system framework. Next, the efficient coding of the system parameters by using a genetic algorithm (GA) is demonstrated. The closed-form formulas of the kinematics and dynamics allow for cost evaluations through numerical simulations with feedback control. Design examples of a robotic platform and a grasping/manipulation system illustrate the proposed optimization process.