This paper presents the clustering and classification of the radar scattering characteristics of vehicles under real-world driving conditions. The classification of 14 distinct vehicle types is achieved through statistical features derived from their radar cross-section (RCS) characteristics, represented as histograms. Various machine learning classification techniques are applied, and their performance is evaluated across different clustering scenarios. The results of the clustering algorithm are in line with the physics-based expectations on the scattering from different vehicle types. The classification results demonstrate the effectiveness of the proposed algorithm, validating the histogram-based feature method as a novel and promising approach for vehicle identification and detection. In addition, the results highlight the potential applications of our methods in millimeter-wave (mmWave) radar technology, illustrating their capability to improve feature extraction by means of RCS histograms and ensure robust classification in diverse and challenging environments.