How human does your robot look?

Our Robot Human-Likeness Predictor helps you forecast the extent to which your robot will be perceived as human-like compared to other real-world robots. Give it a try!

How it works

With this tool, you can get a quick estimate of how human-like your robot will be perceived. The Human-likeness score and Dimension scores you receive are directly comparable to the robots in the core collection of the ABOT Database (link to collection). The predicted scores are based on a multiple regression model calculated with data from 1,240 participants. When you are ready to try the predictor click 'start'!

Make sure to have a photo of your robot for reference.


Elizabeth Phillipsa, Xuan Zhaob,c, Daniel Ullmanb, and Bertram F. Mallea,b, “What is human-like?: Decomposing robots' human-like appearance using the Anthropomorphic roBOT (ABOT) Database”, published in the Proceedings of the 11th Annual ACM/IEEE International Conference on Human-Robot Interaction (HRI '18). The predicted scores are based on a multiple regression model with data collected from 1240 participants.

a – Brown University, Humanity-Centered Robotics Initiative
b – Brown University, Department of Cognitive, Linguistic, and Psychological Sciences;
c – The University of Chicago Booth School of Business.