A Pneumatically Driven Stewart Platform Used as Fault Detection Device

This paper investigates the sensorless detection and evaluation of inner oscillations of unknown test objects mounted on a compliant test bench. The principle of the sensorless analysis is that test objects are not totally rigid in reality. This means one or more parts of the test objects are oscillating with different eigenfrequencies compared to their rigid equivalent. By comparing eigenfrequencies of both (rigid and fault test object) oscillating parts are detectable. The aim of this experiment is to demonstrate the use of a 6 DOF compliant Stewart platform (alternatively used in a simulation environment) to generate frequency sweeps in all degrees of freedom, to get a sensorless detection of vibrations in unknown objects. For this purpose only the preexisting sensors applied for the control of the hexapod should be used. The detection of loose parts by shaking objects can be done by a complex robotic manipulation task. Being designed for flexible use by small and medium-sized enterprises, the robotic Stewart platform (hexapod) will adapt autonomously to different test objects leading to a highly flexible robot.