A Robust Feature Matching Method for Robot Localization in a Dynamic Indoor Environment - Technologies and Applications of Artificial Intelligence - page 359

Experiment 1 will test
criterion of our move track
one location after moving 1

t our method in a static environment. The terminat
king method is 15. That is, if the robot cannot converge
15 times, the algorithm will terminate and report fail.

tion
e to

4.1 Experiment 1: The Static Environment

a . b.

c. d.

e. f.

Fig. 12. Ro obot movement in the feature matching process

An example of move tra
each movement will filter
point. Fig. 13 shows the a
tests, each test with a rand
higher for a higher resolutio

acking method is shown in Fig. 12, from (a) to (f) wh
out some candidates and eventually converge to a corr
average accuracy of finding correct location, out of 10
dom position in the map. We can find that the accuracy
on (smaller distance between points).

here
rect
000
y is