News

Hiromitsu Awano gave a tutorial on the first day of the IEEE Asian Solid-State Circuits Conference (A-SSCC 2026), held November 1–4, 2026 in Taichung, Taiwan.

Robots are moving from hand-written programs to learning from human demonstrations and choosing their next moves directly from what their cameras and sensors see. This trend, often called Physical AI, is also a hardware problem, since a robot must sense, think, and act within a few tens of milliseconds on very little power. Aimed at circuit and architecture designers who are new to robotics, the tutorial started from a simple map of a robot's software, explained imitation learning step by step up to today's vision-language-action (VLA) models, and reviewed recent FPGA and ASIC accelerators for the action generator together with the open problems that remain.