3 ms·
I love these long-form build logs. I just started reading, and I am making the faux-pas of commenting before finishing. But, I'm wondering what the challenges
by fusslo 2y ago
I love these long-form build logs.
I just started reading, and I am making the faux-pas of commenting before finishing.
But, I'm wondering what the challenges are of automating the ROV to map a body of water's floor in a pattern. like a grid pattern, or whatever is most efficient.
At first I was thinking currents would cause displacement. but can't we sense the current moving us in undesirable ways and correct with thrusters?
And then I thought.. do lakes have currents? Do they have tides? can a ROV sense the boundary of a lake?
just further down the rabbit hole, realizing how little I've learned about the natural environment!
- relaxing 2y agoMy brother in christ we cannot even keep a robot on land rolling in a straight line without an external source of ground truth. The way to do it is have a boat with GPS tow your sensor array.
- jarofgreen 2y agoLakes can have currents. I wonder if pedantically speaking the definition of lake would include non-tidal in many countries but .... A) humans use names sloppily and if it's an important detail I wouldn't assume a lake is non-tidal without checking. B) non-tidal bodies of water might still change height over the year, for example after a heavy rainfall. Mainly I'd question the need to automate it. It's difficult, and in many cases the cost of a human to drive it is tiny compared to all the other costs you need to pay so just do that (as in the article - those weren't automated). Also, driving them can be fun :-)
- eythian 2y agoRarely (as far as I know), lakes can be tidal in their own way. For example Lake Wakatipu changes by about 10cm every half an hour.
- tetha 2y agoLübecks university has several projects[1] using swarms of robots for automatic cartography, water measurements and such. Autonomous accurate navigation under water is quite complicated, because after a certain point you need to start relying on local sensors because nothing reaches you anymore. But local sensors tend to be weird, because a straight line underwater is not necessarily a straight line - you are most likely drifting -- and detecting drift isn't easy. From a local observer, the water around you isn't even moving. That was a fun team to talk with. 1: https://www.iti.uni-luebeck.de/en/research-areas/mobile-robotics/maus.html https://www.iti.uni-luebeck.de/en/research-areas/mobile-robo...
- fusslo 2y agoso cool; thank you for the link
- rkangel 2y agoNote that ROV stands for Remotely Operated Vehicle - it is not autonomous, it is controlled by a person. What you are describing would be called a UUV (unmanned underwater vehicle) or AUV (autonomous underwater vehicle).