Re:Kalman filters and functional programming
by taeric on Tuesday January 31, 2017 @11:53AM
(#13532378)
This looks really interesting. The idea of splitting the accumulator from the data seems a clear win. Would love to see evidence confirming it works.
[ Parent
| on HN ]
Re:Kalman filters and functional programming
by davidrupp on Tuesday January 31, 2017 @12:18PM
(#13532728)
Brian Beckman works for Amazon Prime Air (drone delivery service)[1]. I submit the video "First Prime Air Delivery"[2] as circumstantial evidence that it does indeed work.
[1] https://twitter.com/lorentzframe [2] https://www.amazon.com/Amazon-Prime-Air/b?ie=UTF8&node=80377...
Re:Kalman filters and functional programming
by taeric on Tuesday January 31, 2017 @12:32PM
(#13532897)
I'm not sure I follow. Would you let me submit as evidence an old piece of software that is littered with GOTOs all over the place as evidence that that is a great way to write software?
I fully believe this can work. Especially with a talented practitioner. However, the question is if it will lead to more maintainable (e.g., by needing fewer iterations?) and faster (e.g., by lots of people using it) output with fewer bugs (e.g., by surviving the lindy effect).
That is, my question is entirely empirical and will hinge on repeated success. Not initial success. I am personally hopeful for it, but I have reasons not to trust my hopes.