Bo Dong

Papers

Total Citations

H-Index

About

Bo Dong is a researcher whose work spans the intersections of machine learning, signal processing, and robotics. His research focuses on two primary domains: intelligent audio scene analysis and robotic perception systems. In the area of audio classification, Dong has contributed to the development of efficient architectures for real-time audio scene recognition, with applications in smart sensing platforms including autonomous vehicles, medical monitoring, and surveillance systems. His 2020 paper, "At the Speed of Sound: Efficient Audio Scene Classification," introduced a retrieval-based architecture combining recurrent neural networks with attention mechanisms to generate robust embeddings for rapid environmental audio recognition, accumulating 5 citations. Complementing this, his work in robotic visual servoing addresses the challenging problem of real-time pose estimation, proposing the use of Smooth Variable Structure Filters as a more stable alternative to the widely used Extended Kalman Filter, which can suffer from instability under nonlinear conditions. This 2018 contribution has garnered 2 citations. Together, Dong's research reflects a broader commitment to making intelligent systems more reliable, efficient, and capable of perceiving and responding to complex real-world environments.