How is Machine Learning Helping Build Space Rockets
Blog #10 : How
is Machine Learning Helping Build Space Rockets
Name :
Kathleen 蔣慧敏
Student
ID : D0726920
Not only healthcare industries that
are making a good use of big data analytics but manufacturing industries are also
starting to apply big data analytics in finding solutions to their problems. In
this article, big data analytics helps in building space rockets.
Finding Stability of Rocket Engine
Researchers from the
University of Texas at Austin are developing a machine learning-based framework
that blends in scientific computing through a combination of physics modelling
and data-driven learning. These will be used to build simulations in order to
make experiment with different design parameters in a fraction of time. These
models can help speed up the designing process and save a huge amount of time
for the design engineers. Their purpose is looking for the most stable design
for the rocket’s engine.
Tuning A Rocket Engine with RL
A team of engineers from Insights Fellow program uses reinforcement learning (RL) approaches in order to
improve their engines. From experiencing trial and error, the team has been
applying RL method to create a solution which gives result that is faster than
using traditional methods.
Using 3D Printing Combined with AI
The engineers at Relativity Space have
made a good use of ML to monitor the manufacturing of propellant tanks and
other large objects. They also made large metal bodies using 3D printing
techniques with AI which is never done by any manufacturing companies before.
In fact, Relativity Space
is the first company that has integrated the advantages of intelligent
robotics, software, and proprietary metal 3D printing technology to automate
aerospace manufacturing. This 3D printing technology use a laser welds together
layers of ultra-fine stainless steel dust as its method of printing.
First of all, I feel interested with Kathleen’s article choice because of NASA launched the SpaceX rocket last week, which is at the end of May and I’ve followed the live video of SpaceX from the beginning until it has launched to the outer space so I decided to make a comment in this blog. I'm amazed at how technology has promptly grown this past few years, especially on how human can build such huge transportation like space rockets. And in the process of making this huge thing, big data, AI and machine learning took a big part in it. It indicates that big data, AI and machine learning could possess all kinds of sectors. From this article, we know that these emerging technologies have made everything possible, easier, and faster. Researchers in University of Texas have developed technology using machine learning. After that, they build simulation called reduced-order-models or ROM to accelerate rocket designing process. It does help the design engineers a lot also it reduces some cost. Other than that, some engineers also found that these technologies can help in choosing better engines. They believe that it will take only several months of trial and error. And eventually, they can bring the rocket to the launchpad earlier than usual. Also, ML and AI has been combined with 3D printing techniques to manufacture the large metal of aerospace bodies. Engineers can tell the printer what to do and these vast robotic arms will move gracefully to sculpting the metal. I think that this was an appealing innovation because it’s new, unique and will save engineers tremendous amount of time.
ReplyDelete