Szász Csaba

Fault-tolerant Mechatronic Systems Development: a Biologically-inspired Approach


AbstractModern mechatronics embeds sophisticated control systems to meet increased performance and safety requirements. Timely fault detection is a critical requirement especially in safety-critical mechatronic applications, where a minor fault can evolve to catastrophic situations. In such cases it looks a high demand for more reliable, safety and fault-tolerant mechatronic systems development. The alternative to overcome all these bottlenecks was inspired from the biological world. By adapting the remarkable surviving and self-healing abilities of living entities it is possible to develop novel hardware systems suitable to fulfill in all the most demanding high reliability operation criteria’s and requirements. The paper presents a biologically-inspired computing system based on a Field Programmable Gate Array (FPGA) network developed for high reliability mechatronic applications. By choosing a design strategy relying on a multi-cellular concept which outlines the versatility of biologically inspired technologies, task allocation or reliability problems can be solved with high efficiency. Real-time simulations prove that by implementing methods that imitate biological processes, high performance fault-tolerant and self-healing hardware architectures can be experimented and tested. The benefits of this approach are also confirmed by experiments performed on a laboratory-prototype hardware platform. The results underline that techniques which imitate bio-inspired strategies can offer viable solutions in high reliability mechatronic systems development.