Fault Tolerant Control Architecture Design for Mobile Manipulation in Scientific Facilities

Abstract

This paper describes one of the challenging issues implied by scientific infrastructures on a mobile robot cognition architecture. For a generally applicable cognition architecture, we study the dependencies and logical relations between several tasks and subsystems. The overall view of the software modules is described, including their relationship with a fault management module that monitors the consistency of the data flow among the modules. The fault management module is the solution of the deliberative architecture for the single point failures, and the safety anchor is the reactive solution for the faults by redundant equipment. In addition, a hardware architecture is proposed to ensure safe robot movement as a redundancy for the cognition of the robot. The method is designed for a four-wheel steerable (4WS) mobile manipulator (iMoro) as a case study.

Actuator failure compensation for two linked 2WD mobile robots based on multiple-model control

“…As for the application to wheeled mobile robots, some fault diagnosis methods are developed (e.g., Fourlas et al, 2015;Goel et al, 2000;Skoundrianos and Tzafestas, 2004), a sensor fault accommodation scheme is presented by Ji and Sarkar (2007), some fault-tolerant control systems are designed by Koh et al (2012), Zhang and Cocquempot (2014), Rotondo et al (2014), Kim et al (2015), and Aref et al (2015) for four-wheel drive robots, and a hybrid fault adaptive control scheme is designed by Ji et al (2003) to accommodate partial faults and degradation for two-wheel drive (2WD) mobile robots. However, even if it is important to compensate a partial loss of wheel-motor effectiveness, the actuators (motors) of a wheeled mobile robot may also be totally faulty.…”