Kaushik Balakrishnan
Papers
1
Total Citations
6
H-Index
1
About
Kaushik Balakrishnan is a researcher at the forefront of embodied AI and autonomous navigation, with a focus on bridging the gap between simulation and real-world robotic performance. His work centers on reinforcement learning and perception-driven control, particularly for indoor robot navigation using RGB-D sensors. In his highly cited 2021 paper, “An A* Curriculum Approach to Reinforcement Learning for RGBD Indoor Robot Navigation,” Balakrishnan introduced a novel curriculum learning strategy that integrates classical A* path-planning with deep reinforcement learning, enabling robots to navigate complex, photo-realistic environments like Habitat more efficiently. This contribution has garnered 6 citations and is recognized for its practical approach to solving the confluence of mapping, localization, and control. Balakrishnan’s research is notable for its emphasis on scalable, sim-to-real transfer, making his work directly relevant to the development of autonomous systems for homes, warehouses, and assistive robotics. His achievements reflect a deep commitment to advancing intelligent navigation, and his findings continue to inspire new directions in curriculum-based RL for embodied agents.
Research Focus
Key Achievements
Top Papers
- 1