David Doermann
Papers
5
Total Citations
59
H-Index
4
About
David Doermann is a leading researcher at the intersection of computer vision, robotics, and human-machine teaming. His work primarily focuses on human motion prediction (HMP) and multi-robot systems, with a strong emphasis on enabling machines to understand, anticipate, and collaborate with humans in dynamic environments. A key contribution is the development of **PIMNet**, a physics-infused neural network for human motion prediction that integrates physical constraints into deep learning models, achieving more realistic and robust pose forecasts. This work has garnered 26 citations and represents a significant step toward safer human-robot interaction. Doermann also pioneers the study of **human-swarm teaming**, using physiological measurements to analyze tactical decision-making, and has developed scalable algorithms like **SCoPP** for multi-robot coverage path planning in non-convex areas. His research on jointly forecasting human action and pose addresses the critical challenge of predicting *what* a person will do and *how* they will do it, with applications in assisted living and co-robotics. With over 60 citations across his most-cited works, Doermann’s contributions are shaping the future of autonomous systems that can seamlessly integrate with human teams.
Research Focus
Key Achievements
Top Papers
- 1PIMNet: Physics-Infused Neural Network for Human Motion Prediction26 citations · 2022
- 2Learning Robot Swarm Tactics over Complex Adversarial Environments14 citations · 2021
- 3
- 4
- 5What and How? Jointly Forecasting Human Action and Pose2 citations · 2021