Control of Modular Robot with Parameter Estimation Using Genetic Algorithms

Abstract

A novel way to identify friction model and torque sensor parameters of a modular robot joint is proposed and experimentally studied in this paper. The identification method is based on a genetic algorithm (GA). A model based friction compensation method and a real coded GA are integrated in the proposed method, and then applied to an experimental modular robot joint with a harmonic drive and built-in torque sensor. The friction parameters, as well as the torque sensor gain and offset, are identified and used in the control system, and the position tracking error reduction is demonstrated with experimental results.