David Vengerov
Papers
1
Total Citations
5
H-Index
1
About
David Vengerov is a researcher whose work lies at the intersection of reinforcement learning, autonomous systems, and adaptive control. His key research areas include co-evolutionary algorithms, perception-based decision-making, and sensor allocation for unmanned aerial vehicles (UAVs). In his most-cited paper, "Co-evolutionary perception-based reinforcement learning for sensor allocation in autonomous vehicles" (2004, 5 citations), Vengerov tackles the critical challenge of enabling UAVs to dynamically allocate sensors in uncertain, changing environments. He introduces a novel approach that combines perception-based rules for generalizing decision strategies across similar states with reinforcement learning for real-time adaptation. This work is notable for addressing the "curse of dimensionality" in reinforcement learning by using co-evolution to optimize both the perception mapping and the decision policy simultaneously. Vengerov’s contributions have practical implications for autonomous navigation, robotics, and multi-agent systems, where efficient sensor management is key to performance. Though his citation count is modest, his innovative integration of co-evolution and reinforcement learning offers a valuable framework for researchers working on adaptive autonomy and intelligent control in complex, real-world settings.
Research Focus
Key Achievements
Top Papers
- 1