What Problems for 50 year space probe to Alpha Centauri? I'm thinking speculatively.  NASA is doing experiment with ion thrust tech.   Assuming, that inside the next 5 years, we could create a probe which would take on the order of 30 to 50 years to reach Alpha Centauri:  What problems would it have?
ie:  Is the biggest challenge going to be communications back to Earth?  Maybe a small probe can't transmit a strong enough signal (especially since we'd be essentially looking straight at the star, from our point of view) ?
Or Maybe the challenge in transmitting is really the targetting, to keep the Antenna pinpointed on Earth?
Or would there be problems with powering the probe over a 50 year journey (at which point we'd really want it to have several years left of usable lifetime) ?
Or is there problems with space probes, outgassing or other gradual losses which would cause the device to decay after 50 years of vaccum?
Do the long-lasting Voyager probes help answer any of these issues?   I know they are MUCH closer to earth than what I'm talking about, but at least they are still alive and transmitting...
This is a speculative question, so I hope it is allowed by the forum rules.   I'd like to think about how such problems can be overcome -- but first I really need to know what the problems might be.  So I'm looking for thoughts or info...
EDIT:  perhaps I was optimistic.  if a 50 yrs travel time is not practical, even with a long duration thrust like ion drive, then feel free to posit a longer flight time.  
 A: Probably the single biggest obstacle is detection and avoidance of objects in space.  To get to Alpha Centauri in 40-50 yrs the probe would be traveling at relativistic speeds (~1/10 the speed of light).  All collision detection and avoidance would need to be handled by onboard systems since the ground controllers would have no direct control.  You have to realize that at the speeds involved, the probe would cover the distance between the Earth and Moon in about 10 secs or so.  The volume sweep of the probe in that time is enormous, and because of the speeds involved, and the limit of the speed of light, the device will cover at least a tenth to a fifth of the distance to an obstacle after detection in an active system.  Passive systems would be good early warning, but regardless the system will need massive computational power to calculate projected courses of obstacles for avoidance.  
For these reasons, there might need to be a series of probes, a few probes that would be "disposable" that act as pathfinders, followed by more critical probes that contain the more important science packages.  
A: to reach Alpha Centauri in 50 years you need at least nuclear fission power. With nuclear fusion it could reach there in 30 years as proposed by the Daedalus and Icarus projects, but consider that those missions are proposed as flybys: they don't pack enough fuel to brake again at arrival
The biggest challenges are multiple, specially if the mission is flyby, this severely constrains the amount and quality of the data that might be retrieved by the probe. Definitely communications is a problem but it is believed to be solvable. Of course, the probe will not be able to rely on earth communications to take any decision, since the communications latency will be of 8 and a half years at destination.
The Voyager probes are right now entering interstellar space, but they won't get anywhere near other stars for a couple hundred thousand years. Their radioactive thermoelectric generators will be entirely exhausted in a couple decades from now. So in order for a interstellar probe to reach other stars in less than, say, a hundred years, there is a limited list of what power sources can be used:


*

*fission nuclear power (Orion spaceship)

*fusion nuclear power (Icarus and Daedalus)

*antimatter-matter power (we still have to figure out how to store and produce large quantities of antimatter but without adding too much dead weight)


The Icarus project is working on refinements of the original Daedalus project, definitely check that for more details
A: Targeting the antennea on Earth isn't a problem - we can point things at stars rather accurately. And any practical size radio antennea would spread out a signal to such a wide beam that pointing at Earth would be the least of the problems. A moderate size telescope and a laser would probably be better - just pick a wavelength that isn't strong in the star's spectrum.
Powering it with radioisotopes for 50years again isn't much of an issue. The Plutonium RTG on the new Mars rover is good for 100-200W for 15years - so enough to run any on-board electronics.  The power demands of the thruster are bit more of a problem!
Outgassing we can handle, once the probe is in space it reaches equilibrium pretty quickly. In interstellar space the main issue would be background radiation, mostly rare but high-energy cosmic rays. You need to have multiple redundant systems to recover from temporary or permanent damage to electronics.
edit: oops space is surprisingly big. A 5MW ion thruster would take something like 2500 years to reach A Cen.
A: The rocket equation is a cruel tyrant. By far the hardest challenge is just finding a propulsion system that can achieve the necessary velocities. Project Daedalus proposed to use an inertial confinement nuclear fusion system, but we really don't know if that can be made to work. The best existing inertial confinement technology is the National Ignition Facility, which uses lasers and barely achieves breakeven, with a huge building's worth of heavy equipment to make that happen. There are no fundamental principles that must be broken to make it work, but there is also no guarantee it can be engineered.
Compared to that challenge, simply continuing to operate for 50 years is easy. 
Avoiding collisions should not be a problem. No space probe used to date has had any provision for collision avoidance, and we have lost none to accidental collisions (unless you count crashing on a planet). 50 years is a much longer time, but interstellar space is much emptier. 
Fundamentally, the problem is just that we don't know how to make a rocket engine that achieves the necessary exhaust velocities. 
