Methods and Tools to Design and Test Robot Control Systems

Abstract

It is still a major goal in industrial robotics to increase robot speed and enhance the accuracy of the robots by improving the control system. Short product cycle time require a hard- and software environment for state-of-the-art model-based control algorithms where the main effort can be dedicated to high-level design tasks, not to coding details, and results can easily and reliably be transferred from the laboratory test setup to serial products. In this work methods and tools are presented for fast developement, optimization and testing of new feedforward and feedback control laws based on component oriented robot device models. A hardware-in-the-loop environment for optimizing the parameters of new algorithms and validating the resulting perfomance has been established that uses only slightly modified control cabinet and software from series production.