Max Robotics — Global Service Robot Directory + US Certification

About

Bernard Manderick is a pioneering figure in the fields of evolvable hardware (EHW) and reinforcement learning, best known for his foundational work on hardware that can adapt and reconfigure itself using genetic algorithms. His seminal 1996 paper, "Evolvable hardware with genetic learning" (78 citations), introduced the concept of building hardware on programmable logic devices whose architecture evolves through genetic learning to adapt to new environments. This work, along with his research on applying EHW to pattern recognition and fault-tolerant systems (74 citations), established key principles for self-adaptive hardware. Manderick also made significant contributions to reinforcement learning in complex, large-scale environments, notably applying Q-learning and Bayesian networks to simulated robotic soccer—a domain characterized by large state spaces and incomplete information. His research on "Reinforcement Learning in Large State Spaces" (16 citations) and agent modeling with Bayesian networks (3 citations) advanced the practical application of learning algorithms in autonomous systems. With a career spanning hardware evolution and intelligent agent control, Manderick’s work has influenced both theoretical foundations and real-world implementations in adaptive robotics and fault-tolerant computing.