Fault-adaptive control of robotic systems

The goal of this research is to develop fault-adaptive controllers for robotic systems so that the robot can function in the presence of faults during an autonomous operation. It involves real-time fault detection, isolation, and control accommodation. The faults that are of interests include both actuator and sensor faults. We employ model-based fault detection and isolation techniques that are integrated with hybrid control methodologies. Robust control techniques are used to design the individual controllers of the hybrid system. The application of this research is demonstrated both on a single mobile robot and a team of mobile robots, where the robots accomplish a given task in the presence of system faults.

Related publications:

• Sidek, N., Sarkar, N., "Integrating Actuator Fault and Wheel Slippage Detections within FDI Framework", WSEAS Transaction On Systems, Issue 2, Vol. 6, Feb. 2007. (Note: The same paper also appears in the Proceeding: International Conf. of CSECS, Dallas, Texas, USA, Nov. 1-3, 2006).
• Halder. B, Sarkar. N, “Robust Fault Detection of Robotic Manipulator,” International Journal of Robotics Research, vol. 26, no.3, pp. 273-285, 2007.
• Halder. B and Sarkar. N, “Robust Nonlinear Analytic Redundancy for Fault Detection and Isolation in Mobile Robot,” International Journal of Automation and Computing, vol. 4, no.2, pp. 183-188, 2007.
• Ji, M., Zhang,  Z., Biswas, G. and Sarkar, N. "Hybrid Fault Adaptive Control of a Wheeled Mobile Robot," in the IEEE/ASME Transactions on Mechatronics, Vol. 8, No. 2, pp. 226-233, 2003.

Mechanical Engineering Home | School of Engineering Home
Vanderbilt University Home