This contribution presents an approach to automatically classify crop plants and weed based on multi-sensor information with the aim of mechanically removing the weed by an agricul-tural robot. First, a possible sensor setup and its calibration are outlined. As the robot moves forward, the sensor data can be aggregated into a 3D model. To increase model quality, the roll and pitch angles of the robot have to be estimated and compensated. Then, different fea-tures are computed from the sensor data and the discriminative power of the features is evaluated. The feature vector is input to a support vector machine classifier. The considered classes are crop plant, weed, and soil. The result is a 3D representation of plants and weed which can be used for automatic weed removal.
As an example application, tree nurseries are considered, especially the growth of boxwood trees. Classification and mapping results on real data acquired by a robot are reported.