Kadhiravan Umasankar
Papers
1
Total Citations
26
H-Index
1
About
Kadhiravan Umasankar’s research bridges the critical gap between ground and aerial robot navigation, enabling autonomous systems to traverse highly constrained environments with unprecedented speed and reactivity. His most-cited work, “From Agile Ground to Aerial Navigation: Learning from Learned Hallucination” (2021, 26 citations), introduces a self-supervised Learning from Learned Hallucination (LfLH) method that allows robots to generate and learn from imagined trajectories, effectively teaching themselves to navigate cluttered spaces without human intervention. This paradigm-shifting approach extends the Learning from Hallucination (LfH) framework from ground vehicles to agile aerial platforms, demonstrating that robots can internalize safe motion patterns through simulated experience. Umasankar’s contributions are particularly notable for their practical impact: his methods enable drones and ground robots to execute complex maneuvers in real-time, avoiding obstacles while maintaining high speeds. By eliminating the need for extensive human-labeled training data, his work accelerates the deployment of autonomous systems in search-and-rescue, inspection, and logistics applications. With a growing citation footprint and a focus on self-supervised learning for motion planning, Umasankar is establishing himself as a rising voice in robotics, pushing the boundaries of what autonomous navigation can achieve in the physical world.
Research Focus
Key Achievements
Top Papers
- 1From Agile Ground to Aerial Navigation: Learning from Learned Hallucination26 citations · 2021