Affective social touch is increasingly recognized as a critical factor in Human-Robot Interaction, enabling robots to interpret and respond to human emotions. Related to this concept are haptic gestures, which involve both tactile contact with the robot’s surface and the robot’s physical manipulation. These gestures encompass touch gestures, where direct contact is made, and inertial gestures, which involve movements that alter the robot’s orientation or position. This work focuses on inertial gestures as a unique channel for conveying social and affective communication through motion. This research presents the development of an inertial gesture recognition system aimed at identifying and interpreting affective social gestures through the movement of a robotic platform. Using the data provided by an Inertial Measurement Unit, our approach enables precise classification of these gestures, employing the features derived from inertial data alone. Preliminary results demonstrate robust performance, with the system achieving an F-Score of 0.971 in classifying four inertial gestures.