Optimization

MERL is seeking a graduate student to develop a scalable optimization-based framework for feedforward-feedback co-design for nonlinear dynamical systems subject to path constraints. The framework will 1) support modeling and operational uncertainties, and 2) guarantee static and dynamic feasibility in closed-loop. The solution approach will leverage the state-of-the-art in sequential convex programming, contraction analysis, and first-order methods for semidefinite programming. The methods will be evaluated on high-dimensional motion planning problems in robotics. The results of the internship are expected to be published in top-tier conferences and/or journal in robotics, control systems, and optimization.

The internship is expected to start in Spring or Summer 2025 with an expected duration of 3-6 months depending on the agreed scope and intermediate progress.

Required Specific Experience

• Current/Past enrollment in a Ph.D. program in Mechanical, Aerospace, Electrical Engineering, Computer Science, or Applied Mathematics.
• 2+ years of research in at least some of: first-order algorithms for SDPs, contraction analysis, nonconvex trajectory optimization.
• Strong programming skills in Python and/or C/C++.

MERL is seeking a motivated and qualified individual to work on physics-informed scientific machine learning algorithms for problems governed by partial differential equations (PDEs). The ideal candidate will be a PhD student in engineering, computer science, or related fields with a solid background in scientific machine learning for PDEs. Preferred skills include knowledge of physics-informed neural networks, operator learning, nonlinear dimensionality reduction, and diffusion models. Strong coding abilities in Python and a popular deep learning framework such as Pytorch are essential. Publication of the results obtained during the internship is expected. The duration is expected to be at least 3 months with a flexible start date.

MERL is looking for a highly motivated and qualified candidate to work on stochastic control for grid-interactive net-zero energy buildings informed by deep generative models. The ideal candidate will have a strong understanding of optimization-based control with expertise demonstrated via, e.g., publications, in stochastic model predictive control.

Additional understanding of energy systems and machine learning is a plus. Hands-on programming experience with numerical optimization solvers and Python fluency is required. The results of this 3-6 month internship are expected to be published in top-tier energy systems and/or control venues.