In this session of The ROS Developers Podcast, I chat with Sam Pfeiffer, who is a PhD student at University of Technology, Sydney where he leads the team of the University that goes to the Robocup with the Pepper robot. Among other things, he will explain us how he ROSified Pepper for that competition. He will also talk about how other participants of the competition are using Deep Learning to win, and how you can do that with ROS. Finally, he will recommend us some useful tools for programming with ROS, like for example a tool that he coded himself that allows to identify the ROS messages used by a remote ROS system, and create automatically a definition for the ones that your system does not have, so you can talk to the remote ROS system.

Selected quote from the interview

I took the decision of not trusting the wifi of the competition, so we run the whole ROS inside the Pepper robot.

Related links

• Follow Sam on Twitter
• Sam Pfeiffer’s github
• Sam’s ROSCON2016 presentation of Dynamic Dynamic Reconfigure
• Sam’s ROSCON2017 presentation of Stealing ROS messages
• Dynamic Dynamic Reconfigure tool
• Steal ROS messages tool
• Discourse channel of ROS
• Github of Sam’s lab at University of Technology, Sydney
• Webpage of Robocup2017 Korean team AUPAIR (winners, using deep learning)
• Inception network
• Open Pose – ROS
• Dense Cap
• Integrated Perception for Service RObots

Suggested code editors for coding with ROS

• Sublime (Python and C++)
• PyCharm (Python)
• QTCreator (C++)
• ROS Development Studio (if you want to code in your browser from Windows, Linux or Mac without installing anything)

Subscribe to the podcast using any of the following methods

• ROS Developers Podcast on iTunes
• ROS Developers Podcast on Stitcher
• ROS Developers Podcast on iVoox

Podcast: Play in new window | Download | Embed

SUBSCRIBE NOW: RSS