nth degree polynomials joint angle path by approximation of inverse kinematics data using Genetic Algorithm

摘要

This paper proposes an idea to approximate few robot manipulator inverse kinematics data by nth degree polynomials function using a Genetic Algorithm (GA). This paper will find a joint angle path from inverse kinematics data in the form of nth degree polynomials parametric function. The GA is used as an approximation method. It will find proper coefficients of a polynomial function such that a polynomial curve is close to sample nodes. A fitness function is the minimum error between data and a function value. Third, fifth, seventh, and tenth polynomials degree approximation will be carried out. The results show that the GA can be used as the approximation methods with various errors for each degree and there is always the appropriate degree which gives the best result.