Section: New Results

Laser based road obstacle classification

Participants : Pierre Merdrignac, Evangeline Pollard, Oyunchimeg Shagdar, Fawzi Nashashibi.

Vehicle and pedestrian collisions often result in fatality to the vulnerable road users, indicating a strong need of technologies to protect such vulnerable road users. Laser sensors have been extensively used for moving obstacles detection and tracking. Laser impacts are produced by reflection on these obstacles which suggest that more information is available for their classification. This year, we introduced the design of a new system for road obstacles classification that is divided in four parts: definition of geometric features, selection of the best features, multi-class segment classification based on Support Vector Machines (SVM) and track classification from SVM decision values integration. Our study discloses a sorted list of useful features for road obstacle recognition that were used to construct a multi-class SVM. Finally, we tested our system with 2D and 3D laser sequences and shown that it can successfully estimate the class of some road obstacles around the vehicle.