A Fast and Real-Time Digital Method To Sense Fault Conditions in Resonant-Based MEMS Systems

This work presents a novel method to detect fault conditions of the trajectory of resonant-based MEMS with a focus on Laser Beam Scanning systems. The technique is operated in real-time and implemented in the digital domain, making it suitable for systems with limited computational capabilities and small-footprint ASIC implementations. The results demonstrate the tradeoff between fault detection time and quality in the estimation of the MEMS trajectory. Compared with traditional approaches, reaction times to a fault event are improved by 100X, with worst-case values of approximately 13.7% of the 27.4 kHz resonant MEMS mirror cycle that has been used as test scenario.