On control of parallel robots for high dynamic performances : from design to experiments

Abstract

Parallel Kinematic Manipulators (PKMs) have gained an increased popularity in thelast few decades. This interest has been stimulated by the significant advantages of PKMscompared to their serial counterparts, such as better precision and higher accelerationcapabilities. Efficient and performant control algorithms play a crucial role in improvingthe overall performance of PKMs. Control of PKMs is often considered in the literature achallenging task due to their highly nonlinear dynamics, abundant uncertainties, parametersvariation, and actuation redundancy. In this thesis, we aim at improving the dynamicperformance of PKMs in terms of precision and robustness towards changes of operatingconditions. Thus, we propose robust control strategies being extensions of (i) the standardRobust Integral of the Sign of the Error (RISE) feedback control and (ii) the super-twistingSliding Mode Control (SMC). Moreover, an actuator and friction dynamics formulation isproposed within a model-based control strategy to compensate for their resulting errors.Lyaponuv-based stability analysis is established for all the proposed controllers verifyingthe asymptotic convergence of the tracking errors. In order to validate the proposed controllers,real-time experiments are conducted on several parallel robot prototypes: the 3-DOF Delta robot at EPFL, Switzerland, the 4-DOF VELOCE robot, and the 5-DOF SPIDER4robot at LIRMM, France. Several experiments are tested including nominal scenarios, robustnesstowards speed variation, and robustness towards payload changes. The relevanceof the proposed control schemes is proved through the improvement of the tracking errorsat different dynamic operating conditions.