Rohan Smith

Papers

Total Citations

H-Index

About

Rohan Smith is a leading researcher in robotic vision and active scene understanding, with a focus on bridging the gap between simulation and real-world robotics. His major contributions center on developing standardized benchmarking tools that allow researchers to evaluate robotic perception systems consistently across both photorealistic 3D simulations and physical robot platforms. Smith created BenchBot, a software suite that provides a unified interface for testing sensorimotor capabilities, and the BEAR platform, which offers high-fidelity simulated environments for active scene understanding research. These tools have been instrumental in advancing how robots explore environments and identify objects for human interaction. His work on the Robotic Vision Scene Understanding Challenge has further standardized evaluation in this critical area, enabling reproducible comparisons across different approaches. With over 19 citations across his most influential papers, Smith's contributions are shaping the future of robotic perception. His platforms are now used by research groups worldwide to accelerate progress in autonomous navigation and object recognition, making him a key figure in the movement toward more robust, real-world-ready robotic systems.