Adaptive Feed Forward Control of an Industrial Robot
Sprache des Vortragstitels:
Austrian Robotics Workshop 2015
Sprache des Tagungstitel:
Positioning and path accuracy is mandatory for state of the art robotic applications, where most of the path planning is done on computer systems. In most of the industrial robots a simple PD motor joint control is implemented, which suffers from quite high tracking errors. The performance of this controller can be improved by using a model based feed forward control. This model based feed forward control strongly depends on the dynamical parameters of the manipulator which have to be idenfified. However, parameters can vary during operation due to e.g. changes in friction conditions or changing loads. Therefore an online adaptation of the dynamical parameters is recommended. Two methods, namely the Recursive Least Squares (RLS) method and the Square Root Filtering (SRF) method, are presented and compared w.r.t tracking errors in the robots joints. The evaluation is done for a Stäubli TX90L industrial robot mounted on a linear axis leading to 7 degrees of freedom.