This is Beom-Jin Lee, and I was the team leader for AUPAIR at RoboCup@Home SSPL held in Nagoya 2017.
I am emailing to inform RoboCup@Home researchers that we have finally open- sourced our AUPAIR package for everyone’s use.
The AUPAIR package is a Perception-Action-Learning System (PALs) which is a multi-modular system incorporated with state-of-the-art deep learning techniques.
By using the PALs for pepper, you can demonstrate all the functions we used for the RoboCup@Home SSPL competition.
I hope this contribution may ease some difficulties for research groups intending to participate in the future Social Standard Platform Leagues.
We tried to add detail descriptions for any beginners to understand the code, however, if anyone has questions, feel free to contact me.