Robust and Reliable Error Detection and Correction for Autonomous Systems

Abstract

The rapid rise of self-driving cars and drones has raised questions about the safety of autonomous robotics deployed in society. This is due to the large numbers of system state variables involved, the resulting degraded ability to perform accurate error detection and most importantly, loss of the ability to perform accurate error diagnosis. Prior work on robust and adaptive control make assumptions about the boundedness of errors or require the use of full-scale system models running in the background for control reference. In this research, we show how state space checks facilitated by different machine learning algorithms can be used to detect, diagnose and compensate for errors in sensors, actuators and control program execution in linear and nonlinear systems for robotic applications. The primary focus is on low-cost, ultra-fast, efficient, and lightweight methods for mitigation of transient errors in sensor data and control program execution and parametric deviations in sensor circuitry and actuator subsystems. Additionally, the proposed approach should incur minimal hardware and software overhead. The proposed approach has been applied to multiple test-cases which includes DC motor control system, quadcopter as well as automotive subsystems such as steer by wire subsystem. Simulation results indicate that errors can be compensated with high efficiency and low computation overhead.