Robot navigation system using intrinsic evolvable hardware

Abstract

Recently there has been great interest in the idea that evolvable system based on the principle of artificial intelligence can be used to continuously and autonomously adapt the behaviour of physically embedded systems such as autonomous mobile robots and intelligent home devices. Meanwhile, we have seen the introduction of evolvable hardware (EHW): new integrated electronic circuits that are able to continuously evolve to adapt the changes in the environment implemented by evolutionary algorithms such as genetic algorithms (GA) and reinforcement learning. This paper concentrates on developing a robotic navigation system whose basic behaviours are obstacle avoidance and light source navigation. The results demonstrate that the intrinsic evolvable hardware system is able to create the stable robotic behaviours as required in the real world instead of the traditional hardware systems.