Cars can nowadays record several thousands of signals through the CAN bus\ntechnology and potentially provide real-time information on the car, the driver\nand the surrounding environment. This paper proposes a new method for the\nanalysis and classification of driver behavior using a selected subset of CAN\nbus signals, specifically gas pedal position, brake pedal pressure, steering\nwheel angle, steering wheel momentum, velocity, RPM, frontal and lateral\nacceleration. Data has been collected in a completely uncontrolled experiment,\nwhere 64 people drove 10 cars for or a total of over 2000 driving trips without\nany type of pre-determined driving instruction on a wide variety of road\nscenarios. We propose an unsupervised learning technique that clusters drivers\nin different groups, and offers a validation method to test the robustness of\nclustering in a wide range of experimental settings. The minimal amount of data\nneeded to preserve robust driver clustering is also computed. The presented\nstudy provides a new methodology for near-real-time classification of driver\nbehavior in uncontrolled environments.\n