Control of a Noisy Mechanical Rotor

We investigate the control of the dynamics of a periodically kicked mechanical rotor in the presence of noise. It was recently shown that the system dynamics shows the characteristics of a complex multistable system. We demonstrate that it is possible to stabilize the system at a desired attracting state even in the presence of considerable noise level. As control strategy we use a recently developed algorithm for the control of chaotic systems which is based on reinforcement learning. This method finds a global optimal control policy directing the system from any initial state towards the desired state in a minimum number of iterations and stabilizes the system once a neighborhood of the desired state is reached. The algorithm does not use any information about governing equations. Key-Words: Multistability, mechanical rotor, control, chaos, reinforcement learning