Nader Motee
Papers
7
Total Citations
35
H-Index
4
About
Nader Motee is a robotics and systems researcher whose work spans autonomous navigation, multi-robot path planning, and machine learning for perception. His research is characterized by a rigorous mathematical foundation, drawing on optimization theory, information theory, and deep learning to solve complex real-world robotics challenges. Among his most notable contributions is his work on sparse visual feature selection for robot localization, where he exploited the linear structure of information matrices to enable fast, computationally efficient landmark selection — a problem critical for real-time robot navigation. This work, accumulating over a dozen citations, has practical implications for both single and multi-robot systems operating in unknown environments. His 2010 research on multi-robot path planning introduced an elegant duality-based optimization framework for collision and obstacle avoidance under curvature constraints, establishing foundational methods still referenced in the field. More recently, Motee has pushed into deep reinforcement learning for active perception, proposing layered architectures that enable robots to classify environments under partial observability. He has also contributed robustness analysis frameworks for recurrent neural networks, addressing reliability concerns for AI deployed in sequential decision-making tasks. Collectively, his work bridges classical control theory with modern machine learning, offering principled solutions to scalable, intelligent autonomous systems.
Research Focus
Key Achievements
Top Papers
- 1Estimation With Fast Feature Selection in Robot Visual Navigation11 citations · 2020
- 2A duality approach to path planning for multiple robots7 citations · 2010
- 3
- 4Path planning for multiple robots: An alternative duality approach4 citations · 2010
- 5
- 6Estimation with Fast Landmark Selection in Robot Visual Navigation2 citations · 2019
- 7