A bio-inspired analog scheme for navigational control of lightweight autonomous agents

Abstract

We present an approach for the development of lightweight analog cognition for autonomous navigation in unknown environments. We consider a hypothetical differential drive robot, equipped with a limited sensor suite that includes one front and one rear monocular obstacle sensor and a target (bearing and distance) sensor. A robotic motion control scheme is developed for obstacle avoidance and target-seeking using an innovative dynamical systems architecture incorporating a nonlinear controller derived using Lyapunov techniques. We claim that the avoidance-steering behavioral property of the agent is a weakly emergent characteristic. Simulations are provided and discussed.