Experimental robot identification using optimised periodic trajectories

This paper describes a new approach to the parameterisation of robot excitation trajectories for optimal robot identification. The trajectory parameterisation is based on finite Fourier series. The coefficients of the Fourier series are optimised for minimal sensitivity of the identification to measurement disturbances, which is measured as the condition number of a regression matrix, taking into account motion constraints in joint and Cartesian space. This approach allows small condition numbers with few coefficients for each joint to be obtained, which simplifies the optimisation problem significantly. The periodicity of the resulting trajectories and the fact that one has total control over their frequency content are additional features of the presented parameterisation approach. Further optimisation of the excitation experiments is possible through time domain data-averaging and optimal selection of the excitation bandwidth, which both help to reduce the disturbance level on the measurements, and therefore improve the identification accuracy. Application of the method for the identification of the KUKA IR/361 industrial robot proves the validity of the proposed approach. (C) 1996 Academic Press Limited