Dmitriy Miloserdov

St. Petersburg Institute for Informatics and Automation

Papers

Total Citations

H-Index

About

Dmitriy Miloserdov is a researcher at the intersection of artificial intelligence, robotics, and smart urban systems. His primary research areas include recurrent neural network architectures, time-series forecasting, and continuous learning for autonomous systems. Miloserdov’s most notable contribution is his work on urban traffic flow forecasting using recurrent neural networks with spiral structures of layers—a novel architectural innovation that improves prediction accuracy in complex, dynamic environments. This paper has garnered 56 citations, reflecting its significance in the field of intelligent transportation systems. He also explores event forecasting for robots with continuous training, aiming to equip autonomous agents with the ability to anticipate and adapt to changing environments in real time. This work addresses a critical challenge in robotics: enabling machines to operate safely and effectively in unpredictable settings. Miloserdov’s research bridges theoretical advances in neural network design with practical applications in smart cities and robotics, offering promising pathways toward more intelligent, responsive autonomous systems. His contributions are particularly relevant for students and researchers interested in deep learning for spatiotemporal data and lifelong learning in robotics.