The latest innovation we are testing is a camera that can rotate around, up and down to follow predators. There are two main benefits of using this. First, we can see much more than with a static camera. The second, main long-term reason is that this will allow us to implement a kill mechanism that will not require the predator to enter a trap.
Video 1: First prototype of tracking camera
Seeing 4-8 times as much with the same camera
We estimate that the tracking camera will be able to see up to 8 times as much as a static camera. Given the number of times we see predators walking straight past traps completely ignoring them, it seems safe to assume many could be walking behind and to the side of the camera without coming into the camera's view. We also see plenty of shots where the animal appears just in one corner of the view and there would not be enough time to identify or kill the predator. Being able to track the predator will give more time to be able to identify accurately what type of predator it is.
Tracking to kill
One of the main reasons we did this part of the project sooner rather than later was to show that the idea of a device that can track a predator in real time is not actually that difficult. Obviously there is a lot of testing and levels of redundancy required before enabling this. The first part would be a 100% animal identification system for the predators and simultaneously a 100% confirmation it is not a human. All current traps have some possibilities for trapping non-target species and even to allow damage from human mental lapses. We would also need a very highly accurate kill method that has low negative impacts if there was failure.
There are a number of possible kill mechanisms that this could be linked to, but let’s take as an example the idea of a paint ball with poison that would splatter directly onto the predator. They are all groomers, so the theory is that they would then go off and lick the poison from their fur. This means they are likely to ingest poison even if they are not hungry. If we compare this to commonly accepted forms of poisoning such as bait stations or tipping from a helicopter, then it seems vastly more targeted and safer. It is only applied to a confirmed predator and never a pet or person.
Issues and opportunities
The obvious first issue with a tracking camera is the increased number of ways it can break down. However, this is all solvable with fairly standard engineering. If the camera can see and track even 4 times as much then the additional cost will pay for itself.
Noise of the motors could be an issue that scares off some predators, but we will be able to test that and change the design accordingly.
In the short-term it may be that the camera does some regular scanning to pick up predators but in the long-term it could “listen” for predator noises and turn in the appropriate direction to see them coming.
The biggest opportunity long-term is the ability to target and kill any type of predator. The larger the area for the predator kill, the less targeted the lure needs to be. If we needed to get predators to stand in a very small space (eg one meter by one meter) this could limit how effective sound lures could be.
In the short-term we are likely to keep using static cameras mostly to get the data for artificial intelligence identification, but hopefully this little demonstration shows the ultimate device for predator removal is eminently possible.