Scholar
Junhong Xu
Google Scholar ID: wBvO3LIAAAAJ
Indiana University
Citations & Impact
Citations
331
H-index
9
i10-index
9
Publications
20
Co-authors
0
Contact
Publications
Resume (English only)
Academic Achievements
- October 2024: One paper accepted by CoRL 2024, enabling generating realistic and challenging environments for training generalizable robot navigation policies.
- October 2024: Work on using the diffusion model to imagine unknown regions to provide more informative context for the downstream planner is accepted by IROS 2024.
- July 2024: Published a new blog post on how Nuro.ai combines safe RL and imitation learning for self-driving.
- April 2024: Boundary-Aware Value Function Generation for Safe Stochastic Motion Planning got accepted by the International Journal of Robotics Research (IJRR).
- January 2024: One paper accepted by the International Journal of Robotics Research (IJRR), proposing a novel kernel-based approach for solving stochastic optimal control problems and its application to autonomous navigation on unstructured terrains.
- October 2023: A new arXiv paper on enhancing the sampling-based MPC algorithm (MPPI) with uncertainty propagation.
- October 2023: Work on combining informative prior policies trained by goal-conditioned RL with the bounded-rational game-theoretic framework is accepted by IROS 2023.
- August 2023: Successfully defended PhD dissertation titled “Robust Motion Planning and Control for Autonomous Robots Under Uncertainty”.
- October 2022: One paper on learning a causal model of the robot’s dynamics is accepted by ICRA 2023.
- September 2022: Paper Decision-Making Among Bounded Rational Agents on explicit modeling the computational limits in multi-agent motion planning using information-theoretic bounded rationality is accepted by DARS 2022.
Research Experience
- Working on safe reinforcement learning methods and generative models at Nuro.ai; focused on leveraging model uncertainty to improve deployment time robustness during doctoral research, and developed planning algorithms that adopt conservative behaviors in high-uncertainty regions; also worked on generating diverse and trainable environments for reinforcement learning, aiming for deployment time generalization over environment distributions.
Education
- PhD from Indiana University, Bloomington, where he worked with Lantao Liu in the Vehicle Autonomy and Intelligence Lab.
Background
- Currently at Nuro.ai working on safe reinforcement learning methods and generative models for addressing challenging, safety-critical problems in autonomous driving. Research focuses on designing methods that enhance the safety and efficiency of robotic systems at deployment time.
Miscellany
- Powered by Jekyll with al-folio theme. Hosted by GitHub Pages. Photos from Unsplash. Last updated: February 13, 2025.
Co-authors