have to make a decision to choose one. Unfortunately a wrong 
decision often happens. Here several typical examples can be 
given. 
Example 1: Location of gross errors 
Fig.1a gives a forward intersection from two known stations. 
If the intersection angle is equal to 100 gon, the coordinates 
of new point is very accurate. But in this case the gross 
error can not be detected. For this reason we must increase 
the observations ( see Fig.1b and 1c ). Both in Fig.1b and 1c 
a single gross error i8 detectable obviously. The difference 
is that in the case of Fig.1b the gross error is not locatable. 
A gross error can be correctly located only in Fig 1c. 
» 
Ns 
A 
NS: 
Fig. 12 
c 
és 
Example 2: Wrong separation betw 
[P 
se 
N^ 
^ 
4 
^. 
| CX 
1b 
Fig. 
- / 
Yu 
Xk 
/ 
| 1 
Tig. 1C 
een gross and systematic errors 
Tab.1 shows a mutual confusion of gross and systematic errors 
in the bundle block adjustment which geometry is given in fig.2. 
Tab.1 
Adjustment Method 
Wrong sera 
errors in the bundl 
With A.P. 
( Ebner,1976 ) 
ration between gross and systematic 
e block adjustment 
Without A.P. 
  
Real Model Errors 
Gross Errors in 
Control Point 1 . 
VX,=VY4= 10 m v 
Systematic Image 
Error 
  
Results 
= = 0.000 
y YY m 
A. P. ba=-54.22 
1,252.69 t 
y. «0,238 m 
1 
wy my" 20 873.29 
  
Wrong Conclusion 
  
There is no blunder _ 
in control point 1, 
bois very significant 
  
There are gross 
errors in control 
point.1. 
  
- 187 -