© Ongrade 2018 Disclaimer
Welcome to the revolution, in this section we aim to provide useful information on many things machine control, both technical and otherwise.

August -September

August- September

We’re mixing things up again this month and looking at the practical side of developing and verifying a machine control system like DigPilot more particulalry the new 3D version of the product. This months contribution is from Jan Floberg the man behind DigPilot and details part of the verification process in the development of DigPilot 3D, which is now available in the UK . This article is unedited from the original so you will see Jan’s English is better than my Norwegian!   

Verification of DigPilot 3D with single and dual GPS/GNSS  

(July 2010) After another 6 months of hard work with our 3D software for DigPilot we are happy to report from our last installation. The accuracy of both the single and dual antenna solution was systematically tested and the results could not have been better.      

DigPolit 3D installed on a Komatsu PC240LC












Tormod Kristensen in Alta founded his new contractor company in Alta 5 years ago. Alta lies well north of the polar circle in Norway, and the machines of Tormod Kristensen Maskin AS are expected to work under the most severe conditions as the climate is harsh this far north: It changes from very hot in the summer down to extreme cold in the winter.   Tormod Kristensen Maskin AS just acquired a brand new Komatsu PC 240, and chose DigPilot 3D excavator guidance system with a dual antenna GPS/GNSS solution from Leica.      

Guttorm Torsetnes and I spent the weekend from 2 – 4 July 2010 at the worksite in Alta to make sure that the excavator was ready to be put to work with DigPilot 3D. I am happy and extremely proud to report to all you gentlemen that the accuracy tests showed extremely good results.      

Normally we are happy when the X and Y coordinates are within 50 mm of the reference points, which means that the horizontal distance from the reference point could be as much as 70 mm since sqrt(5^2 +5^2) = 7.07.      

In all tests, DigPilot performed brilliantly, always keeping well within the required 50 mm for the X and Y coordinate, but mostly being inside just 10 to 20 mm!      

(It must be said that this accuracy test was made easier since the excavator was brand new and had no noticeable play in the bolts and a new firmly anchored rotating tilt bucket).      

Here is a detailed description of what we did:      

Make sure that DigPilot 2D measures accurately  

If the DigPilot does not measure accurately in 2D mode (without GPS), then it is impossible to make the DigPilot measure accurately in 3D mode with GPS. The first thing you need to do is therefore to make sure that DigPilot 2D measures accurately which means that excavator and bucket measurements are correct, and that the angle sensors are correctly calibrated.  Make sure that the antenna and boom offsets with respect to the slew center are correct      

These offsets are critical for the accuracy obtained at the bucket tip and must be measured with millimeter accuracy. We therefore hired a surveyor who made the measurements with a total station as shown below:      

Measuring point nr 1 of 2 of the reference line along the centre of the boom with a total station


Measuring point nr 2 of 2 of the reference line along the centre of the boom with a total station


Measuring the Position of antenna nr 1


Measuring the position of the boom pin

Make reference points around the excavator   

Place the reference points as far away from each other as possible, and as far away from the excavator as possible, but make sure that they can all be reached with the bucket tip without moving the undercarrige. Then measure the position of each reference point. We did this by extending the antenna cabel of antenna nr 1 (master antenna), and placed the GPS/GNSS antenna ontop of the reference points. By doing so you can store the coordinates of the reference points in the DigPilot and then make the DigPilot calculate the measurement errors with respect to these reference points.        

We made 5 reference points around the machine.        

Measuring the coordinates of a reference point and storing it in DigPilot












In addition we also wrote down the coordinates of all the reference points on a sheet of paper.              

Measure the reference points with the bucket tip     

Mark the center of the bucket blade so that you can easily see exactly which point on the bucket to place on the reference points, and then carefully and accurately position it ontop of the reference point. Then save the bucket tip measurement in DigPilot.        

Note that the height is not so important at this stage as the height error can be adjusted afterwards using a reference point on which the bucket tip can be placed without any risk of displacing the reference point. The main objective at this stage is to make sure that the horizontal deviation is as small as possible, and you may therefore position the bucket tip slightly above the reference point to reduce the risk of moving it.        

Measuring the position of a reference point with the bucket tip


Review the results on the DigPilot     

When done we viewed the resulting deviations between the reference points (P1, P2, P3, P4 and P5) and the measurements made with DigPilot by placing the bucket tip on them.        

If the results are not satisfactory the DigPilot also calculates what the deviations would have been if the the boom and antenna offsets are altered without you having to make new measurements by placing the bucket tip on the reference points all over again. In this way you can thus make the final accuracy adjustments directly on the DigPilot.        

Measuring the position of a reference point with the bucket tip


Adjust the height error     

We then adjusted the antenna nr 1 height offset to get rid of the height error. The height that was measured from the boom pin to the top of the antenna was too high as the position outputted from the GPS is with respect to the metal plate at the bottom of the GPS/GNSS antenna.        

Accuracy test when the excavator is rolling and pitching and the bucket is tilted        

The first tests were performed when the excavator was more or less horizontal and the bucket was not tilted sideways. To make sure that the DigPilot was working properly we therefore tilted both the machine and the bucket, and then measured the reference points all over again.        

The X and Y coordinates were still within 50 mm of the reference points, even though placing the bucket tip accurately on the reference points was somewhat more difficult due to the roll angle        

Perform accuracy test with respect to installed project     

Finally we made sure that the LandXML (.xml) project installed in the DigPilot agreed with the stakes of the real project. We therefore tested the accuracy once more by carefully placing the bucket tip on the centerline of the road that had just gotten its first layer of asphalt. The DigPilot informed us that the current height of the was a few centimeters too low. That however corresponded very well since the road was about to get two more layers of asphalt, and we therefore considered the job well done.