VSLAM ACCURACY AND DRAGONFLY PRECISION

Dragonfly, our Visual SLAM technology, provides high level of accuracy.

Dragonfly is a Visual SLAM technology that uses computer vision to provide the precise location of vehicles, robots, drones, forklifts and moving assets. Dragonfly provides a high level of accuracy. For this technology there are two different types of accuracy and precision to consider.

PRECISION FACTORS

Drift

The drift is the accumulated error.

The drift applies to both monocular, stereo and RGB-D installations, though there are some differences. 

The drift, when using a stereo camera, can be immediately calculated. Instead, for monocular cameras, measuring the drift is harder since the scale of the environment is unknown beforehand.

The drift typically applies to “unknown” environments, meaning venues and spaces where Dragonfly has not been used before, and for which there is no existing 3D map created by Dragonfly, yet.

For example, the first time you drive a forklift with a camera and Dragonfly installed, inside a warehouse, you will be exploring an unknown environment.

The drift inside known environments, instead, corresponds to the difference between the accumulated error, corrected with loop closing, and the ground-truth. 

The drift is expressed as a percentage: this number indicates the amount of accumulated error, and can be considered as the amount of meters of error for each 100 linear meters.

  • Stereo-camera installations: Dragonfly’s drift has been verified to range between 0.6% and 1.3%. This means that if you drive a forklift with Dragonfly installed on board along a 100 meters linear path, at the end Dragonfly will report a location that can be 60 cm to 130 cm inaccurate.
  • Monocular installations: there are other ways to measure the accuracy: the Range of Confidence (ROC). In this case is not the result of the accumulated error over time. In any case, several tests in real life environments demonstrated that it is possible to approximate the accuracy of monocular cameras inside an unknown environment to 1%. Empirically, inside a venue of 12,000 sqm the initial drift of the monocular camera is 1.9% but this gets corrected over time with loop closings.

Do you want to know more about Dragonfly?

The drift and the error for monocular cameras’ systems are however automatically corrected by Dragonfly each time there is a loop-closing.

 

This means that each time Dragonfly recognizes an area that has already been mapped, the loop closes, and Dragonfly corrects the location and the map is also updated.

 

Loop-closings are extremely useful to improve the overall accuracy of the system, and it is strongly recommended to perform frequent loop-closings during the initial mapping, for both monocular and stereo cameras architecture.

PRECISION FACTORS

Loop Closing

PRECISION FACTORS

Linear accuracy

When navigating inside a known environment, Dragonfly re-locates the device inside a previously created map.

The accuracy in this case depends on the precision of the triangulation of known points (features).

The linear accuracy, the radius of confidence (ROC), is typically 5-10 cm, and depends on several factors, including:

  • The quality of the camera;
  • The lightning of the environment;
  • The real-world reference;
  • The dimension of the map;
  • The camera’s calibration.
 

As an example, imagine a drone navigating inside a hangar that has already been “mapped” before: in this case, the ROC of the drone will be 5-10 cm.

Do you want to know more about Dragonfly?

Are you ready to integrate the Dragonfly engine?

Get your Dragonfly Stramer now

Questions? Connect with one of our experts​!

We would love to hear from you and clarify all your doubts.

THANK YOU FOR YOUR MESSAGE.

One of our experts will connect with you as soon as possible.

Our average response time is 24 hours.

Let us help you.

Explain your requirements, and our experts will be happy to connect with you.

Don't miss our updates!

Subscribe to our newsletter and make sure to be the first to receive news and updates. Beat the competitors, stay one step ahead.

We don't spam, it's a promise.