HAART: The Human-Animal Affective Robot Touch

HAART contains 7 gestures: pat, contact without movement (press), rub, scratch, stroke, tickle and `no touch'. These gestures are the most heavily used of those in Yohanan et al’s Touch Dictionary [2], gathered to communicate emotion in human-animal interaction. For the HAART data set (collected from 10 participants), each touch action was performed on a 10x10 pressure sensor [1] for 10 seconds. To assess feature robustness under realistic operating conditions when installed on a robotic animal model, each participant contributed gestures with the sensor mounted on all permutations of 3 substrate conditions (firm and flat; foam and flat; foam and curve) and 4 cover conditions (none; short minkee; long minkee; synthetic fur). The resulting dataset includes 840 gesture-captures (12 conditions x 7 gestures x 10 participants). Each capture is 10 seconds of a continuously repeated gesture, sampled at 54 Hz and trimmed to the middle 8 sec (432 frames); there are generally 10-15 gesture instances per capture. This dataset is provided as a CSV file and includes the center 8x8 frame (trimmed for consistency with CoST) with pressure values ranging from 0 to 1023. Labels include gesture type, condition, and subject number.


[1] A. Flagg and K. E. MacLean. Affective touch gesture recognition for a furry zoomorphic machine. In Proceedings of the International Conference on Tangible, Embedded and Embodied Interaction (TEI), (Barcelona, Spain), pages 25–32. 2013.

[2] S. Yohanan and K. E. MacLean. The role of affective touch in human-robot interaction: Human intent and expectations in touching the haptic creature. International Journal of Social Robotics, 4(2):163–180, 2012.