Karan Mahesh

PhD Student at MIT MechE & WHOI

Autonomy, Robotics, Controls, Machine Learning

prof_pic.jpg

I am a Ph.D. student in the MIT–MechE/WHOI Joint Program, where I work on multi-agent planning and autonomy. I am advised by Dr. Michael Benjamin in MIT’s Marine Autonomy Laboratory and I am graciously supported by the NSF Graduate Research Fellowship.

Before starting my Ph.D., I was an AI/ML Research Scientist at Aurora Flight Sciences, where I worked on novel multi-agent autonomy and motion planning research, including the DARPA Learning Introspective Control program. I received my B.S. in Mechanical Engineering from UC Berkeley in 2022. As an undergraduate, I conducted research in TAFLab with Dr. Reza Alam, HiPeRLab with Dr. Mark Mueller, and as a visiting research intern with the University of Oxford’s Computational Biology Group.

My broader interests include robotics, control, and machine learning for autonomous systems. More details are available in my CV.

news

Oct 1, 2024 Our paper on safe marine autonomy using adaptive control and reachability analysis has been accepted to IEEE Transactions on Control Systems Technology
May 22, 2024 Talk at Boeing Technical Excellence Conference on Fast Adaptive Learning and Control Online (FALCON), our work for the DARPA LINC program
Jan 29, 2024 Our paper on online safety certification for systems in dynamic environments was accepted to ICRA 2024!
Aug 15, 2022 I started as an AI/ML Research Scientist at Aurora Flight Sciences

publications

  1. Online Data-Driven Safety Certification for Systems Subject to Unknown Disturbances
    Rober, Nicholas*, Mahesh, Karan*,  Paine, Tyler M. and 5 more authors
    ICRA, 2024
  2. Safe Autonomy for Uncrewed Surface Vehicles Using Adaptive Control and Reachability Analysis
    Mahesh, Karan*, Paine, Tyler M.*,  Greene, Max L. and 6 more authors
    IEEE Transactions on Control Systems Technology, 2024