K. Rathbone

Papers

Total Citations

H-Index

About

K. Rathbone’s research focuses on the intersection of evolutionary computation, neural networks, and robotics, with a particular emphasis on lifelong learning and adaptive control. Their major contributions lie in developing controllers that enable robotic arms to adapt continuously to changing environments, such as equipment drift or repositioning, without requiring recalibration. In their 2002 work, “Evolving lifelong learners for a visually guided arm,” Rathbone pioneered a fast-learning dynamic controller that combined neural networks with evolutionary methods for uncalibrated visual guidance. This was extended in 2003 with “Evolving robot arm controllers for continued adaptation,” where they employed genetic algorithms to evolve neural network controllers capable of robust, ongoing adaptation to real-world disturbances. Though their citation counts are modest—3 and 2 respectively—these papers represent foundational steps toward autonomous, self-improving robotic systems. Rathbone’s work is notable for its forward-looking approach to lifelong machine learning, anticipating later developments in continual adaptation and embodied AI. Their research remains relevant for students and engineers interested in evolutionary robotics, adaptive control, and the challenge of building machines that learn throughout their operational lifetime.