John Selinsky

Papers

Total Citations

H-Index

About

John Selinsky is a pioneering figure in intelligent control systems, whose work bridges the gap between neural networks and robotics. His research focuses on the intersection of neurocontrollers, adaptive robot control, and the strategic use of a priori knowledge in dynamic systems. Selinsky’s foundational paper, “Neurocontroller design via supervised and unsupervised learning” (1989, 27 citations), established early frameworks for training neural networks to govern complex plant dynamics. His most influential contribution, “The role of a priori knowledge of plant dynamics in neurocontroller design” (2003, 15 citations), introduced a novel architecture that guarantees neurocontroller performance by leveraging known system structures to design exploratory schedules—a critical advance for reliable real-world deployment. In “A learning/adaptive robot controller” (2003, 2 citations), Selinsky further clarified the distinct yet complementary roles of learning (identifying dynamics for long-term tracking) and adaptation (real-time adjustment) in robotics. Though his citation counts reflect a focused, specialist audience, his work remains essential for researchers tackling the challenge of embedding prior knowledge into learning-based control, offering a principled path toward safer, more predictable autonomous systems.