![](bgc.png)
Table 2. Characteristic values of absolute errors on SPIDAR location in generalization with data
set 1.
Data set 1 Raw NN PF1
mean (mm) 13.23 5.31 7.91
std (mm) 8.41 5.14 7.68
max (mm) 37.60 29.42 32.51
6 Conclusions
In this paper we propose a method to calibrate SPIDAR using a feedforward neural
network coupled with a semi-automatic initialization. The semi-automatic initialization
allows us to place the SPIDAR referential at the same 3D position at each startup with an
accuracy of 1.2 mm. This way, we can use a method for calibrating the SPIDAR which,
don’t need to be updated at each startup. We choose a feedforward neural network
in order to compensate non linear errors on location and their abilities to estimate a
targeted output from a source without any knowledge on the mathematical model. We
obtain good results and our whole calibration procedure is efficient. Testing our neural
network in generalization shows us that our calibration is quite robust, even if we reset
the SPIDAR. We plan to make the initialization procedure fully automatic.
References
1. Kim, S., Ishii, M., Koike, Y., Sato, M.: Development of tension based haptic interface and
possibility of its application to virtual reality. Proceedings of the ACM symposium on Virtual
reality software and technology (2000) 199–205
2. Boudoin, P., Otmane, S., Mallem, M.: Fly over, a 3d interaction technique for navigation in
virtual environments independent from tracking devices. VRIC ’08 (2008) 7–13
3. Nixon, M., McCallum, B., Fright, W., Price, N.: The effects of metals and interfering fields
on electromagnetic trackers. Presence, 7, (1998) 204–218
4. Ghazisaedy, M., Adamczyk, D.: Ultrasonic calibration of a magnetic tracker in a virtual
reality space. Virtual Reality Annual International Symposium (VRAIS’95) (1995)
5. Bryson, S.: Measurement and calibration of static distortion of position data from 3d trackers.
Siggraph’92 (1992)
6. Ikits, M., Brederson, J., Hansen, C.D., Hollerbach, J.M.: An improved calibration framework
for electromagnetic tracking devices. IEEE Virtual Reality Conference 2001 (VR 2001)
(2001)
7. Kindratenko, V.: A survey of electromagnetic position tracker calibration techniques. Virtual
Reality,5, (2000) 169–182
8. Saleh, T., Kindratenko, V., Sherman, W.: On using neural networks to calibrate electro-
magnetic tracking systems. Proceedings of the IEEE virtual reality annual international
symposium (VRAIS ’95) (2000)
9. Ikits, M., Hansen, C., Johnson, C.: A comprehensive calibration and registration procedure
for the visual haptic workbench. Proceedings of the workshop on Virtual environments 2003,
39 (2003) 247–254
10. Leotta, D.: An efficient calibration method for freehand 3-d ultrasound imaging systems.
Ultrasound in medicine & biology (2004)
11. Bishop, C.: Neural networks for pattern recognition. Oxford University Press (2005)
98