Title:  Self-Checking Error-Resilient Smart Autonomous System Design

Committee: 

Dr. Chatterjee, Advisor  

Dr. Coogan, Chair

Dr. Mukhopdhyay

Abstract: The objective of the proposed research is to develop methodologies, support algorithms and infrastructure for cross-layer failure detection, diagnosis and correction in cyber-physical autonomous systems which consist of multiple interacting subsystems acting to meet commands from a supervisory controller. These failures can be in actuation, sensing, control computation or in machine learning subsystems. This objective is motivated by the proliferation of autonomous sense-and-control real-time systems in safety-critical roles such as autonomous driving. A key focus is on cross-layer synergies for system resilience, enhancing the autonomous system's safety through information sharing and coordination between its subsystems. On-line adaptation to failures in actuators, sensors, control computation errors, machine learning system computation errors and security threats to ML systems are investigated and methods are proposed which leverage cross-layer subsystem interactions for rapid, on-line failure adaptation.