Max Robotics — Global Service Robot Directory + US Certification

About

Akinobu Hayashi is a robotics researcher whose work centers on advancing robot manipulation through multi-modal perception, particularly the fusion of vision and tactile sensing. His most impactful contribution, the 2022 paper "VisuoTactile 6D Pose Estimation of an In-Hand Object Using Vision and Tactile Sensor Data" (51 citations), tackles the critical challenge of determining an object's precise 6D pose while it is held in a robot gripper—a scenario where heavy occlusion often defeats purely vision-based methods. By integrating tactile data, Hayashi’s approach enables robust state estimation for in-hand manipulation, directly improving a robot's ability to handle objects dexterously. Beyond this, his research explores how robots can reason under uncertainty. In works like "Reasoning about uncertain parameters and agent behaviors through encoded experiences and belief planning" (2019) and "Online adaptation of uncertain models using neural network priors and partially observable planning" (2019), he develops frameworks for planning and adapting behaviors when models are incomplete or observations are noisy. His earlier work on "Receding horizon optimization of robot motions generated by hierarchical movement primitives" (2014) combines fast, reactive motion primitives with optimal control. Collectively, Hayashi’s research pushes toward more capable, perceptive, and adaptable robotic systems for real-world manipulation tasks.