TR2022-039

PRISM: Recurrent Neural Networks and Presolve Methods for Fast Mixed-integer Optimal Control


    •  Cauligi, A., Chakrabarty, A., Di Cairano, S., Quirynen, R., "PRISM: Recurrent Neural Networks and Presolve Methods for Fast Mixed-integer Optimal Control", Learning for Dynamics and Control Conference (L4DC), April 2022, pp. 34-46.
      BibTeX TR2022-039 PDF
      • @inproceedings{Cauligi2022apr,
      • author = {Cauligi, Abhishek and Chakrabarty, Ankush and Di Cairano, Stefano and Quirynen, Rien},
      • title = {PRISM: Recurrent Neural Networks and Presolve Methods for Fast Mixed-integer Optimal Control},
      • booktitle = {Learning for Dynamics and Control Conference (L4DC)},
      • year = 2022,
      • pages = {34--46},
      • month = apr,
      • publisher = {Proceedings of Machine Learning Research (PMLR)},
      • url = {https://www.merl.com/publications/TR2022-039}
      • }
  • MERL Contacts:
  • Research Areas:

    Control, Machine Learning, Optimization

Abstract:

While mixed-integer convex programs (MICPs) arise frequently in mixed-integer optimal control problems (MIOCPs), current state-of-the-art MICP solvers are often too slow for real-time applications, limiting the practicality of MICP-based controller design. Although supervised learning has been proposed to hasten the solution of MICPs via convex approximations, they are not designed to scale well to problems with >100 decision variables. In this paper, we present PRISM: Presolve and Recurrent network-based mixed-Integer Solution Method, to leverage deep recurrent neural network (RNN) architectures such as long short-term memory (LSTMs) networks, in conjunction with numerical optimization tools to enable scalable acceleration of MICPs arising in MIOCPs.
Our key insight is to learn the underlying temporal structure of MIOCPs and to combine this with presolve routines employed in MICP solvers. We demonstrate how PRISM can lead to significant performance improvements, compared to branch-and-bound (B&B) methods and to existing supervised learning techniques, for stabilizing a cart-pole with contact dynamics, and a motion planning problem under obstacle avoidance constraints.