*********************************
There is now a CONTENT FREEZE for Mercury while we switch to a new platform. It began on Friday, March 10 at 6pm and will end on Wednesday, March 15 at noon. No new content can be created during this time, but all material in the system as of the beginning of the freeze will be migrated to the new platform, including users and groups. Functionally the new site is identical to the old one. webteam@gatech.edu
*********************************
Title: Real-Time Error Detection and Correction for Robust Operation of Autonomous Systems using Encoded State Checks
Committee:
Dr. Abhijit Chatterjee, ECE, Chair , Advisor
Dr. Mark Davenport, eCE
Dr. Samuel Coogan, ECE
Dr. David Anderson, ECE
Dr. Haomin Zhou, Math
Abstract: The objective of the proposed research is to develop methodologies, support algorithms and software-hardware infrastructure for detection, diagnosis, and correction of failures for actuators, sensors and control software in linear and nonlinear state variable systems with the help of multiple checks employed in the system. This objective is motivated by the proliferation of autonomous sense-and-control real-time systems, such as intelligent robots and self-driven cars which must maintain a minimum level of performance in the presence of electro-mechanical degradation of system-level components in the field as well as external attacks in the form of transient errors. A key focus is on rapid recovery from the effects of such anomalies and impairments with minimal impact on system performance while maintaining low implementation overhead as opposed to traditional schemes for recovery that rely on duplication or triplication. On-line detection, diagnosis and correction techniques are investigated and rely on analysis of system under test response signatures to real-time stimulus. For on-line error detection and diagnosis, linear and nonlinear state space encodings of the system under test are used and specific properties of the codes, as well as machine learning model based approaches were used are analyzed in real-time. Recovery is initiated by copying check model values to correct error for sensor and control software malfunction, and by redesigning the controller parameter on-the-fly for actuators to restore system performance. Future challenges that need to be addressed include viability studies of the proposed techniques on mobile autonomous system in distributed setting as well as application to systems with soft as well as hard real-time performance constraints.
