While doing my background research, I found a forum containing a discussion on the use of UAVs for falconry (http://www.falconryforum.co.uk/showthread.php?s=8dfe66d7d5c5df7d09fdef9d68642c31&t=133013). The initial poster, TomOlivia, describes the very positive experience he has had with it. He points out that there are limiting issues in the range of the UAV, but that these do not cause any issues in practice, as the UAV can fly much higher than a bird of prey would for lure training. He hightlights the numerous advantages of using UAVs for falconry, including the ease of use, the ability to collect heights and timing metrics, and the speed of getting everything in to position. He comments that he has had some issues with the release mechanism. I am assuming that he has used a similar release mechanism to that seen at the Falconry School. This issue should be overcome by the custom release method that I will be creating in my project.
An important issue highlighted is that of the use of a parachute during the bird’s descent. When using a kite/balloon, the kite/balloon is tethered. The lure can be looped around this tether, preventing the bird from flying away with the lure. This cannot be done, and so the parachute is attached to the lure to prevent the bird from being able to fly away with the lure. Birds of prey do not feel safe eating on the floor generally, and so for larger meals (such as a lure), they will try and fly somewhere safe to eat. Once they have eaten they will then lose interest in eating further (and so flying to the falconer to be rewarded with food), making it difficult for a falconer to recover their bird. One potential issue that I observed during a test at the Falconry School was that the bird hangs upside down during descent (as seen in the below picture):
Birds of prey do not have diaphragms as we humans do. Because of this, they cannot breathe upside down. This means that if a bird is upside down for an extended period of time, it can potentially suffocate. I will need to discuss this issue with staff at the Falconry School, but it seems that the short duration of the fall may prevent this from being a real threat to the bird. The UAV will certainly not be dropping the lure from excessive heights. Calculations could perhaps be done on the time taken for a bird+lure+parachute to reach the floor, to assess the maximum safe height. There does not appear to be an immediately obvious alternative to using a parachute on the lure, but it is possible that an alternative will be created during this project. The parachute will also likely have issues if the weather is too windy, as the bird+lure+parachute could be blown away from the intended landing site becoming, for example, tangled in a tree. On a windy day however, the kite based lure system could instead be used. Birds additionally tend to dislike flying on excessively windy days, and so lure training may simply not be done on windy days.
TomOlivia mentions in the thread the benefits of waypoints being set in the UAV. He states that it is very easy to retrieve the bird, as you don’t need to focus on controlling the UAV. He even recommends switching off the controller to force the UAV to return by itself to the launch site.
Another user, Florrain, mentions that in Doha (capital city of Qatar), the use of balloons for falconry training has been banned. This user states that falconers there have begun experimenting with the use of small quadcopters for falconry as an alternative. This user also highlights the need for a method of releasing the lure from the UAV, to reduce the risk to both the UAV and the bird. They then go on to describe methods of modifying off the shelf UAVs to release lures.
The user lorrain additionally references a modified safety UAV, as can be seen here: http://www.borisboege.com/drone-safety/. The website mentions a company that has already gone out of business, Safeflight Copters LLC. They highlight issues in their article that I have previously mentioned within my project, such as the hampering of aerodynamics caused by putting protection over the blades. They also highlight the issue of increase cost in manufacturing. This is not an issue in this project however. In the pictured UAV in this article, the quadcopter appears to have wire mesh over the rotors. This could potentially also be investigated as a method of bird-proofing the blades within my project. A user, Old Wolfie Gosser, provides an interesting anecdote. He states that one of his birds tends to fly above the kite when doing kite-lure training, and stoops on the lure. He would not like to use a UAV with this bird, due to the increased risk of impact. If a bird exhibited this behaviour at the Falconry School, this could be problematic. However, as all birds that are flown to lures at the centre have already been flown to kites/balloons already, the Falconry School staff should be able to predict any birds which may behave in this manner, and avoid using them for UAV-lure training.
There seems to be general discussion that flying a UAV in GPS mode is much easier than flying in manual mode. At this point in time I am unsure as to what exactly GPS mode flying is (I’m assuming it isn’t entirely waypoint based), and so this could be a useful area for me to investigate to feed in to a more useable control system.
The idea of using a line on the UAV which the lure tethered to it to “guide” the bird to the ground is discussed. This would remove the need for a parachute, and would reduce the risk of the bird drifting in to undesired areas while coming down. However, it is pointed out that you would need a quite powerful UAV to be able to withstand the weight of a bird of prey hanging off of its line.