Lessons that DroneJournalism.org developers learned from designing, building, testing, and re-designing JD-01:
- Hobby-Class is good, but not quite good enough. What we're trying to accomplish at DroneJournalism.org is a happy medium between the $50,000 commercial drone with all the bells and whistles, and the $500 toy. We don't need it to have all the capabilities and sensors of a warzone-ready drone. At the same time, it has to be able to take some abuse. We still believe it's a reasonable goal to start with hobby-class equipment, such transmitters, metal-gear servos, batteries, autopilots, and in some cases airframes. But what we're finding is that the standard foam and/or balsa craft simply can't cope with our demands. Yes, some of that can be blamed on inexperienced pilots. But journos we introduce to this equipment also will be inexperienced, and they need something that can take some punches. And we don't entirely chalk it up to pilot error, either -- our experience shows that some equipment and hardware can be damaged by anything other than light handling. We can start with a hobby-class airframe, but we need to at least strengthen it out of the box. This means reinforcing an existing hobby airframe with lightweight and high-strength material, or it could mean fabricating a simple airframe to spec, or preferably both.
- Drones must be serviceable in the field. Every time we had a setback, even a minor one, we were required to return to the bunker, and the drone would be down for days. This was not only due to the frailty of the small parts involved, but also because many parts could not be swapped out in the field. This means moving away from micro-sized components (such as tiny servos, small bolts and screws), and moving towards the kind of standard hardware you could find in a hardware store (bolts that are at least 3/16 in, hefty mounting plates). That also means substituting wood for metal, and plastic for carbon-fiber reinforced polymers. This obviously would increase the weight and size of these aircraft, but we feel that the enhanced up-time in the field would be well worth it.
- Landing gear, launching platforms are not an option. The mode of launch is a decision that balances the ease and reliability of the launch with the price, transportability, and complexity of the system. Small UAS for military applications and many entry-level hobby gliders are launched by hand, but this can be difficult to accomplish if you also are piloting the drone. A launcher might be a good investment as it reduces the chance for human error, but it introduces additional complications and comes with a severe penalty to transportability. Meanwhile, ground takeoff may be the easiest for beginning pilots, but the landing gear always comes with a weight and performance penalty, not to mention the problem of finding a suitable runway. Hand launch seems to be the best option, and that is the method we will pursue in the future.