Akshay Thirugnanam
Robotics Research (United States), University of California, Berkeley
Papers
5
Total Citations
150
H-Index
4
About
Akshay Thirugnanam is a roboticist forging new pathways in safety-critical control and autonomous navigation. His research masterfully bridges the gap between formal safety guarantees and agile, real-world performance, focusing on enabling robots—particularly legged systems like quadrupeds—to operate in cluttered, narrow environments. Thirugnanam’s most significant contribution is a novel, duality-based optimization framework that reformulates the complex problem of obstacle avoidance between polytopes into a convex optimization. This breakthrough, detailed in his highly cited 2022 paper (78 citations), allows for real-time, provably safe trajectory planning, a task previously limited to slower, offline methods. He further extended this work to legged locomotion in his 2023 paper (26 citations), introducing Exponential Discrete Control Barrier Functions to enable a quadruped to "walk in narrow spaces" with formal safety guarantees. By also exploring the integration of model-based safety with model-free reinforcement learning (19 citations), Thirugnanam is pioneering a future where robots are both provably safe and highly agile. His work represents a critical step toward deploying autonomous robots in the tight, human-centric spaces of the real world.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3
- 4
- 5