I am a Ph.D. Candidate at the Paul G. Allen School of Computer Science and Engineering at the University of Washington, advised by Byron Boots. My research focuses on developing practical algorithms and systems that enable robots to operate in dynamic real-world environments with minimal human supervision. Specific topics include integrating model-predictive control with reinforcement learning, learning-based methods for accelerated motion planning, and offline reinforcement learning.

Before transferring to University of Washington, I spent a year as a Ph.D. Robotics student at Georgia Institute of Technology, prior to which I was a Robotics Engineer at Near Earth Autonomy, Inc. In 2017, I worked as a researcher at The Air Lab at The Robotics Institute, Carnegie Mellon University. advised by Sanjiban Choudhury and Sebastian Scherer. I received my Master's in Robotic Systems Development from RI in 2016, and my B.Tech in Mechanical Engineering from Indian Institute of Technology, Varanasi.

During the course of my Ph.D, I have had the amazing opportunities to intern and collaborate with various industry research labs. In summer 2022, I was a Research Scientist Intern at Google DeepMind, London in the Controls Team lead by Martin Reidmiller. Prior to that I spent Fall 2020 and Summer 2019 as an intern at NVIDIA Seattle Robotics Lab working with Dieter Fox, Fabio Ramos and Byron Boots.

When not busy with research, you'll catch me doing stand-up and improv comedy or practicing Capoeira.



  • Adversarial Model for Offline Reinforcement Learning, M. Bhardwaj*, T. Xie*, B. Boots, N. Jiang, C. Cheng [Paper]

Journal Publications

  • Lazy SP AuRo

    Leveraging Experience in Lazy Search

    M. Bhardwaj, S. Choudhury, B. Boots, S. Srinivasa

    Autonomous Robots, 2021

    [Paper] [BibTex]
  • Data-Driven Planning IJRR

    Data-driven Planning via Imitation Learning

    S. Choudhury, M. Bhardwaj, S. Arora, A. Kapoor, G. Ranade, S. Scherer and D. Dey

    International Journal on Robotics Research, 2017

    (Finalist for Paper of the Year)

    [Paper] [BibTex]

Conference Publications


    STORM: An Integrated Framework for Fast Joint-Space Model-Predictive Control for Reactive Manipulation

    M. Bhardwaj, B. Sundaralingam, A. Mousavian, N. Ratliff, D. Fox, F. Ramos, B. Boots

    (Selected for Oral Talk - 6% Acceptance)

    [Paper] [BibTex] [Website] [Code]
  • MPQLambda

    Blending MPC & Value Function Approximation for Efficient Reinforcement Learning

    M. Bhardwaj, S. Choudhury, B.Boots

    International Conference on Learning Representations (ICLR), 2021

    [Paper] [BibTex]
  • MPQ

    Information Theoretic Model Predictive Q-Learning

    M. Bhardwaj, A. Handa, D.Fox, B.Boots

    Learning for Dynamics and Control, 2020

    [Paper] [BibTex] [Website]
  • DGPMP2

    Differentiable Gaussian Process Motion Planning

    M. Bhardwaj, B. Boots and M. Mukadam

    International Conference on Robotics and Automation, 2020

    [Paper] [BibTex] [Website]
  • LazySP

    Leveraging Experience in Lazy Search

    M. Bhardwaj, S. Choudhury, B. Boots and S. Srinivasa

    Robotics:Science and Systems, 2019

    [Paper] [BibTex] [Talk]
  • SaIL

    Learning Heuristic Search via Imitation

    M. Bhardwaj, S. Choudhury and S. Scherer

    Conference on Robot Learning (CoRL), 2017

    [Paper] [BibTex] [Website] [Talk]
  • VisServo

    Real-time dynamic singularity avoidance while visual servoing of a dual-arm space robot

    P. Mithun, V.V. Anurag, M. Bhardwaj, S.V. Shah

    Advances in Robotics (AIR), 2015