MagicDog-W
Quad Quadruped by MagicLab

MagicDog-W

Moving Fearlessly on Any Terrain — hybrid quadruped-wheeled robot with intelligent posture control.

The MagicDog-W is MagicLab's quadruped-wheeled hybrid robot — combining legged agility with wheeled efficiency. With dual locomotion forms and synchronized intelligent control, it moves fearlessly across any terrain: steps, slopes up to 40°, and obstacles under 60 cm.

Intelligent posture control enables natural handstands, bidirectional movement, and dynamic gait switching. A versatile expansion backplate supports LiDAR, gimbals, and other peripherals. With a full sensor suite including 4K HD camera, depth camera, and fisheye lens, the Dog-W is ready for complex real-world deployments.

Key Features

16+1 Motors
22.5 kg / 10 kg Payload
0–3 m/s
2–4h Battery
Handstand Capable

Hardware Specifications

Standing Dimensions670 × 350 × 650 mm
Prone Dimensions720 × 500 × 290 mm
MaterialAluminum Alloy + High-Strength Engineering Plastic
Tire Type7-inch Pneumatic Tire
Motor Count16 + 1 (Head Motor)
Head Rotation≥ 100°
Max Joint Torque37.5 N·m
Weight22.5 kg (with battery)
PayloadMax 10 kg
Speed0–3 m/s
Obstacle Height< 60 cm
Max Slope≤ 40°
Battery8,200 mAh, 29.6 V
Battery Life2–4 hours
Standby TimeUp to 8 hours
Operating Temp-5°C ~ 40°C

Software & Integration

Computing6 TOPS onboard
Sensors2D LiDAR, Dual Camera, Depth Camera, 4K HD Camera, Fisheye Camera, Ultrasonic
AudioOmnidirectional Microphone + Speaker
TouchTouch sensors — ≥4 touch interaction methods
ConnectivityWiFi 5, 4G Module, Bluetooth 5.0
Expression DisplayFacial expression display
SdkMagicDog SDK + MagicDog-Motion SDK + MagicDog-ROS2 SDK
Ota UpdatesSupported

Application Scenarios

All-Terrain Patrol

Security and inspection patrol across stairs, slopes, indoor and outdoor surfaces — where standard wheeled robots cannot go.

Interactive Deployment

Public-facing applications with touch sensing, facial expression display, and voice interaction.

Research & Development

Open SDK platform for locomotion research, sensor fusion, and embodied AI development.