A High-Strength Neuromuscular System That Implements Reflexes as Controlled by a Multiquadrant Artificial Efferent Nerve.

Abstract

We demonstrate an artificial efferent nerve that distinguishes environment-responsive conditioned and unconditioned reflexes, i.e., hand-retraction reflex and muscle memory, respectively. These reflex modes are immediately switchable by altering the polarity of charge carriers in a parallel-channeled artificial synapse; this ability emulates multiplexed neurotransmission of different neurotransmitters to form glutamine-induced short-term plasticity and acetylcholine-induced long-term plasticity. This is the successful control of high-strength artificial muscle fibers by using an artificial efferent nerve to form a neuromuscular system that can realize curvature and force simultaneously and in which all these aspects far surpass currently available neuromuscular systems. Furthermore, the special four-quadrant information-processing mechanism of our artificial efferent nerve allows complex application extensions, i.e., relative-position tracking of sound sources, immediate switchable learning modes between fast information processing and long-term memory, and high-accuracy pattern cognition. This work is a step toward development of human-compatible artificial neuromuscular systems.