3D point clouds acquired with lidars are an important source of data for the classification of outdoor environments by autonomous terrestrial robots. We propose a two-layer classification model. The first layer consists of a Gaussian mixture model. This model is determined in a training step in an unsupervised manner, and classifies into a large set of classes. The second layer consists of a grouping of these classes. This grouping is determined by an expert during the training step, and leads to a smaller set of classes that are interpretable in a considered target task. Because the first layer relies on unsupervised learning, manual labelling of data is not required. Supervision is only necessary for the second layer, and in this case is assisted by the classes provided by the first layer. The evaluation is done for two datasets acquired with different lidars and possessing different characteristics. It is done quantitatively using one of the datasets, and qualitatively using another. The system design follows a standard learning procedure with a training, a validation and a test steps. The operation follows a standard classification pipeline. The system is simple, with no requirement of pre-processing or post-processing stages. Note to practitioners. The classification model is a predictive model and can be used to classify new data. An implementation of the approach would consist in: (a) data acquisition; (b) composition of the learning datasets; (c) feature extraction, unsupervised training and supervised grouping for a few different systems to be tested; (d) validation consisting of a qualitative, visual inspection of the results of the tested systems; (e) selection of the system which performed the best; (f) runtime operation with the selected system. Applications of our system include terrain traversability analysis, rapid production of an operational semantic model in a case of search and rescue, reference in a comparison of different classification systems, labelling of a dataset, or equivalently, the production of a ground-truth.