Energy-Optimal Motion Planning with Robust Residual Oscillation Suppression on a Translating Flexible Beam

Motion planning is probably the most cost-effective approach for the reduction of the energy consumption of automatic machines and manipulators. Here a model-based direct motion planning approach is introduced and tested numerically: its aim is to generate energy-efficient motion profiles for uncertain systems. The impact of uncertainties on the outcome of the execution of the planned motion can be minimized by the proposed approach, while minimizing also the energy consumption. The method is based on the development of a direct planning framework augmented with the sensitivities of the dynamic system for which the motion is planned. The testcase under consideration is slender beam moved by a linear actuator. The numerical results indicate that energy efficiency can be enhanced with limited residual oscillations, even in the presence of sensible model-plant mismatches.