Az | EN

ASUS Is Teaching Its Gallop Humanoid Robot to Work in Factories

Cəmil Hüseynzadə
16 November 2026 10:49
122 views
ASUS Is Teaching Its Gallop Humanoid Robot to Work in Factories

Taiwanese technology company ASUS is developing a humanoid robot called Gallop to automate repetitive tasks in electronics manufacturing. Customer pilot programmes are planned for 2027, while commercial sales are targeted for the second half of 2028.

The robot is currently being tested at an electronics manufacturing services (EMS) factory belonging to the ASUS group. There, it is learning to package motherboard accessories at a workstation designed for human employees.

According to ASUS, Gallop is being developed to demonstrate that humanoid robots can consistently perform real factory operations rather than simply complete scripted demonstrations.

Robot Learns to Package 15 Types of Accessories

Gallop’s first task is to package 15 different motherboard accessories, most of which are small, thin items contained in transparent plastic bags. Their varied shapes and flexible packaging present challenges for conventional robotic grippers.

ASUS selected this workstation to test whether Gallop can take responsibility for most of an operation without requiring a human worker to remain alongside it throughout the process.

The company has developed a training system combining human demonstrations, simulation and hands-on practice. Experienced factory workers voluntarily record how they perform specific operations using small cameras mounted on their heads. Gallop uses this footage to learn the movements and gripping techniques required for the tasks.

Simulation software then generates variations of the demonstrations to help the robot handle minor differences in real working conditions.

Virtual reality (VR) is also part of the training process. An operator wearing a headset and gloves guides Gallop through tasks using the same components and cameras found at the actual workstation. Once the robot is deployed, its recorded experiences will also feed back into the training process.

ASUS has set a target of achieving a 99.5 per cent task success rate within 20 seconds, equivalent to the time a human worker would need to complete the same operation.

Gallop's AI Architecture Has Four Layers

ASUS has divided Gallop’s AI architecture into four layers: planning, supervision, skills and control.

The planning layer interprets the surrounding environment and breaks an operation down into individual steps. A separate supervisory AI monitors these actions and can update the plan when something goes wrong. The skills layer manages movements such as reaching for, grasping and placing objects, while the control layer operates the robot’s arms and hands.

The architecture uses fixed interfaces between its different layers. This allows ASUS to replace or upgrade AI models without redesigning the entire robot. Alongside standard robotics software, the system runs on a single NVIDIA Jetson AGX Thor computer embedded inside Gallop.

ASUS is developing the robot’s body and dexterous hands in-house. The hands are particularly important for handling small components and use tactile sensing to detect physical contact. Other components, including the motors, are sourced from established industry partners.

Gallop currently operates on a fixed base with two arms. ASUS plans to introduce a wheeled version next year and integrate its internally developed dexterous hand. The company also intends to demonstrate the robot to customers at Computex 2027, ahead of its planned pilot deployments.

Electronics Manufacturing Is the First Target Market

Although ASUS sees potential applications for Gallop in hospitals and other service environments, electronics manufacturing will be its first target market. The company believes skills learned at one workstation can be transferred to other operations involving similar pick-and-place tasks.

ASUS says Gallop is intended to take over repetitive and physically demanding tasks rather than replace factory workers. Employees involved in training the robot could instead move into roles covering robot deployment, supervision and performance improvement.

The Gallop programme began in early 2026. ASUS is targeting customer pilot programmes in 2027 and a commercial launch in the second half of 2028. However, the robot’s ability to meet its factory performance targets remains a key part of the development process.

© copyright 2022 | tech.az | info@tech.az