
- Arm Count
- 4.0
- Payload Capacity
- 1.5 kg
- Teleoperation method (reported)
- Leader-follower kinesthetic teleoperation (operator back-drives 2 leader arms; 2 follower arms mirror in real time), with simultaneous mobile-base driving, per the Mobile ALOHA architecture.
- Mobile Base
- Yes
Overview
Cobot Magic is AgileX Robotics' productized implementation of Stanford's Mobile ALOHA whole-body bimanual manipulation system. An operator physically back-drives the two leader arms and the follower arms mirror the motion in real time (leader-follower teleoperation), while the wheeled base can be driven simultaneously — enabling collection of coordinated whole-body demonstration data. The two follower arms and mobile base form the deployable robot; the two leader arms are the teleoperation input. The system ships with the Mobile ALOHA / ACT (ACT++) imitation-learning codebase adapted by AgileX and a ROS integration layer. It exists in a full mobile version (Tracer base, catalogued here) and a split version (Ranger Mini 3.0 base + docking station + Livox Mid-360 LiDAR).
Highlights
- Productized Mobile ALOHA: four 6-DoF AgileX PiPER arms in a 2-leader + 2-follower teleoperation rig
- Simultaneous whole-body teleoperation — arms plus AgileX Tracer mobile base — for demonstration data collection
- Three Orbbec Dabai RGB-D cameras (2 wrist + 1 top) for imitation-learning perception
- Ships with the open-source Mobile ALOHA / ACT imitation-learning stack and ROS integration
- Available as a full mobile version (Tracer base) and a split version (Ranger Mini 3.0 + Livox Mid-360 LiDAR)
Specifications
Configuration
- Arm Count
- 4.0
- Teleoperation method (reported)
- Leader-follower kinesthetic teleoperation (operator back-drives 2 leader arms; 2 follower arms mirror in real time), with simultaneous mobile-base driving, per the Mobile ALOHA architecture.
- Mobile Base
- Yes
- Leader Arms Included
- Yes
Per-Arm Specs
- Payload Capacity
- 1.5 kg
- Arm Degrees of Freedom (per arm)
- 6.0
- Maximum Reach
- 626.75 mm
- Repeatability
- 0.1 mm
Perception
- RGB-D Camera Count
- 3.0
Software & Ecosystem
- Onboard Compute
- Industrial-grade computer; configurations documented as NVIDIA Jetson Orin Nano Developer Kit (8GB) or an Industrial PC (Intel i7-9700, NVIDIA RTX 4060, 32GB RAM), Ubuntu 20.04 / CUDA 11.3.
- ROS/ROS 2 Integration
- Yes
- Open-Source Assets Included
- Mobile ALOHA imitation-learning framework (ACT / ACT++ policy codebase) adapted by AgileX, plus a ROS-based teleoperation and data-collection stack (Stanford Mobile ALOHA lineage).