Ankur Deka

Carnegie Mellon University, Intel (United States)

Papers

4

Total Citations

18

H-Index

3

About

Ankur Deka is a robotics researcher whose work bridges multi-agent systems, haptic manipulation, and cloud-based robot learning. His key research areas include leader-follower navigation, contact-rich manipulation, and scalable robotic fleets. Deka’s major contributions focus on enhancing resilience and adaptability in multi-robot teams. In his most-cited work, “Hiding Leader’s Identity in Leader-Follower Navigation through Multi-Agent Reinforcement Learning” (8 citations), he developed algorithms that conceal which robot is the leader, protecting mission-critical information from adversaries. He further explored human and AI performance in identifying leaders (“Human vs. Deep Neural Network Performance at a Leader Identification Task,” 4 citations), advancing swarm security. Deka also pioneered haptics-based object insertion policies in “Zero-Shot Transfer of Haptics-Based Object Insertion Policies” (5 citations), enabling robots to exploit tactile feedback for contact-rich tasks without retraining. Most recently, he introduced OpenBot-Fleet (1 citation), an open-source cloud robotics system that uses smartphones for collective learning with real robots, democratizing multi-robot research. With a growing citation impact, Deka’s work is shaping secure, tactile, and scalable robotic systems for real-world deployment.

Research Focus

Key Achievements

3
H-Index
4
Papers
18
Total Citations
5
Avg Citations/Paper
🏆 Most Cited Paper
Hiding Leader’s Identity in Leader-Follower Navigation
\nthrough Multi-Agent Reinforcement Learning
8 citations · 2021
📈 Most Prolific Year: 2021 (2 Papers)
🤝 Key Collaborators: 12
🏛 Institutions: Carnegie Mellon University, Intel (United States)

Top Papers

  1. 1
  2. 2
  3. 3
  4. 4

Key Collaborators

Contact & Links

Available for collaboration
Content generated · 13 days ago