Papers
6
Total Citations
29
H-Index
3
About
Junbo Tan is a robotics and control systems researcher whose work spans autonomous robot navigation, fault-tolerant control, and reinforcement learning-based robot manipulation. His research addresses some of the most demanding challenges in modern robotics, combining theoretical rigor with practical application across diverse platforms. Tan's most-cited contribution introduces a hybrid trajectory optimization framework for articulated tracked robots, enabling autonomous terrain traversal that reduces operator cognitive load — a significant advance for deploying robots in complex, unstructured environments. Complementing this, his work on active fault-tolerant control integrates neural network-based fault diagnosis with reinforcement learning, offering robotic manipulators greater resilience against actuator failures. His earlier research on multi-sensor switching strategies using linear-parameter-varying models further demonstrates a sustained commitment to robust, reliable robotic systems. Beyond manipulation and locomotion, Tan has explored medical robotics through a visuotactile-sensor-driven pneumatic device for oropharyngeal swab sampling, highlighting his versatility. His more recent investigations into offline goal-conditioned reinforcement learning and behavior cloning for continuum space robots reflect a forward-looking focus on data-efficient, safety-critical control. With citations accumulating across multiple research threads, Tan is emerging as a thoughtful contributor bridging classical control theory with modern machine learning in robotics.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3
- 4
- 5
- 6