John Pohovey
Papers
2
Total Citations
39
H-Index
2
About
John Pohovey is a rising researcher at the intersection of robotics, artificial intelligence, and human-robot collaboration. His primary research areas include sampling-based path planning, learning-based motion planning under constraints, and proactive human-robot interaction using large language models (LLMs). Pohovey’s most impactful contribution is his work on “Neural Informed RRT*,” which integrates neural networks with the classic RRT* algorithm to dramatically accelerate convergence in path planning. By leveraging point cloud state representations and admissible ellipsoidal constraints, his method enables more efficient and optimal motion planning in complex environments—a critical advance for autonomous systems operating in the real world. This paper has already garnered 37 citations, signaling strong interest from the robotics community. In parallel, Pohovey is pioneering proactive human-robot collaboration through his “LIT” framework, which uses LLMs and vision-language models to enable robots to anticipate human intentions rather than requiring constant prompting. Demonstrated in a robot sous-chef application, this work reduces cognitive load on human partners and paves the way for more natural, long-horizon teamwork. Pohovey’s dual focus on algorithmic efficiency and intuitive interaction marks him as a promising voice in next-generation autonomous systems.
Research Focus
Key Achievements
Top Papers
- 1
- 2