An Improved Multi-Classifier Road Detection Applied to Navigation of Automatic Land Vehicle through Remote Calling
Szu Yu Shen, Rong Chin Lo · Advanced materials research · 2011
This paper proposes a technique of calling ALV remotely. Firstly, sending the GPS information of the target to the ALV by wireless network, then path planning is completed by prepared map data. If the target is changing in navigation, a new path to the terminal point will be to re-plan. AdaBoost algorithm [1] is introduced for implementingroad detection; different types of classifiers are trained to detect different road surfaces. In conclusion, the ALV is able to determine driving directions on different road surfaces.