Group items tagged

More information on the future of UAVs and the space industry.

Not explicit but smells like RTGs

"The effort contrasted and compared eight heat sources technologies, three power conversion, two dual cycle propulsion system configurations, and a single electrical power generation scheme." Yes, it's about RTGs, at least partially.

The quote "it was disappointing to all that the political realities would not allow use of the results" may refer to RTGs too :-)

If you've ever flown an R/C plane, you know how nerve wracking it can be. Navigating in three dimensions opens up the possibility to run into so many things, and a single crash could be your last. Now, a team from EPFL's Laboratory of Intelligent Systems has been working on a UAV called the AirBurr.

An approach to robot design that makes sense. I can see applications to planetary explorers here.

Demand for drones is exploding! Dupin wants to aggregate aerial imagery from around the globle at Dronestagr.am. In the near future we could experience something close to google maps, made with aerial pictures.

Remember that drones have top-view camera and front-view camera which gives more possibilities in terms of what you can do with collected data. With such a huge database and a little bit of 3D geometry we could get e.g. a 3D map of the world... I guess google can derive something like that already from their streetview images however obviously street view covers some relatively small part of the globe and also can not access places that UAV can.

Here there's an example of a 3D picture of Tokyo (in high resolution). It is not made by drones, but I think it is something we can use them for. http://360gigapixels.com/tokyo-tower-panorama-photo/

StarTiger: similar, yet different from the way the ACT does things. Seems like a very interesting programme.

Nice initiative and also a good approach, problem-oriented within a fixed time frame. Could definitely be a highly motivating approach, similar to GTOC... I think the ACT should do this more often, targeted at future technologies and/or missions. The team could be structured around 'problems' instead of 'research areas', this will promote multidisciplinary work as well, plus it will also focus activities more. The problems, or more broadly concepts, are identified by the team and a few get chosen as main activities. Subsequent RF and YGT hiring is then done to strenghten the research team. These projects have a maximum lifetime maybe of 1 year? Thoughts?

I'm impressed already by what an innovative group of experts and researchers was able to achieve when resourcefully targeted at coming up with the project's name...

"A new lightweight, energy-efficient tool for analyzing a material's chemical makeup could improve the detection abilities of various technologies, ranging from bomb-detecting drones to space rovers searching for signs of life". Raman Spectroscopy is about measuring vibrational modes in molecules. This vibrational modes are in the meV typically, turning Raman Spectroscopy into a high precision technique. This impressive work shows a new technique based on the use of optical fibers coupled to photomultipliers allowing its use, author's word, in extreme conditions such as unmanned aircraft vehicles (UAVs) and Mars/Moon rovers.

OpenKinect project http://openkinect.org/wiki/Main_Page

An interesting research centre in California! Focus areas: Business Services Electronic Materials, Devices, & Systems Information & Communication Technologies Biomedical Systems Cleantech

XEROX PARC was definitely heavily involved in computer development: eg. mouse, GUI, ethernet, OO programming, all came out of PARC, and all that without focusing on computers but printers...

aaah its the XEROX centre, didn't know. Yep they made the mouse and then handed it over nicely to Apple after IBM thought it was useless

Prediction-based method for encouraging reinforcement learning agents to explore their environments through curiosity (reward for unfamiliar states). Learns some games without any extrinsic reward!

Fun failure case: agent gets stuck in front of TV.

Not read this article but on a related note: Curiosity and various metrics for it have been explored for some time in robotics (outside of RL) as a framework for exploring (partially) unfamiliar environments. I came across some papers on this topic applied to UAVs when prep'ing for a PhD app. This one (http://www.cim.mcgill.ca/~yogesh/publications/crv2014.pdf) comes to mind - which used a topic modelling approach.