School of Advanced Technology would like to welcome and introduce our new faculty member: Dr. Yuqing Chen from the Department of Mechatronics and Robotics.
Optimal control theory provides a systematic approach to control complex robotic systems. Assuming model of the robotic system is known, the optimal control theory can be applied to control complex robotic systems with high degrees of freedom by minimizing a pre-defined cost function. However, the applicability of optimal controllers to robotic systems is mainly limited two factors. First, it is nontrivial to obtain precise model of the complex robotic systems, especially modern robot with nonlinear actuators. Imprecise model would lead to inevitably model-bias and deteriorate the optimal control performance. Another factor is the excessive online computation required by optimal control algorithms. The online computational resource in real-time robotics systems is limited to 1ms, but controlling high-dimension robotic systems would require far more computational time than the provided interval.
Yuqing has developed novel hardware-in-the-loop optimal controllers to overcome these limitations. The model bias is reduced by eliminating model-based future prediction in numerical computation. The proposed controllers could be efficiently computed online by requiring less than 0.1ms computation per time step. With numerical simulation and robot experiment, the proposed algorithm is demonstrated to provide improved performance compared to state-of-the-art optimal control methods.
Expectation for XJTLU and School of Advanced Technology:
It is my great honor to join this international team in School of Advanced Technology, XJTLU. I am looking forward to collaborating with our staff and students to develop intelligent robotic systems, and helping our students to develop independent learning ability so that they will possess their unique advantages in competitive society.
（By Huatian Jin, photo provided by Dr Yuqing Chen）