Learning to Innovate in Space

Somehow, at some time, I enrolled in CICADA's online course "Space Foundations for Elevate 2025". This is introductory start up training for budding space entrepreneurs. The course has 10 modules with a study time of four and a half hours. There are 3 minute videos some readings and quizzes. This is all designed for the student to do alone in their own time, it is complemented by face to face workshops (the first of which as yesterday and I missed). 

Some of the introductory slides juddered on my screen. All I could do is hit "start" and move on. The videos are the usual talking heads with some slides and quizzes. Unfortunately some of the videos don't have closed captions or transcripts, so if you can't hear, you are out of luck. 

There appear to be a generic set of innovation videos, supplemented by ones specifically on the space industry. This can be a little jarring. The entrepreneurial ethos of quick to market and failing fast doesn't necessarily fit with space engineering, where development can take decades and failure can be fatal. 

I managed to get half way through the course. At that point I noticed I had only completed 90% of the first module. There were a few multiple choice quiz questions. I went back and attempted them several times, but could not get past 90%, so at that point gave up. 

Huston, we have a solution!

Greetings from the AI, ML and Friends Seminar at the Australian National University in Canberra, where Dr. Zak Kingston from Rice University is speaking on "Scaling Multi-Modal Planning".

That doesn't sound very exciting, but he is speaking from the USA, where he has been planning how to use a robot on the International Space Station. Robots can help with routine housekeeping tasks, on a space station, and presumably on earth, but it is tedious to have to program the robot to make every tiny motion. That works fine for a production line, where every car body will be locked into the same place, with an area reserved for the robot. But Dr. Kingston explained a space station is a confined space cluttered with stuff, which gets moved about. He is describing the use of the Open Motion Planning Library (OMPL) to solve this problem.

Austral autonomous ship
concepts for USN & RAN.

This work has application for the Australian Defence Force. Suppliers to the ADF are working on ships, and submarines designed to be operated without a crew for weeks. These drone submarines and ships will need routine maintenance while at sea, with no one on board. A robot sailor could save the vessel having return to port to have a minor problem fixed.