I’m co-organizing an open invited track at the IFAC world congress 2026 on learning interpretable control policies. Find details and submission code here!

I’m a postdoc in the Mesbah Lab at UC Berkeley. My works is driven by the need for built-in safety of the architectures and algorithms governing decision-making agents. In that vein, my recent work develops a complementary framework inspired by deep reinforcement learning and model predictive control. These two areas are often viewed as opposites, but really they have a common core in dynamic programming and Markov decision processes. Essentially, RL represents one branch that solves decision-making tasks through trial and error and function approximators, while MPC is another branch that is based on dynamics, constraints, and optimization. RL has proved to be incredibly versatile but is not necessarily safe by-design; MPC puts safety and robustness at the forefront, but it can be difficult to design toward high performance. A more unified perspective of these two areas would make RL more appealing for real-world applications while also making MPC more flexible and scalable under general learning algorithms. Please see my recent works on RL and MPC and don’t hesitate to get in touch.

Publications

Most of my papers are linked to arXiv or a DOI. You can also find my articles on my Google Scholar profile.

1. Journal Papers
   1. A view on learning robust goal-conditioned value functions: Interplay between RL and MPC
      Nathan P. Lawrence, Philip D. Loewen, Michael G. Forbes, R. Bhushan Gopaluni, Ali Mesbah
      arXiv. 2025.
      [Link] [arXiv]
   2. Stabilizing reinforcement learning control: A modular framework for optimizing over all stable behavior
      Nathan P. Lawrence, Philip D. Loewen, Shuyuan Wang, Michael G. Forbes, R. Bhushan Gopaluni
      Automatica. 2024.
      [Link] [arXiv]
   3. Machine learning techniques for industrial sensing and control: A survey and practical perspective
      Nathan P. Lawrence, Seshu Kumar Damarla, Jong Woo Kim, Aditya Tulsyan, Faraz Amjad, Kai Wang, Benoit Chachuat, Jong Min Lee, Biao Huang, R. Bhushan Gopaluni
      Control Engineering Practice. 2024.
      [Link] [arXiv]
   4. Automated deep reinforcement learning for real-time scheduling strategy of multi-energy system integrated with post-carbon and direct-air carbon captured system
      Tobi Michael Alabi, Nathan P. Lawrence, Lin Lu, Zaiyue Yang, R. Bhushan Gopaluni
      Applied Energy. 2023.
      [Link] [arXiv]
   5. Meta-reinforcement learning for the tuning of PI controllers: An offline approach
      Daniel G. McClement, Nathan P. Lawrence, Johan U. Backström, Philip D. Loewen, Michael G. Forbes, R. Bhushan Gopaluni
      Journal of Process Control. 2022.
      [Link] [arXiv]
   6. Deep reinforcement learning with shallow controllers: An experimental application to PID tuning
      Nathan P. Lawrence, Michael G. Forbes, Philip D. Loewen, Daniel G. McClement, Johan U. Backström, R. Bhushan Gopaluni
      Control Engineering Practice. 2022.
      [Link] [arXiv]
   7. Toward self‐driving processes: A deep reinforcement learning approach to control
      Steven Spielberg, Aditya Tulsyan, Nathan P. Lawrence, Philip D. Loewen, R. Bhushan Gopaluni
      AIChE Journal. 2019.
      [Link] [arXiv]
2. Conference Proceedings
   1. MPCritic: A plug-and-play MPC architecture for reinforcement learning
      Nathan P. Lawrence, Thomas Banker, Ali Mesbah
      arXiv. 2025.
      [Link] [arXiv]
   2. Local-Global Learning of Interpretable Control Policies: The Interface between MPC and Reinforcement Learning
      Thomas Banker, Nathan P. Lawrence, Ali Mesbah
      arXiv. 2025.
      [Link] [arXiv]
   3. Guiding reinforcement learning with incomplete system dynamics
      Shuyuan Wang, Jingliang Duan, Nathan P Lawrence, Philip D Loewen, Michael G Forbes, R Bhushan Gopaluni, Lixian Zhang
      IROS. 2024.
      [arXiv]
   4. Deep Hankel matrices with random elements
      Nathan P. Lawrence, Philip D. Loewen, Shuyuan Wang, Michael G. Forbes, R. Bhushan Gopaluni
      Learning for Dynamics & Control Conference. 2024.
      [Link] [arXiv]
   5. Reinforcement learning with partial parametric model knowledge
      Shuyuan Wang, Philip D. Loewen, Nathan P. Lawrence, Michael G. Forbes, R. Bhushan Gopaluni
      IFAC World Congress. 2023.
      [Link] [arXiv]
   6. A modular framework for stabilizing deep reinforcement learning control
      Nathan P. Lawrence, Philip D. Loewen, Shuyuan Wang, Michael G. Forbes, R. Bhushan Gopaluni
      IFAC World Congress. 2023.
      [Link] [arXiv]
   7. Meta-reinforcement learning for adaptive control of second order systems
      Daniel G. McClement, Nathan P. Lawrence, Michael G. Forbes, Philip D. Loewen, Johan U. Backström, R. Bhushan Gopaluni
      AdCONIP. 2022.
      [Link] [arXiv]
   8. A meta-reinforcement learning approach to process control
      Daniel G. McClement, Nathan P. Lawrence, Philip D. Loewen, Michael G. Forbes, Johan U. Backström, R. Bhushan Gopaluni
      AdCHEM. 2021.
      [Link] [arXiv] | Keynote Paper
   9. Reinforcement learning based design of linear fixed structure controllers
      Nathan P. Lawrence, Gregory E. Stewart, Philip D. Loewen, Michael G. Forbes, Johan U. Backstrom, R. Bhushan Gopaluni
      IFAC World Congress. 2020.
      [Link] [arXiv]
   10. Optimal PID and antiwindup control design as a reinforcement learning problem
       Nathan P. Lawrence, Gregory E. Stewart, Philip D. Loewen, Michael G. Forbes, Johan U. Backström, R. Bhushan Gopaluni
       IFAC World Congress. 2020.
       [Link] [arXiv]
   11. Almost Surely Stable Deep Dynamics
       Nathan P. Lawrence, Philip D. Loewen, Michael G. Forbes, Johan U. Backström, R. Bhushan Gopaluni
       NeurIPS. 2020.
       [Link] [arXiv] | NeurIPS Spotlight
   12. Modern machine learning tools for monitoring and control of industrial processes: A survey
       R. Bhushan Gopaluni, Aditya Tulsyan, Benoit Chachuat, Biao Huang, Jong Min Lee, Faraz Amjad, Seshu Kumar Damarla, Jong Woo Kim, Nathan P. Lawrence
       IFAC World Congress. 2020.
       [Link] [arXiv]
3. Theses
   1. Deep reinforcement learning agents for industrial control system design
      Nathan P. Lawrence
      The University of British Columbia. 2023.
      [Link]
   2. Convex and nonconvex optimization techniques for the constrained Fermat-Torricelli problem
      Nathan Lawrence
      Portland State University. 2016.
      [Link]