Max Robotics — Global Service Robot Directory + US Certification

About

Yutaka Maeda is a robotics and computational intelligence researcher whose work sits at the intersection of neural networks, machine learning, and robotic control systems. His research has made notable contributions to solving some of robotics' most persistent challenges: teaching machines to understand their own bodies and movements through data-driven approaches rather than rigid analytical models. Maeda's most recognized contributions center on applying advanced neural network architectures — including complex-valued and quaternion neural networks — to the learning of inverse kinematics and inverse dynamics in robotic manipulators, particularly SCARA robots. His 2014 and 2007 papers on high-dimensional neural network control and SCARA robot learning have each garnered 9 citations, reflecting steady influence in the field. A distinctive thread throughout his work is the use of the simultaneous perturbation optimization method, an efficient gradient-approximation technique that simplifies neural network training with minimal computational overhead. Beyond manipulator control, Maeda has explored fuzzy logic-based visual feedback systems that eliminate the need for complex camera calibration, central pattern generators for rhythmic motion generation, and micro tactile sensor development. Collectively, his portfolio reflects a career dedicated to making intelligent, adaptive robot control more practical and accessible.