HN user
Mechanical Engineering graduate student studying robotics
Looks good for robotics applications. The power density is comparable to the low end of brushless motors, and it can use hydro-carbon fuels which are about 25 times more energy dense than current batteries. At 3 hp, it might be a reasonable drop in for the type of weed-whacker two-stroke that drives big dog.
You've made some good points in this thread about actuation. Can you point me to more info on SCHAFT's actuation strategy? I've been trying to turn up info, but haven't found anything great.
I have a plan for overhauling/replacing Geometric Dimensioning and Tolerancing. The way it's done now consumes way too much design time, and honestly is a bit silly considering the inputs and outputs of the system.
I'd like to talk to someone with experience interfacing with factories in China about how the current approach could be replaced. Any suggestions?
Also, if I developed this tool-chain, what is the best way to attach to the market?
Yep, and it makes sense for Tesla to do this. It's a new product, and the information they collect from maintenance and inspections is going to be invaluable for future revs.
There was no difference between the control experiment and fish exposed to parts made with FDM.
Read the article to see how FDM could become toxic/harmful, turns out it didn't.
Very hype heavy article. No mention of Form Labs who is already on the market with a similar product.
So there was a DARPA initiative to do stuff like this back in the 90's/early 2000's:
http://archive.darpa.mil/darpatech99/Presentations/dsopdf/ds...
Turns out it's easy in controlled environments, and doesn't work at all outside. Once the animal sees something it wants to eat/mate with, it's just going to leave.
The claimed contribution of this work is direct ganglial stimulation as opposed to the antennal stimulation which was used previously. But I'm not seeing anything new in terms of capability/new science, and their data collection methods seem hinky:
The controllability of a roach, in terms of success rate, was subjectively rated based on user observation. If the user was able to successfully have the cockroach turn in a desired direction, then a run was deemed successful.
After the single lethal attack in 1945: "Several Japanese civilians have visited ... to offer their apologies for the deaths that took place here, and several cherry trees have been planted around the monument as a symbol of peace" [1]
I was looking for this workflow for my research, and ended up going with an ipython notebook.
Ah, well here is where I could use some advice. I disagree with the way they have gone about their hardware both in the sensing electronics and the hive box itself. I don't want to go into it here; I'd feel a bit rude criticizing someone else's work in a public forum.
Should you approach someone for collaboration if you don't agree with how they're going about the project?
Weight is another good one, it drives our complexity and cost up but is hugely important. Most extant hive monitoring tools record weight. NASA uses some of these measures to track mass flow rates of nectar.
It's hard to say with CCD being so large and varied. One thing we know is that beehives exposed to certain pesticides [1] are much more likely to collapse.
[1] http://www.hsph.harvard.edu/news/press-releases/study-streng...
We certainly have a microphone. There is such a richness in the sounds a colony produces. Our minimal sensing package also contains temperature and humidity sensing. We're shooting for collecting a labeled data set this coming season to see what we can pull out of the information we collect.
One of my side projects is building a sensing electronics package to monitor the health and activity of beehives. The idea is to make it easier for someone just starting to maintain their hive.
Any one have feedback on the concept? Or features we haven't thought of?
Okay... but actual animals do way better:
http://www.ted.com/talks/robert_full_the_secrets_of_nature_s...
As a rule of thumb, any walking robot that uses servos can't approach the performance of its animal colleagues. The leg impedance is way too high.
Neat. Looking forward to it.
This looks great! I'd like to use it to ease up the tech overhead for some mobile robotics applications. Is there anyway that the user's position in the world frame can be accessed? i.e. can this do SLAM over longish distances?
Basically I'm looking for Google's Project Tango, without the hardware constraints.
So this has popped up on HN before, and again I would caution that this is snake oil. The sparse technical details they give are misleading, and they still haven't shown that their fiber reinforced parts are anything special.
Kinda disappointed the total emphasis is on him being a "hell of a good pilot" instead of the control engineers who allowed the plane to compensate for losing a wing. There's no way that plane would have stayed in the air without their work.
Where did you get that yield stress number? Is it from that graph with Aluminum and the plastics?
Because that graph shows modulus(stiffness), not strength. Which is pretty hinky when placed next to the statements about strength. Looks to be a bit misleading.
On top of that "stronger" has no meaning when it comes to composites. There are about 9 different failure modes, and you can't just choose one and report the "strength". I'm guessing they're talking about axial tension numbers (because I'm cynical) where their material would perform the best. Even then, their yield strength is about 200MPa (based on the "5x stronger than ABS" statement) whereas aluminum's yield strength tops out around 600MPa.
I'm skeptical too, but don't count it out because the fibers aren't woven together. The weaving is not where the strength comes from, and it's usually not omnidirectional. If this machine can set the wrap angle, and build out several layers, the fiber layout might be comparable to a filament winding process. http://en.wikipedia.org/wiki/Filament_winding
But the thing that I'm concerned about is the curing of the stuff. A composite gets its strength by having a low resin content, and I'm not sure this machine/process can pull it off. Based on the few available details, it's right to be skeptical.
Does this even work? It's just an IR detector, and you can be pretty sure that the sun is a bigger emitter than security spotlights...
I did my undergrad with Jon Clark! Great guy. It's cool to see a few bio-robotics people on HN
Great! How far does the book go in terms of advanced approaches? Up to the current state of research?
What about convolutional neural nets? They weren't mentioned, but that's really what most of the deep learning approaches use...
This is interesting because they could potentially avoid the thermal expansion problem that would otherwise sink the hyperloop. Carbon fiber has a negative coefficient of thermal expansion in its axial direction, and you can pair it with a metal to get a net zero expansion structure.
This is the second worst application of the "animal-inspired" moniker that I have seen.
Nike calling their sneakers "bio-inspired" because they had leopard spots on them was the worst.
Shall we not forget the MIT cheetah: http://biomimetics.mit.edu:8100/wordpress/
I like the design much better than BD's weighty quadrupeds.
I agree that it's core research, but it isn't published anywhere. Their control algorithms are not public; even plenary talks from Mark Raibert don't go into the details.
(wheels are fine)
I disagree with this. Try constraining yourself to a wheelchair for a week or so, and then tell me nothing can be gained from a good set of legs.