Bio-inspired robotics: application of a CNN-based CPG VLSI chip to control an autonomous mini-hexapod robot

Abstract

This demo focuses on the description of a VLSI chip for locomotion control and its applications. The chip realizes a Central Pattern Generator through a CNN-based structure [1], The chip has been already used to control different bio-inspired structures like hexapods, octopods, lamprey-like and worm-like robots [2]. In the proposed demo the demonstration platform is a completely autonomous mini-legged robot. The minihexapod structure will be presented describing the mechanical design and the electronics boards. The robot has a battery pack equipped on board and can move autonomously on different kinds of terrains. Moreover the chip permits the integration of exteroceptive sensors that can be used to close the feedback loop with the environment implementing simple navigation control strategies. We have endowed a suite of distance sensors on the robot that can accomplish two different tasks: obstacle avoidance and target pursuit. The robot is able to change the direction of motion stopping in stance phase one of the middle legs; this behaviour is obtained modifying the control inputs of the chip and then the locomotion pattern that emerges in the CNN lattice. The fully working minihexapod will be available for a hardware description and a set of experiments (for details see [3]).