There are lots of good ways to do things and my advice is only out of my own tiny slice of experience.
My one caution is that if you pick a plane that needs a catapult or bungee to launch, that is not insignificant. That is extra stuff to load up in the morning, extra stuff to transport, extra stuff to setup before your flights, extra stuff to take down, pack up, and finally unload back at home or the office. Also, every plane I've personally flown that needed launch assist had pretty high wing loading and thus was at least somewhat sketchy. (There are some great commercial/military launch and recovery systems if you don't mind taking an extra humvee and trailer and six extra guys with you every where you go.) If the launch is even a little bit sketchy, that will translate into launch fails once in a while. Failed launches result in broken airplanes, field repairs, and often downtime. The thing that sucks the worst is going through the whole extended process of prepping, traveling, setting up ... and then you prang your airplane on launch and you are done for the day. It would be easy to think you'll just bring an backup airplane and spare parts ... and that may be great. But in my experience I'm often down to one flyable plane, or one airplane that is currently setup for the thing I need to do. Sometimes field repairs don't work out if everyone is on a tight schedule. It's hard to make quality decisions and do quality work under pressure, sometimes it is just easier to call a 3rd strike on yourself, go home, and regroup.
Flying wings typically do not have friendly stall characteristics. The one 7.6' flying wing I helped design would go around twice after a snap/stall before you could recover it.
Pusher props can't blow air over your own control surfaces. Your control surfaces are only as effective as your actual airspeed. This means that it can be very difficult to power out of a slow/sketchy situation if your launch system (and wind conditions) aren't perfect every time.
High wing loading often happens when you increase battery for longer endurance, and get the complete payload package installed.
It would be easy to find yourself in a situation where your power system worked great for the first empty test flight, but might suddenly be fairly under-powered when you finally get your airplane fully outfitted at it's mission weight. Our flying wing on a high density altitude day could do about 100-200 fpm climb which was pretty white knuckle if you hit a couple up/down drafts and start wondering if it's actually able to climb or not. (We had nice endurance, cruise speed, and payload specs though ...)
So mix all that together ... flying wing, pusher prop, high wing loading, high stall speed, possibly under powered == sketchy launches.
It all depends on the actual details of course, but for any uav work I do now and in the future, I'd never pick a flying wing unless the situation had no alternative. Personally I have a couple fun flying (pure RC) flying wings that are easy to launch and fly great and are super fun (the sonic 64 is great) ... but those are very lightly loaded, way over powered, and much smaller than your typical uav would be.
My 2 cents ...
