Evaluation Platform for Testing Fault Tolerance Properties: Soft-core Processor-based Experimental Robot Controller

conference paper

English

Various electronic systems play an important role in our everyday lives. Some of them serve for fun or to make our lives easier. These systems are useful but not necessary; when they malfunction, the consequences are not critical. On the other hand, there are systems which are more or less critical, and their failure can cause undesirable consequences. For example, a failure in medicine, aviation, the army or automotive systems can cause high economic losses and/or endanger human health. These systems must be protected against the impact of faults, and flawless operation must be ensured. Fault tolerance is one of the techniques that will ensure this. There are many fault-tolerance methodologies targeted towards various systems and technologies, and new methodologies are being investigated. It is also important to verify these techniques; this is the main topic of this paper. An evaluation platform for testing fault-tolerance methodologies targeted towards SRAM-based FPGAs (Field Programmable Gate Arrays) is presented and demonstrated. A robot for seeking a path through a maze and the processor-based robot controller serve as an experimental system case study. Experimental results with the unhardened and hardened versions of the processor-based robot controller are presented and discussed.

Soft-core Processor, NEO430, TMR, FPGA, Fault Tolerance, Robot Controller, Reconfiguration.

PODIVÍNSKÝ, J.; LOJDA, J.; ČEKAN, O.; KOTÁSEK, Z.

29. 8. 2018

IEEE Computer Society

Praha

978-1-5386-7376-8

Proceedings of the 2018 21st Euromicro Conference on Digital System Design

229

236

8

https://www.fit.vut.cz/research/publication/11705/

evaluation_platform.pdf