|
Ambihelical Hexnut posted:We're all weak pussies compared to somebody: Is that laser completely unenclosed? 
|
#
¿
Sep 26, 2015 23:31
|
|
|
|
| # ¿ Apr 29, 2024 14:37 |
|
|
Parts Kit posted:Looks like an industrial power one so it's likely in a huge enclosure with a light and camera inside for the video. The Locator posted:You can see at about 1:15 that some sort of a cover is opened to grab the cut part. At 0:45 he pans the camera from the laser to the control display and then back again---the camera is definitely not inside of an enclosure. The camera then stays fixed on the laser without any (apparent) cuts. At 1:15 I think the shadows moving are just from the cutting head being lifted up.
|
|
#
¿
Sep 27, 2015 00:41
|
|
|
DreadLlama posted:I am from a place where highschool textbooks are from 1967. By the time I graduated from university, you could send an "E-mail" (in quotation marks) to your professor (but not in lieu of a written copy) if you wished to prove on-time submission. Yeah, I have a couple "optical flow" sensors for my drones. Optical mice use high speed (but low resolution) camera sensors combined with a digital image processor to sense 2-dimensional movement fairly accurately (high speed specialized hardware can determine how far an image has shifted from the previous frame), They are intended to have lenses (usually a cheap molded polycarbonate piece), so using a bigger lens is fairly simple. As long as the surface has enough contrast (and you're not moving too fast) it will still be able to measure the translational movement. However, these things ended up being not-great for drones, because they are unable to handle rotational movement(rotational transforms are somewhat more complex, and dont go well with a low resolution sensor). Anyway, PID controllers have been around for like 100 years. The math is very simple, and originally they were used in mechanical systems, and then electromechanical systems, and then (analog) electronics, before modern digital PID control came about. Analog PID control is pretty cool IMO, since it involves the use of operational amplifiers (op-amps), which are a special type of amplifier than can be configured to perform various mathematical operations--addition/subtraction, integrating, differentiating and gain are all trivial, so if you can represent your sensor signal(s) and control signal(s) as voltages, you can make an analog PID controller that will function pretty much like a digital one. (Op-amps can also be used to build analog computers, which allow you to perform complicated math and solve differential equations and stuff, often by using logarithmic amplifiers to give you a mathematical shortcut to make doing stuff like multiplying 2 signals together easier. All you need to do is dial in your equation's constants as voltages, wire the computer up correctly for your equation, and then measure an output voltage to get the solution). Slanderer fucked around with this message at 03:43 on Oct 6, 2015 |
|
#
¿
Oct 6, 2015 03:41
|
|