英飞凌在SEDEX 2026介绍面向人形机器人灵巧手的MCU、GaN功率器件与触觉传感器方案

Infineon is targeting the market for robotic hands, a critical component of humanoid robots, by leveraging its expertise in industrial and automotive semiconductors. Its key offerings include microcontrollers (MCUs), power semiconductors and tactile sensors.
Sarah Wu, an Infineon representative, introduced the company’s semiconductor solutions for robotics at a session titled “Semiconductors That Build Humanoids,” held alongside the Semiconductor Exhibition (SEDEX) 2026 at COEX in Seoul on Oct. 16.
Humanoid robotic hands must grasp and manipulate objects like human hands. Each finger and joint requires precise, independent control, while gripping force must be adjusted according to the object being handled. Excessive force can break fragile objects, making hands one of the most difficult humanoid components to develop.
Robotic hands require motor control and tactile sensing technologies. Motor-control MCUs and drivers move finger joints, while sensors detect finger positions and the force applied when grasping objects. These are technologies Infineon has supported through its industrial and automotive semiconductor products.
Robotic hand control architectures fall into three categories: centralized, hybrid and fully distributed. Infineon explained that the architecture depends on the hand’s degrees of freedom, which represent the number of joint movements a robot can control independently. Higher degrees of freedom enable movements that more closely resemble those of a human hand. The company said it can support all three architectures.
Tactile sensing relies on magnetic, pressure and capacitive sensors. Three-dimensional (3D) magnetic sensors and pressure sensors installed at the fingertips measure pressure when the hand contacts an object. Capacitive sensors detect finger positions and the force applied to the fingertips. Infineon has adapted technologies previously used in smartphones and home appliances for robotic hands.
The company is also exploring the use of microelectromechanical systems (MEMS) microphones to identify the characteristics of objects touching the hand through sound and vibration.
Infineon also demonstrated a board equipped with an MCU, gallium nitride (GaN) power semiconductors, and power and position sensors capable of driving motors rated at up to 1.2 kilowatts. The board measures approximately 66 millimeters.
GaN enables faster switching and lower power losses than silicon. Because humanoid robots require motors at individual joints, Infineon said GaN technology can help reduce the size of motor-drive circuits.