Avinash Kumar
Papers
1
Total Citations
45
H-Index
1
About
Avinash Kumar is a leading researcher in robot control, with a particular focus on bridging the gap between classical feedback methods and modern machine learning. His most influential work, "Residual Reinforcement Learning for Robot Control" (2019, 45 citations), addresses a critical challenge in manufacturing: controlling robots during contact-rich tasks involving friction and complex dynamics. Kumar’s key contribution is demonstrating that reinforcement learning can be layered on top of conventional feedback controllers, allowing robots to learn residual corrections for behaviors that are difficult to model analytically. This hybrid approach preserves the efficiency of traditional control while gaining the adaptability of learning-based methods. His work has significant implications for automating assembly, polishing, and other high-precision industrial tasks. By showing how to combine the best of both worlds, Kumar has opened new pathways for deploying robots in unstructured environments. His research continues to influence both the control theory and robotics communities, making him a notable figure in advancing practical, deployable robot intelligence.
Research Focus
Key Achievements
Top Papers
- 1Residual Reinforcement Learning for Robot Control45 citations · 2019