A Bio-inspired Spiking Neural Network for Control of A 4-DoF Robotic Arm

Abstract

This paper explores the control task of a 4-DoF robotic arm via a Spiking Neural Network (SNN). Inspired from the biological neuron control mechanism, the SNN is proposed and analyzed for the robotic arm control. The SNN adopts a data-driven way to estimate the kinematic properties of the robotic arm and further spares the difficulty of analytical model building. Biologically, the desired target position and sensory information are processed into the network, and the patterns of motor commands are able to extract from the readout layer of the SNN. Finally, numerical studies are conducted to verify the effectiveness of the proposed SNN.