Evolving Genetic Regulatory Networks for Hardware Fault Tolerance

We present a new approach that is able to produce an increased fault tolerance in bio-inspired electronic circuits. In order to do so, we designed hardware-friendly genetic regulatory networks based on a bio-inspired hardware architecture called POEtic tissue. To assess the functionality of the system, the parameters of the genetic regulatory networks were evolved using genetic algorithms to achieve specific behaviors, including patch growth and oscillations at various frequencies. The tolerance to faults was explored by inflicting several types of damage, and results show that the system exhibits capabilities to recover from them.


Published in:
Parallel Problem Solving from Nature 8 (PPSN'04), 561-570
Presented at:
Parallel Problem Solving from Nature 8 (PPSN'04), Birmingham (UK), 18-22 September
Year:
2004
Keywords:
Note:
Sponsor: POEtic IST-2000-28027 (OFES 00.0529-1)
Other identifiers:
Laboratories:




 Record created 2006-01-12, last modified 2018-03-17

External link:
Download fulltext
URL
Rate this document:

Rate this document:
1
2
3
 
(Not yet reviewed)