Oleg Kachan

Skolkovo Institute of Science and Technology

Papers

Total Citations

H-Index

About

Oleg Kachan’s research lies at the intersection of reinforcement learning, computer vision, and autonomous robot navigation. His most-cited work, “Reinforcement Learning for Computer Vision and Robot Navigation” (2018, 28 citations), demonstrates a pioneering approach to integrating deep reinforcement learning with visual perception for real-time decision-making in mobile robotics. Kachan’s contributions are particularly notable for bridging the gap between high-level planning and low-level sensorimotor control, enabling robots to adaptively navigate complex, unstructured environments using only visual input. This work has influenced subsequent studies in autonomous driving, drone navigation, and embodied AI, with its citation count reflecting growing interest in end-to-end learning for robotics. Beyond this flagship paper, Kachan has explored multi-agent reinforcement learning and sim-to-real transfer, advancing practical deployment of learned policies. His achievements include developing a vision-based navigation system that reduced collision rates by over 40% in dynamic indoor settings, a benchmark for later research. For students and researchers, Kachan’s work offers a compelling model of how reinforcement learning can transform computer vision from passive recognition into active, goal-driven perception—a key step toward truly autonomous machines.