Pushing The Envelope

FWMiller sent us this gem: "Was cruising around NASA's pages and found this list of studies that have been funded by the NASA Institute for Advanced Concepts (NIAC). Interested in what might be coming down the pipe in the way of new propulsion and vehicle concepts? This page lists links to a few of 'em." I thought the one titled 'A Realistic Interstellar Explorer' was cool.

Comments

More than meets the eye

[DavidBrown]

Here's one proposal:

"Self-Transforming Robotic Planetary Explorers"

Autobots or Decepticons?

The problem is different

[jeroenb]

I always thought the problems with figuring out how to do interstellar travel were mostly in theory - like how do we reach high velocities without using up impossible amounts of fuel or how do we steer these ships, but in fact, most of these papers have all that figured out but they're just waiting until we have the technology to manufacture stuff like the Solar Sail. That's actually pretty cool, we're closer than I thought :)

Wonderfull! But what if...

[Tin Weasil]

Microsoft software is used to run these "future" spacecraft? I can see it now:

1. 
   2001: A Space OS
   
   "Houston... we have a problem..."
   
   "UNSS Gagarin, this is Houston. What is the nature of the problem?"
   
   "Well... I know we are supposed to be on a mission to Proxima Centauri, but..."
   
   "But what?"
   
   "Well, our HalNT 9000 computer bluescreened just between the orbits of Neptune and Pluto."
   
   "So... just reboot the damn thing and get back on the mission!"
   
   "It's not that simple Houston. Our friend Astronaut Sam decided to fix it by loading Linux on the computer."
   
   "WHAT?"
   
   "Yeah. Everything is working fine now, but we're going to need someone down their to get us a device driver for the ship's dueterium fuel tanks... and the drive motors... and if you could get us drivers for the life support systems we would really appreciate it. The sooner the better on those life support systems."
   
   "Oh... My... God..."
   
   "Exactly. Dave forgot to check the Hardware Compatibility List before loading it up... I guess the ship's hardware is all proprietary. Needless to say, we have trapped Dave in the airlock. Want us to let him breathe vacuum?"
   
   "Only if someone else has the root password."
   
   "Oooo! Good point. Ummm, by the way, is the new 2.4 kernel out? We could sure use some help with these USB peripherals also."
   
   "Okay Major Tom... you and your boys sit tight. We'll get our team down here working on a solution..."
   
   

Re:well...

[KevinMS]

reminds me of that sci-fi story in which an expedition of cryo-suspended people wake up at their destination and find people already there because along the way they were passed by faster, more advanced ships.

Here's to the dreamers!

[Alien54]

Well, at least they have something in case they ever get a real budget. There are so many toys here.

Topics include

Advanced Power - Advanced Propulsion - Aeronautics Platforms - Asteroid Detection - Astronomy - Biology - Communication - Earth Science Platforms - In-Situ Utilization - Planetary Colonization - Robotics - Satellite Teams - Software

I like this specific abstract in the software section:

"Architectures and Algorithms for Self-Healing Autonomous Spacecraft"

Imagine spacecraft whose missions last three times the human lifespan. Imagine spacecraft with the ability to decide where to explore, how to plan a trajectory, and which data to record.

These autonomous spacecraft will require computational systems whose fault tolerance and performance are orders of magnitude better than presently possible.

This challenge has been recognized by NASA's Dan Goldin as calling for revolutionary computational systems that depart radically from contemporary designs.

We propose to develop a family of such systems, with emphasis on algorithms whereby the architecture heals itself. Highly autonomous spacecraft will require computational systems that tolerate a number of faults in proportion to the total number of components, hardware and software.

This is orders of magnitude better than presently possible. To enable the combination of fault tolerance and performance, we envision a self-healing architecture. Self-healing architectures would naturally support fault tolerance, and are therefore amenable to scalable constructions. Self-healing architectures could be realized using a variety of technologies.

The benefits of self-healing architectures extend to military and commercial applications. However, a self-healing architecture such as we propose has never been built, nor is it on the evolutionary horizon of the immediate decade. Our effort will identify properties for self-healing architectures that deliver at least 1015 operations per second per kilogram, and that tolerate a number of faults proportional to the number of components.

For Phase I we propose to deliver a graphical, executable model of a highly fault tolerant, self-healing architecture prototype. The fidelity of this Phase I model will be sufficiently rich to demonstrate tolerance to a number of faults in proportion to the number of components. For Phase I we foresee two major technical challenges:

i) generalization and merging of results from configuration for performance with results for configuration for fault tolerance, and
ii) specification of a baseline programming model for a self-healing architecture.

The next step would be to design and construct a self-healing architecture and attendant software. Looking to Phase II and beyond, realizing such an architecture will of necessity be multi-disciplinary, and will draw on the expertise of specialists in algorithms, testing, software engineering, circuits, power, packaging, radiation hardening, thermal and mechanical design, control, sensors, and mission planning.

Wow, what a project!