Humanoid Database
Disclaimer: All content presented on this site, including robot profiles, images, and descriptions, is provided for editorial, educational and informational purposes only. Humanoid Press does not claim ownership of any third-party trademarks, designs, or intellectual property featured herein. All rights belong to their respective owners.
Gallop (ASUS)
|
type: |
type 1 |
|
{{variant.name}}:
|
{{opt.name}}
{{opt.name}}
|
SUMMARY:
Gallop is ASUS's first humanoid robot, developed by the company's Humanoid Robotics Business Unit within the Physical AI Solutions Business Group. Officially unveiled in November 2026, Gallop is designed specifically for real-world manufacturing environments, where it can perform repetitive, dexterous assembly and packaging tasks using existing human-designed workstations.
Unlike many humanoid projects focused on demonstrations, Gallop was introduced directly on an active ASUS electronics manufacturing line. Its first assignment involves packaging motherboard accessories, including handling thin plastic bags and delicate components that are challenging for conventional industrial automation systems.
Powered by NVIDIA Jetson AGX Thor and ASUS-developed Physical AI software, Gallop learns from experienced workers through visual demonstrations, simulation training, and VR-guided teleoperation. The robot uses a modular four-layer AI architecture that separates planning, supervision, motion skills, and machine control, enabling rapid upgrades as AI models improve.
ASUS plans customer pilot programs in 2027 and is targeting broader commercial availability during the second half of 2028. Gallop represents the company's effort to bring Physical AI directly into production environments where human workstations, tools, and procedures already exist.
EDITORIAL:
Gallop stands out because ASUS deliberately avoided the usual robotics showcase strategy. Rather than prioritizing dancing, running, or entertainment demonstrations, the company deployed its robot directly onto a manufacturing task with measurable business value. This approach reflects ASUS's belief that Physical AI must be proven in practical environments before it can become commercially relevant.
The motherboard-packaging workstation chosen for Gallop is intentionally difficult. The task involves handling numerous small accessories packaged in thin transparent bags that challenge traditional vision systems and robotic grippers. By targeting one of the most manipulation-intensive operations in electronics manufacturing, ASUS is testing whether humanoids can realistically integrate into existing factories without requiring major facility redesigns.
Equally important is the robot's learning methodology. Gallop is trained by observing skilled workers, generating variations in simulation, and refining performance through VR teleoperation before transitioning to autonomous operation. This creates a continuous learning pipeline that mirrors how humans acquire industrial skills and aligns with broader trends in embodied AI development.
While Gallop remains a prototype platform, ASUS possesses a unique advantage: access to its own manufacturing facilities for large-scale testing. If the company achieves its performance targets, Gallop could become an early example of a commercially viable humanoid specifically optimized for electronics manufacturing and factory automation.
SPECIFICATIONS:
Dimensions & Build
Height: Not published
Weight: Not published
Design Intent: Industrial humanoid for electronics manufacturing and workstation automation
Configuration: Dual-arm humanoid manipulation platform
Current Deployment: Fixed-base manufacturing workstation configuration
Mobility
Current Locomotion: Stationary fixed-base operation
Future Variant: Wheeled mobile platform under development
Primary Focus: Workstation-based manipulation and assembly tasks
Not published: Navigation performance metrics
Degrees of Freedom
Dual articulated robotic arms
Five-finger dexterous hands
Industrial manipulation architecture
Not officially published: complete joint-level DoF breakdown
Actuation
Actuators: Combination of ASUS-designed systems and industrial components
Custom tactile dexterous hand development
High-precision manipulation for electronics manufacturing tasks
Safe motion control with immediate stop functionality
Manipulation
Primary Task: Packaging motherboard accessories
Target Throughput: Approximately 20 seconds per motherboard kit
Target Success Rate: 99.5%
Capabilities: Pick-and-place operations, packaging, object handling, multi-step workflows
Handles flexible and transparent packaged components
Perception & Sensors
Vision-based perception system
Wrist-mounted cameras for manipulation guidance
Head-mounted sensor visor
Tactile sensing integrated into dexterous hands
Multimodal observation and learning capabilities
Compute & AI
Compute Platform: NVIDIA Jetson AGX Thor
Learning Method: Human demonstrations, simulation, and VR teleoperation
AI Architecture:
Planning Layer
AI Supervision Layer
Motion Skills Layer
Control Layer
Modular architecture enables AI-model replacement without redesigning hardware
Battery & Power
Power specifications: Not published
Battery specifications: Not published
Operational endurance: Not published
Connectivity
VR teleoperation support
Digital-twin and simulation integration
Image: @SectionA-A-AI
Compnay Source:
https://press.asus.com/blog/factory-proven-physical-ai-asus-gallop-humanoid-robot/