Jonathan Chang

About

Jonathan Chang is a leading researcher in off-road and agricultural robotics, with a focus on perception, navigation, and human-robot interaction. His seminal work includes the creation of the National Robotics Engineering Center agricultural person-detection dataset, which addressed a critical gap by providing the first large-scale benchmark for pedestrian detection in off-road environments—a paper that has garnered 35 citations and become a foundational resource in the field. Chang also pioneered leader-tracking systems for walking logistics robots, enabling intuitive human-robot collaboration in challenging terrains, and advanced scene understanding algorithms for high-mobility walking robots, allowing them to autonomously navigate complex, unstructured environments. His contributions have directly supported the deployment of robots in agriculture and disaster response, where wheeled vehicles fail. More recently, Chang has explored perception robustness testing, developing frameworks to predict system behavior across diverse conditions, ensuring safety and reliability in real-world deployments. With over 87 citations across his key works, Chang’s research continues to shape the future of autonomous systems in rugged, off-road settings.

Research Focus

Key Achievements

Top Papers

1. 1
   Comparing apples and oranges: Off‐road pedestrian detection on the National Robotics Engineering Center agricultural person‐detection dataset

   35 citations · 2017
2. 2
   Leader tracking for a walking logistics robot

   27 citations · 2015
3. 3
   Scene understanding for a high-mobility walking robot

   22 citations · 2015
4. 4
   Perception Robustness Testing at Different Levels of Generality

   3 citations · 2021

Key Collaborators

Contact & Links