Skip to content
Droid/Glove evidence record

MANUS Metagloves Pro

A hand-capture candidate with versioned C++ and ROS 2 documentation. Check the runtime architecture before choosing the capture workstation.

Pro; separate from Pro Haptic

Can a Linux robotics stack use Metagloves Pro without a Windows capture host?

The documented integrated SDK talks directly to the hardware. Remote mode uses MANUS Core, which the 3.0 guide places on Windows. Your architecture determines whether a separate Windows host is needed.

MANUS ↗

What the sources establish

Manufacturer documentation is attributed below. These values are not independent measurements by Droid.

Model and interface facts, with the conditions needed to interpret them.
FieldReported valueConditions and evidence
Tracking approachElectromagnetic hand trackingManufacturer description; no independent accuracy benchmark reproduced.
MANUS ↗
Hand model25 degrees of freedomReported anatomical model; not a count of robot actuators or independently measured sensors.
MANUS ↗
Linux pathC++ SDK: integrated or remoteThe application handles calibration and pairing in integrated mode; remote mode requires MANUS Core on Windows.
MANUS ↗
SDK license scopeRobotics edition lists SDK and ROS 2Confirm the quoted edition, release and subscription or perpetual terms.
MANUS ↗
ROS 2 interfacePackage included with SDK 3.0Publishes glove data; does not itself establish a robot-hand controller.
MANUS ↗
Reported data rate120 Hz default ROS 2 publicationTopic publication rate, not a measured sensor-to-actuator latency.
MANUS ↗
Finger force feedbackNot established in reviewed sourcesThis record covers Pro, not the separately named Pro Haptic model.
Evidence status: unknown

What this means for an integration

Droid’s analysis of the cited documentation:

  1. Start with the runtime decision: integrated mode trades a separate Core host for application responsibility for pairing, calibration and tracker handling. Price that engineering work into the capture stack.
  2. A 120 Hz publisher can repeat or queue measurements. Record device timestamps, arrival timestamps and dropped frames before treating the advertised rate as useful training data throughput.
  3. The Sharpa repository provides a concrete retargeting route to investigate. It is evidence of published integration code, not a Droid-tested deployment or compatibility with every dexterous hand.

Documented relationship

Sharpa publishes a Linux C++ acquisition client and Python retargeting examples for Metagloves Pro and SharpaWave. The repository identifies its purpose as integration and demonstration.

Inspect the Metagloves Pro → SharpaWave evidence →

Still to establish

  • Configured price and the license included in the specific quote
  • Latency distribution under the intended host and network
  • Accuracy under the intended magnetic environment
  • Exact supported hardware, firmware and SDK combination

Before a larger capture run

  1. Capture the same hand pose repeatedly after recalibration.
  2. Compare device time with robot and camera clocks.
  3. Verify the chosen SDK mode and license with the supplier.
  4. Record hardware revision and software versions in every session.

Source record

  1. SDK getting started ↗MANUS · Documentation 3.0.0 · Checked 2026-11-10
  2. Metagloves Pro product page ↗MANUS · Metagloves Pro; undated page · Checked 2026-11-10
  3. Core license types ↗MANUS · Robotics and Motion Capture editions; undated terms overview · Checked 2026-11-10
  4. ROS 2 getting started ↗MANUS · Documentation 3.0.0 · Checked 2026-11-10

Make the next decision.

START WITH THE WORKFLOW

Bring us the task.
Let’s find the fit.

Share your environment, the job to be done, and the outcome you need to measure.

Request a project review