Anandsingh Chauhan

Anandsingh Chauhan

PhD student, Mechanical and Aerospace Engineering, Arizona State University.

I am a PhD student in Mechanical and Aerospace Engineering at Arizona State University, advised by Prof. Kunal Garg in the Safe Robotics Lab. I work on learning-based control with safety guarantees: making learned controllers satisfy hard constraints reliably when they are deployed, not only on average.

My current work combines control barrier functions with sequence models. BarrierFormer (CoRL 2026) enforces barrier constraints over a predicted horizon, so the policy acts before a violation occurs and needs no online optimization at inference. FIND-R locates safety failures inside neural controllers by probing their internal representations and repairs the faulty component directly. I am now extending these ideas to multi-robot teams that operate under limited and delayed communication.

Before my PhD, I spent five years at TCS Research in Mumbai applying reinforcement learning to safety-critical infrastructure, including power grids under adversarial attack, multi-robot task allocation, and railway scheduling.

News