Payam Nikdel
Papers
7
Total Citations
251
H-Index
6
About
Payam Nikdel is a leading researcher in human-robot interaction, specializing in autonomous navigation, motion prediction, and deep reinforcement learning. His most impactful contribution is the development of relational graph learning for crowd navigation, which enables robots to reason about interactions between multiple agents using Graph Convolutional Networks and model-based planning, earning over 146 citations. Nikdel has also pioneered the challenging problem of “following in front,” where an autonomous robot stays ahead of a walking user—a task far more complex than traditional follow-behind approaches. His work on the Hands-Free Push-Cart and the LBGP framework combines deep RL with classical trajectory planning to achieve this, demonstrating practical applications for assistive robotics. More recently, he introduced STPOTR, a non-autoregressive transformer architecture that simultaneously predicts human trajectory and pose, enabling robots to anticipate motion with high accuracy and speed. His DMMGAN model further advances multi-modal human motion prediction using attention-based GANs. With over 250 total citations, Nikdel’s research bridges the gap between theoretical AI and real-world robotic systems, making him a key figure in developing socially aware, autonomous robots.
Research Focus
Key Achievements
Top Papers
- 1Relational Graph Learning for Crowd Navigation146 citations · 2020
- 2
- 3
- 4LBGP: Learning Based Goal Planning for Autonomous Following in Front15 citations · 2021
- 5
- 6Relational Graph Learning for Crowd Navigation9 citations · 2019
- 7