Evolvable Hardware and its application to pattern recognition and fault-tolerant systems

Abstract

This paper describes Evolvable Hardware (EHW) and its applications to pattern recognition and fault-torelant systems. EHW can change its own hardware structure to adapt to the environment whenever environmental changes (including hardware malfunction) occur. EHW is implemented on a PLD(Programmable Logic Device)-like device whose architecture can be altered by re-programming the architecture bits. Through genetic algorithms, EHW finds the architecture bits which adapt best to the environment, and changes its hardware structure accordingly.

Two applications are described: the the pattern recognitionsystem and the V-shape ditch tracer with fault-tolerant circuit. First we show the exclusive-OR circuit can be learned by EHW successfully. Then the pattern recognition system with EHW is described. The objective is to take the place of neural networks, solving its weakness such as readability of learned results and the execution speed. The results show that EHW works as a hard-wired pattern recognizer with such the robustness as neural nets. The second application is the V-shape ditch tracer as part of a prototypical welding robot. EHW works as the backup of the control logic circuit for the tracing, although the EHW is not given any information about the circuit. Once a hardware error occurs, EHW takes over the malfunctioning circuit.