Max Robotics — Global Service Robot Directory + US Certification

About

Tomoki Isobe is a leading researcher in dexterous robotic manipulation, specializing in multi-fingered in-hand manipulation enhanced by tactile sensing. His work addresses one of robotics' most fundamental challenges: enabling robot hands to manipulate objects with the same dexterity and adaptability as humans, particularly when using low-cost hardware. Isobe's major contributions include pioneering the use of Graph Convolutional Networks (GCNs) to process tactile data from distributed sensors of varying sizes and shapes on multi-fingered hands, achieving stable manipulation across diverse object properties—a breakthrough detailed in his most-cited paper (46 citations, 2022). He further advanced the field by integrating Convolutional Neural Networks (CNNs) with 3-axis tactile sensors to enable robust in-grasp manipulation on low-precision, affordable robot hands (20 citations, 2020), and developed CNN-LSTM architectures for variable in-hand manipulations without task-specific training (14 citations, 2020). Isobe's research is notable for democratizing dexterous manipulation by proving that sophisticated tactile-driven control can be achieved with accessible hardware, significantly reducing the barrier to entry for advanced robotic hands in both research and practical applications.