
Eric Kerrigan
Biography
Biography
Eric Kerrigan is Professor of Control and Optimization at Imperial College London, where he has a joint appointment in the Department of Electrical and Electronic Engineering and Department of Aeronautics. His research is centered on the development of efficient numerical methods for real-time optimal control and estimation, with applications spanning aerospace, renewable energy, and information systems. He has served as a Senior Editor of the IEEE Transactions on Control Systems Technology, an Associate Editor of the IEEE Transactions on Automatic Control, and as member of the IEEE Control Systems Society Board of
Is the solution to my optimal control problem correct?
Many challenging problems in control, estimation, and design can be formulated as dynamic optimization tasks, which involve optimizing an objective subject to differential equations and other constraints. While collocation-based numerical methods are widely used, they only enforce the governing equations at discrete points, often leading to difficulties in problems with high-index differential–algebraic equations, overdetermined constraints, or singular arcs.
In this talk, I will present integrated residual methods, a generalization of collocation that retains the strengths of standard approaches but overcomes their limitations. By incorporating information from the integrated residuals of the differential equations, these methods provide more accurate error estimates and handle otherwise intractable problem structures. I will show how integrated residual methods not only succeed where traditional collocation fails, but can also be computationally more efficient, making them attractive for real-time and large-scale applications. Numerical experiments and theoretical results will be discussed, illustrating the flexibility and robustness of these methods in practice.