SESSION 3: 
  
COMBINED POINT DETERMINATION 2 
  
  
CHAIRMAN: H. Ebner (FRG) 
AUTHOR: P. Friess (FRG) 
TITLE: A SIMULATION STUDY ON THE IMPROVEMENT OF AERIAL 
IRIANGULATION BY NAVIGATION DATA 
DISCUSSION: 
Ebner: 
Thank you, Mr. Friess, for this clear presentation 
which was an extension of the corresponding part of 
Prof. Ackermann's paper from yesterday and has shown 
which possibilities are existing in this field. We 
have some minutes for questions and comments. 
Molenaar (the Netherlands): Just one question in your simulation 
Friess: 
Ebner: 
Molenaar: 
- yOu introduced the accuracy of your camera 
positions and the attitudes. Do you also introduce 
in your simulation correlation between these 
quantities? Because it may be that if you introduce 
correlation - you introduce an extra instability 
in the solution of your system. That with corre- 
lation maybe you get other results in accuracy. 
Yes, we didn't consider any correlation between 
observations. 
So far as the working group in general is concerned, 
this is considered as a very important question 
because what we first have to learn is what are the 
accuracy properties of this new data and accuracy 
properties means - you have said it to some extent 
- Some systematic errors which have to be modelled 
and on the other hand the stochastic model of the 
data. The only one which is certain is that your 
assumptions are not really justified in detail but 
I also think that the general findings of your 
simulations will remain valid. If you speak with 
people dealing with navigational data even they 
are not able immediately to answer those questions 
concerning stochastic properties, so it is not so 
easy. 
Maybe one additional remark. What you learn from 
the analysis of connections of blocks to ground 
control can also be transferred to this kind of 
information and I think, what also Prof. Ackermann 
mentioned actually yesterday in his paper, is that 
44