Jiaqi (Catherine) Li

Jiaqi (Catherine)
Li
李嘉琦

Email: 

lijq@shanghaitech.edu.cn

Room Number: 

1D-204

Advisor: 

Boris Houska

Bio & Research: 

In June of 2015, Jiaqi graduated from Shanghai Institute of Technology. Her bachelor degree is in Electrical engineering and automation. Since September of 2015, she is pursuing a master degree at ShanghaiTech University. Her research area is Robust Control and she is interested in robotics and control of autonomous vehicles.

 

Education

  • 2015.9-Now: School of Micro-Systems Chinese Academy of Sciences @ ShanghaiTech University; Robust Optimal Control Master of expected (2018.6)
  • 2011.9-2015.7: Shanghai Institute of Technology Electrical engineering and automation  Bachelor.

 

Research

  • Towards rigorous robust optimal control via generalized high-order moment expansion.

This paper is concerned with the rigorous solution of worst-case robust optimal control problems having bounded time-varying uncertainty and nonlinear dynamics with affine uncertainty dependence. We propose an algorithm that combines existing uncertainty set-propagation and moment-expansion approaches. Specifically, we consider a high-order moment expansion of the time-varying uncertainty, and we bound the effect of the infinite-dimensional remainder term on the system state, in a rigorous manner, using ellipsoidal calculus. We prove that the error introduced by the expansion converges to zero as more moments are added. Moreover, we describe a methodology to construct a conservative, yet more computationally tractable, robust optimization problem, whose solution values are also shown to converge to those of the original robust optimal control problem.

  • A computational procedure for ellipsoidal robust forward invariant tubes in nonlinear MPC.

Min-max differential inequalities (min-max DIs) can be used to characterize robust forward invariant tubes with convex cross-section for a large class of nonlinear control systems. The advantage of deriving set-propagation via min-max DIs is that—unlike other existing approaches for tube-based MPC—they avoid the discretization of control policies. Thus, the conservatism of min-max DI tube MPC arises from the discretization of sets in the state space, while the control law is never discretized and remains defined implicitly via the solution of a min-max optimization problem. The contribution of this paper is the development of a practical implementation of min-max DI based tube MPC using ellipsoidal set approximations.