Anuj Pokhrel
Papers
4
Total Citations
31
H-Index
3
About
Anuj Pokhrel is a robotics researcher pushing the boundaries of off-road autonomy, with a focus on enabling wheeled robots to navigate the most extreme and vertically challenging terrain. His work centers on kinodynamic modeling, self-supervised learning, and multi-modal perception, addressing the fundamental challenge of how robots can safely and efficiently traverse environments like rugged boulders and low-light landscapes. His major contributions include CAHSOR, a competence-aware planning framework that allows high-speed navigation in SE(3) space, and Terrain-Attentive Learning, which models 6-DoF vehicle dynamics on difficult terrain. With his self-supervised approach, VertiCoder, he has demonstrated that a single pre-trained model can handle multiple downstream tasks, from forward to inverse kinodynamics learning. His work on the M2P2 dataset tackles the critical problem of perception in extreme low-light conditions, reducing reliance on active sensors. Collectively, his papers have garnered over 30 citations in just two years, reflecting the immediate relevance of his research. Pokhrel’s work is not only advancing the theoretical foundations of off-road mobility but also providing practical tools and datasets that will enable the next generation of autonomous exploration robots.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3
- 4