Duro Inertial Position Accuracy

[Steve Wells]

Hi,

   Does Duro Inertial still maintain RTK centimeter-level accuracy during GNSS outage or in challenging environments?  For how long will it maintain centimeter-level accuracy?

Thank you,

Steve

[Dennis Zollo]

Dear Steve,

You'll notice on the product summary that we included a table of drop-out times vs position accuracy on page 2 to address this very question: https://www.swiftnav.com/sites/default/files/duro_inertial_product_summary.pdf

This table was generated in real test conditions by actively disabling the GNSS antenna while driving on a test vehicle in relatively open sky conditions, so it represent ideal conditions for the accuracy during a dropout.

Ultimately, the accuracy during the "Dead Reckoning" position mode is determined by the GNNS positioning accuracy immediately before the dropout, and the physical limits of the relatively low cost MEMS accelerometer and Gyro that we use on Piksi Multi. If you have RTK fixed or high quality RTK Float positioning before a dropout, you should maintain high accuracy positioning during short dropouts.  In very challenging GNSS environments like urban canyons, you should see a smoother, more robust, and more available positioning solution, but the absolute accuracy and centimeter-level position availability is limited of course by GNSS performance during periods of better skyview.

In the product, we were targeting maintaining sub-meter positioning for short dropouts (i.e. less than 5 seconds) due to overpasses, canopies, trees, barns, and buildings.  This product is not aimed at maintaining sub-meter navigation for very long dropouts like parking garages, long tunnels, or very high-rise urban areas like Manhattan or Tokyo, but rather to smooth and improve outdoor positioning with occasional, short excursions into GNSS denied or very GNSS challenged environments.  Imagine maintaining centimeter level accuracy under a highway overpass or short highway tunnel, or while a tractor turns around at the end of a row under tall trees; these are the use cases that we have targeted and for which we have tried to optimized performance.  

Best,

Dennis