Arth Shukla
Papers
2
Total Citations
16
H-Index
2
About
Arth Shukla is an emerging researcher at the forefront of embodied AI and robotics simulation, with a focused research agenda centered on scalable robot learning and sim-to-real transfer. His most notable contribution is ManiSkill3, an open-source GPU-parallelized robotics simulation and rendering framework designed to overcome critical limitations in existing platforms — particularly their narrow task coverage and poor scalability for generalizable robotics research. By harnessing GPU parallelization, ManiSkill3 enables compute-scalable approaches to robot learning that were previously impractical, supporting a broader range of scenes, tasks, and sim2real pipelines. This work has rapidly gained traction in the research community, accumulating 13 citations shortly after its 2025 demonstration paper, reflecting strong interest from both academia and industry in scalable robot learning infrastructure. Shukla's research addresses a foundational bottleneck in embodied AI: the need for fast, flexible, and realistic simulation environments that can train generalizable robot policies at scale. For students and researchers working in robot learning, reinforcement learning, or embodied intelligence, his work on ManiSkill3 represents an important open-source resource and a meaningful step toward practical, deployable robotic systems.
Research Focus
Key Achievements
Top Papers
- 1
- 2