Johannes Doellinger

Papers

Total Citations

H-Index

About

Johannes Doellinger is a leading researcher in socially compliant robot navigation, focusing on how autonomous systems can safely and intuitively share spaces with humans. His work bridges computer vision, pedestrian behavior modeling, and multi-target tracking to enable robots to anticipate and adapt to human motion. Doellinger’s most cited paper, “Predicting Occupancy Distributions of Walking Humans With Convolutional Neural Networks” (2018, 19 citations), introduced a novel deep learning approach that forecasts where pedestrians will be in the near future, directly addressing the challenge of natural motion path understanding for socially aware robots. He further advanced the field with “Environment-Aware Multi-Target Tracking of Pedestrians” (2019, 9 citations), which integrates environmental context—such as obstacles and pathways—into pedestrian tracking systems, significantly improving prediction accuracy in dynamic settings. By combining CNNs with environment-aware tracking, Doellinger has laid critical groundwork for robots that navigate crowded spaces without disrupting human flow. His contributions are foundational for the next generation of service robots, autonomous vehicles, and assistive technologies, making human-robot interaction more seamless and intuitive.