Optimum boresight procedure

This guide will cover an optimum workflow to get the best results out of the OxTS boresight calibration.

For a more in-depth explanation of the boresight procedure please see this guide, 

https://support.oxts.com/hc/en-us/articles/360011661759-Performing-a-Boresight-Calibration

Data collection workflow 

The data collection should look like the following:

Start the dataset with a good initialisation, driving in a straight line.

100 percent of the 10-minute boresight should be in RTKInteger position mode, and the rest of the dataset should also be in RTK as much as possible; this can be done in post with RINEX files. 

Perform a 15-minute warmup with the following manoeuvres:

After this, the unit should have achieved optimum measurement accuracies,

• Position accuracy < 2 cm
• Heading accuracy < 0.1 degree
• Pitch/Roll accuracy < 0.05 degree  

These can also be checked in post-process, by opening the NCOM in NAVGraph. 

Now you can start logging LiDAR data and perform the boresight, driving around the two retro-reflective boards, for example (blue lines are the boards):

The objective is to scan the targets from as many different angles and ranges as possible. 

The distance from the targets should stay within ~50m. 

Then, perform a warm-down with the same manoeuvres as the warm-up, finishing the data collection in a straight line - this is necessary when processing in Combined mode. 

NCOM Processing

Process the RD file in Combined mode to get the best data possible.

If the unit does not have an improved configuration, this can be done in post-process, by processing the RD file in Simulated mode, clicking "Review configuration" and then "Improve configuration". 

Point NAVSolve to the Simulated NCOM from the data collection and an improved configuration will be calculated, then click save.

Then process in Combined (Forwards+Backwards) mode. This NCOM will then be used in Georeferencer. 

Boresight Processing

Load the NCOM, LIP+LIR and PCAP file into Georeferencer. 

Before processing the boresight, the following filters are recommended in "Processing options":

The minimum range should be increased to remove any features of the vehicle that may be included in the point-cloud, i.e a car bonnet or parts of the roof. The minimum elevation angle can also be increased especially if the car is very reflective. 

In "Boresight options",

It is highly recommended to use the Target dimensions option, by measuring the size of the retro-reflective boards and inputting the measurements into Georeferencer. 

Reflectivity threshold can be increased to remove reflective features that are not the boards, such as wet surfaces or street signs. 

It is recommended to repeat the boresight procedure at least three times by doing the following,

• Only "Optimise LIR"
• Only "Optimise LIP" 
• Only "Optimise LIR" 

Georeferencer may struggle to optimise the vertical axis of the LIP, so this may need to be reset to the physical measured value after optimising the LIP. 

The first page of the boresight will look like this, 

Click "Set Position" and select the two boards in the cluster plot. 

Then adjust the Search radius to fit around the boards, making sure to exclude any reflective features that are not the boards. 

Click "Continue Boresight". A good boresight path will look the following plot,

I.e. lots of hits in as many azimuthal angles as possible. 

A boresight processing folder should be produced in the location of the NCOM, within this folder will be a "mobile_targets.laz" file. 

This will be a point cloud of just the retro-reflective targets. A good LIP/LIR and NCOM will produce targets in the point cloud that are thin and square. 

The LIR may need to be optimised several times to produce this. 

Contact OxTS

Contact OxTS for information on troubleshooting a boresight calibration, obtaining survey-grade INS devices or to share your boresighting data with us or test some of our data sets.

• support@oxts.com for technical support with performing a boresight calibration
• sales@oxts.com for sales enquiries
• info@oxts.com for data sets or information