James R. Slagle
University of Minnesota, Government of the United States of America
Papers
5
Total Citations
49
H-Index
4
About
James R. Slagle is a pioneering researcher in robotics and connectionist learning, whose work has significantly advanced autonomous control systems. His primary research areas include reinforcement learning, neural network-based control, and real-world robotic applications. Slagle’s major contributions center on developing rapid, unsupervised connectionist learning algorithms that enable robots to master complex tasks, such as backing a vehicle with multiple trailers—a notoriously challenging control problem. His 2002 paper on this topic, which has garnered 25 citations, demonstrated a system that could form useful two-dimensional mappings quickly on an autonomous mini-robot, overcoming severe constraints in computing power, memory, and battery life. This work built on earlier studies, including a 2002 paper with 14 citations that applied similar learning to a real robot. Slagle also explored visionary concepts, such as an underwater naval robot in a 1980 paper, showcasing his forward-thinking approach to robotics. His integrated connectionist methods for reinforcement learning, detailed in papers from 1998 and 2000, have laid foundational groundwork for efficient, on-board robotic control. Slagle’s research remains influential for students and engineers seeking to bridge theoretical learning algorithms with practical, resource-limited robotic systems.
Research Focus
Key Achievements
Top Papers
- 1
- 2Fast connectionist learning for trailer backing using a real robot14 citations · 2002
- 3THE PROSPECT OF AN UNDER WA TER NA VAL ROBOT4 citations · 1980
- 4
- 5Connectionist reinforcement learning for control of robotic systems2 citations · 1998