Helena Project Update 25

Week 8

Project Update 25

• If you were to redo the project from Fall term, what would you tell yourself?

If I were to redo the project from Fall term I would have us divide up the work differently. I would also want to be better about creating timelines for the whole team (including the ME's and EE's) so that we could understand our dependencies better from the beginning. We were making a lot of it up as we went along since OSU hasn't ever made a rocketry payload for space. I also would have requested that we remove one degree of freedom from the arm since it would have made things a lot easier, but the challenge has been fun.

• What's the biggest skill you've learned?

I have learned a lot about configuration spaces, pathfinding, and have been able to wrap my head around 4D arrays. I have simultaneously been learning about Machine Learning (I wonder how an AI would feel about us learning to make them learn), so I have been thinking about how to incorporate some ML into this project in the future, especially in reacting to changing environments.

• What skills do you see yourself using in the future?

I think that I will use my space robotics skills in the future. It's a pretty niche skill, but I think Intelligence and Space Research might be the right place to use that. I will, if nothing else, continue to use my embedded C skills to do hardware projects.

• What did you like about the project, and what did you not?

I liked getting to think about pathfinding and 4D arrays, but I did not like dealing with data storage or telemetry, so I was pretty happy when Michael and Amber took the lead on those two things. I did not enjoy working with so many other engineers at first as it could be pretty frustrating when we didn't all know each other very well, but now that is one of my favorite things about it, because we are a team.

• If your project were to be continued next year, what do you think needs to be working on?

I hope that this project will continue next year. There are a lot of improvements to be made. I would like to see the following improvements (in order):

1. Dynamic pathfinding on orbit - requires more processing power or longer experiment duration or both

2. Dynamic targeting on orbit - requires 1.

3. Dynamic CSpace Mapping on orbit - requires data from multiple angles and Inverse Kinematics. requires 1.

4. A second arm - requires 1, 2, 3, and a longer mission, probably on a satellite

5. Tool use with one arm

6. Tool use with two arms interacting with each other

7. Image processing and object recognition

...

I could detail my whole idea for construction in space, but I think that is enough for one night.