Autonomous Control of a Mobile Robot using a Mechanical Metamaterial “Brain”

We present an autonomous mobile robot that is controlled exclusively by a flexible mechanical metamaterial without any digital electronics. The metamaterial — based on the rotating square mechanism — is mounted onto a wheeled mobile base and acts simultaneously as the sensory system that detects contact with obstacles and as the “brain” that computes appropriate motor control voltages to react to those contacts, seeking to free the robot from the obstacle and continue moving through the environment. This tactile control paradigm is loosely inspired by the concept of “thigmotaxis” in biological locomotion.



We outsource the control of the motors to the metamaterial using the principle of Physical Reservoir Computing. As the robot runs into an obstacle, the metamaterial deforms nonlinearly due to the contact with the obstacle. Strain sensors measure the deformation, then the strain measurements are linearly superimposed by weighing them with experimentally trained weights through an analogue circuit to produce the adequate motor control command. Hence, no digital computation is required. The “intelligence” and sensing capability of the robot lies in the nonlinearly deforming mechanical metamaterial.