Srinath Mahankali
Papers
1
Total Citations
6
H-Index
1
About
Srinath Mahankali is a robotics researcher whose work centers on the intersection of reinforcement learning and agile locomotion, with a particular focus on optimizing the energy efficiency of quadrupedal robots. His most-cited paper, "Maximizing Quadruped Velocity by Minimizing Energy" (2024, 6 citations), introduces a novel approach that rethinks traditional reward-shaping in RL. Instead of relying on complex, handcrafted reward terms to guide algorithms like Proximal Policy Optimization (PPO), Mahankali demonstrates that directly minimizing energy consumption can paradoxically lead to higher velocities, simplifying the training process while achieving superior performance. This contribution challenges conventional wisdom in robot locomotion, offering a more elegant and computationally efficient pathway to high-speed movement. His work has implications for reducing the engineering burden in deploying RL for real-world robots, making agile skills more accessible. As an emerging voice in the field, Mahankali’s research is paving the way for more intuitive and effective control strategies in legged robotics.
Research Focus
Key Achievements
Top Papers
- 1Maximizing Quadruped Velocity by Minimizing Energy6 citations · 2024