Compact research hardware
A desktop mounting structure and integrated controller support small manipulation workstations.
Atlas A6
Robotics
Project configuration
Six-axis research manipulator
Discover the detailsSystem overview
Built around
your research.
Start manipulation research with a single arm, then add vision, end effectors, demonstration devices and recording according to the task.
- Degrees of freedom
- 6
- Arm mass
- 4.2 kg
- Repeatability
- +/-0.1 mm
Configuration, lead time and delivery scope confirmed with your inquiry.
Design & capability
Designed for the task.
Research toolchain
Reference materials list Python, ROS 1 / ROS 2, URDF, Gazebo and Isaac Sim support. Confirm drivers and versions.
Separate commands from observations
Record action commands, measured robot state and task outcomes separately. Commands are not execution ground truth.
Specifications
Configuration details and measurement conditions.
| Degrees of freedom | 6 |
|---|---|
| Rated payload | 1.5 kg |
| Arm mass | 4.2 kg |
| Reach | 626 mm |
| Repeatability | +/-0.1 mm |
| Power | DC 24 V |
| Base | 70 x 70 mm, 4 x M5 |
| Installation | Desktop |
| Materials | Aluminum-alloy frame; plastic / resin enclosure |
| Controller | Integrated |
| Joint | Range | Maximum speed |
|---|---|---|
| J1 | -154 to 154 degrees | 180 deg/s |
| J2 | 0 to 195 degrees | 195 deg/s |
| J3 | -175 to 0 degrees | 180 deg/s |
| J4 | -100 to 112 degrees | 225 deg/s |
| J5 | -75 to 75 degrees | 225 deg/s |
| J6 | -170 to 170 degrees | 225 deg/s |
| Temperature | -20 to 50 C |
|---|---|
| Humidity | 25-85%, non-condensing |
| Noise | <60 dB |
| Development | Python, ROS 1 / ROS 2, URDF, Gazebo, Isaac Sim |
| External compute | RTX 4070 / 4080 / 4090 / 5070 / 5080 / 5090 workstation options. Power, cooling, installation and software compatibility require engineering review; high-power GPUs may use an external workstation. |
Delivery and validation
Agree on the complete configuration.
Project integration configuration. Images are concept illustrations, not delivery photographs or engineering CAD. Reference specifications do not establish completed integration, proprietary ownership or brand authorization. Final hardware and performance are subject to project validation and acceptance.
- Confirm the original manufacturer, procurement and branding permissions, then freeze the bill of materials.
- Specify cameras, compute, end effectors, control interfaces, software versions and licenses.
- Verify the workspace, emergency stop, collision controls and communication-loss response before experiments.
- Validate recording, calibration, timing and task performance on the delivered configuration.
Training and deployment
Plan compute for this platform.
Plan GPU compute, model training and hosted inference, with control kept on the robot.
Training & Compute →Related products
Robotics / Project configuration
Atlas A7
Seven-axis lightweight research manipulator
- Redundant kinematics
- 7 axes
- Nominal maximum payload
- 3 kg
- Nominal maximum reach
- 739.45 mm
Robotics / Project configuration
Atlas D
Desktop bimanual learning platform
- Leader-follower setup
- 4 arms
- Reference single-arm payload
- 1.5 kg
- Reference single-arm reach
- 626 mm
Capture / Specification-based configuration
UMI Gripper
Handheld manipulation demonstrations
- Global-shutter video
- 60 FPS
- IMU sampling
- 500 Hz
- Width and GT tactile output
- 100 Hz
Build with Actrium
Start with your task.
Define the devices, data and evaluation your project needs.