About

Amrit Singh Bedi is a robotics and machine learning researcher whose work spans online optimization, multi-robot systems, and language-guided autonomous navigation. His most influential contribution, "Online Learning With Inexact Proximal Online Gradient Descent Algorithms" (2019, 109 citations), established efficient low-complexity optimization frameworks for nondifferentiable, time-varying problems critical to robotics and signal processing. This foundational work shaped subsequent research in tracking moving agents through decentralized multi-robot systems, where his inexact gradient descent methods enabled resource-constrained robots to collaborate on complex tasks like search-and-rescue and intrusion detection. Bedi has made significant strides in distributed robotics, developing ASAPP, the first asynchronous algorithm for distributed pose graph optimization in multi-robot SLAM (45 citations), and pioneering reinforcement learning approaches for warehouse automation through multi-robot task allocation systems like RTAW and DC-MRTA. More recently, his research has ventured into trustworthy human-robot interaction, integrating large language models into navigation systems via TrustNavGPT and LANCAR, addressing uncertainty and context-awareness in unstructured environments. His cumulative body of work reflects a coherent trajectory from theoretical optimization to practical, deployable robotic intelligence, making him a notable voice at the intersection of learning theory and autonomous systems.

Research Focus

Key Achievements

6
H-Index
16
Papers
259
Total Citations
16
Avg Citations/Paper
🏆 Most Cited Paper
Online Learning With Inexact Proximal Online Gradient Descent Algorithms
109 citations · 2019
📈 Most Prolific Year: 2024 (4 Papers)
🤝 Key Collaborators: 31
🏛 Institutions: Indian Institute of Technology Kanpur, DEVCOM Army Research Laboratory, University of Maryland, College Park, Park University, University of Central Florida

Top Papers

  1. 1
  2. 2
  3. 3
  4. 4
  5. 5
  6. 6
  7. 7
  8. 8
  9. 9
  10. 10

Key Collaborators

Contact & Links

Available for collaboration
Content generated · 14 days ago