A view to a kill: Making NZ predator free with an autonomous kill-bot
Making New Zealand predator free by 2050 sounds a crazy ambition. But maybe some crazy technology could achieve that. JOHN McCRONE reports on the Cacophony Project.
What could possibly go wrong? Achieve New Zealand's Predator Free 2050 goal in double quick time by building an artificial intelligence-controlled killing machine.
A robot system which sits in the bush with a night vision camera, animal recognition software, and an adapted paintball gun that can spit 1080 pellets.
I am talking to Menno Finlay-Smits of the Christchurch-based Cacophony Project – a Millennial-styled "open-source" technology collective.
Finlay-Smits says the target recognition part of the job is already half-done, cobbled together in the last few months. The Department of Conservation (DoC) could have working devices in the field by the end of the year.
* Recording Christchurch's birdsong cacophony
* The science behind a predator-free NZ
* Predator-free New Zealand: No fences allowed
* 'Impossible to say' if predator-free 2050 is realistic, experts say
The next step would be to strap on a turret with a 360 degree range of motion, laser sights, a firing distance of 10 metres.
You would have a machine that thinks for itself as long as its battery lasted, blatting every passing possum, rat or stoat with a sticky micro-dose of poison.
"It would break on their fur. That wouldn't kill them, but all these animals groom," he says. Lick the mess and they die. Gone. Finito. Problem solved.
It sounds like a tech fantasy, so I joke: why not mount them on drones? Surely every decent technology story has to feature drones these days?
Funny you should say that, Finlay-Smits replies. That is the advantage of Cacophony's open source development philosophy.
"The drone thing has already been floated. We've been talking to some people who are quite active in drone research."
So yes, why not. Add that too to the plan. What could possibly go wrong?
Everyone remembers what a wild promise it seemed when the Predator Free 2050 goal was announced by former Conservation Minister Maggie Barry in 2016.
Part of the incredulity was simply that it was a National Party government proposing it as New Zealand's "technological Moon shot". People were immediately looking for the catch.
But also, wiping out every introduced pest animal, from the deepest bush to the suburban backyard, just looked a hopelessly difficult task.
Barry tried to sound convincing. She said the government was setting up a new business entity, Predator Free New Zealand Ltd, to develop the means. This would "work alongside the private sector" to perfect new pest control techniques.
There would be a Biological Heritage National Science Challenge to fund basic research. And DoC was supporting a new agency, Zero Invasive Predators (Zip). So the longterm infrastructure was being created, she said.
The programme also had some kind of roadmap – four interim targets to hit by 2025.
Barry listed them. One million hectares with pests removed. At least one science breakthrough to completely eliminate one mammalian predator. Proof that 20,000ha could be kept predator free without needing predator fencing. And all our off-shore island reserves fully protected.
However not a lot of new money appeared involved. The initial budget was $7m a year when conservationists were saying $9 billion was the realistic cost.
Then someone mentioned genetic modification as one of the possible coming science solutions. Someone else said feral cats and garden hedgehogs would of course have to be on the extermination list.
Hackles rose. Barry's Predator Free ambition started to look like an ill-considered publicity stunt soon to be forgotten.
However in fact the search for new cost-effective pest control techniques did get going. And some are feeling the progress has been surprisingly rapid.
It has helped that there has been some supercharged philanthropy getting behind it.
In particular, the Next Foundation – set up by Kerikeri businessman Neal Plowman to give away $100m of the fortune he made from the family industrial cleaning firm – has emerged as a key player.
It now helps fund both Zip and the Cacophony Project. So two approaches are being supported.
One is incremental. Zip is focused on research to get maximum results out of old school technology – 1080 poison drops and trapping. Then Cacophony is the "let's rip up the script/use Kiwi No 8 fencing wire ingenuity to solve a wicked problem" approach.
And obviously, the way-out story is the one to talk about first.
Cacophony is the brainchild of Christchurch inventor Grant Ryan, a serial entrepreneur who rode the DotCom boom of the late 1990s and has continued with ventures like the YikeBike.
His project began with more modest goals. Ryan had moved his family to Akaroa after the earthquakes and decided to rid his new bush plot of pests. After laying traps, he noticed how the volume of bird song increased.
It seemed a neat idea to be able to measure this effect on a national scale. It would give an accurate picture of how well our predator control efforts were actually doing.
Working with a team of techie volunteers including his engineer son, Cameron Ryan-Pears, Ryan lashed together a device – the Cacophonometer – that was basically a cheap cellphone with an app, housed in a waterproof casing with a small solar panel on top to keep it charged.
The plan was to be crowd-sourced. People all over the country would stick one of these in their gardens to start building a national database.
A next step would be to add sound recognition software to help count the different species present. But then came Predator Free 2050 and the project turned into something much larger.
Ryan had set up thermal imaging cameras to monitor his traps and seen that possums had often wandered right past the bait as if uninterested.
So he began playing around with the possibility of using sound lures. A speaker was added to the Cacophonometer to see if possums or rats could be attracted by recordings of mating calls or other "social" noises.
From there came the thought of a completely autonomous killing machine. If AI can drive cars, then bowling weasels or rats in the dark using a heat-sensitive camera and paint-ball gun should be a cinch.
Ryan did a back of an envelope calculation to show the device would be 80,000 times more price-efficient than regular trapping. Serial entrepreneurs certainly understand how to make their elevator pitches.
It quickly got serious with support from Next and Zip arriving mid last year.
Finlay-Smits – a software expert who cut his teeth on self-steering tractors in Queensland – was recruited as project leader. Prototyping began in earnest this January.
Now Finlay-Smits says it seems a reasonable expectation that a first AI animal recognition system will reach the field this year, and a $1000 kill-bot within five. Well, why not?
Getting down to the tech details, the Cacophony Project does seem a distinctively Kiwi lo-fi approach.
Finlay-Smits and the handful of other funded staff operate out of their own homes. No corporate offices or government research institutes. All garage and backyard development.
Finlay-Smits explains open source as an ethos. He says it is now standard in software – the world of apps. People form loose collaborations based on their skills and ideas they can contribute.
And the driver more than money is working on something cool, something of general social impact.
Cacophony is an example of the same philosophy now being applied to hardware development. Because of Moore's Law – the exponentially falling cost of any computer-based technology – the world suddenly finds itself flooded with cheap sensors, cheap processing boards, cheap communications.
There is an abundance of off-the-shelf tech waiting to be taken advantage of. You don't need a corporate R&D department building entire systems from scratch, just a team with the smarts to glue sophisticated components together, he says.
"It's the start-up mentality. Agile, or 'fail fast'. Try it and see if it works. Then move on to the next idea if it doesn't."
Finlay-Smits says Cacophony breaks down into three sub-projects, each of which will be a useful predator control achievement in itself.
The first is the AI recognition system. This has been knocked up quickly using available hardware.
The thermal camera is a US-made $230 Flir Lepton 3 – a single chip device smaller than your finger-nail. The computer is a $50 Raspberry Pi board with built-in wireless. The software is an open source version of Google's TensorFlow neural network programme.
Finlay-Smits says it is just a question of integration. "Really, 80 per cent of the work is getting the data into a format the system can use for its machine learning."
To train the Cacophonator, as they are now calling it, the team have installed units in a number of Banks Peninsula locations to film animals rummaging about at night.
Finlay-Smits says the system can be very dumb. At the beginning, it started to label every tree as a possum.
"Early on, the model realised any time there was a tree visible in the frame, there was probably going to be a possum. It's like a child almost. It will find whatever it is that there is in common with all the videos you've tagged for it to look at."
But quickly, as the parameters were adjusted so it only tracked small, moving, hot objects, it has learnt to distinguish possums by their signature shape and movement. Likewise stoats, weasels and rats.
Finlay-Smits says the system is also getting reliable on cats, hedgehogs and birds. A camera was set up at the Willowbank Wildlife Reserve to train it on kiwis in the nocturnal enclosure.
The Cacophonator should be completely trained by Christmas. "It's come along much faster than we expected."
And DoC is keen to use it just to help with the existing problem of how to accurately count the presence of predators in an area of land it wants to control.
Currently rangers rely on chew cards – bits of hollow board baited with peanut butter. Or for mustelids like stoats, tracking tunnels where the animals crawl across an ink pad and leave paw prints on paper as they exit.
The Cacophonator would also help with live trapping. An issue with traps is they mostly rely on a pest species sticking their heads inside something small and confining. Often animals are just too wary.
So a large open capture trap is more effective. However then weka or pukeko can blunder into them as well.
Radio transmitter versions can tell DoC staff that a trap has been triggered. But they still need to rush in and check what they have caught.
A Cacophonator could realise it was a native bird nosing about, says Finlay-Smits. "If it sees a weka, it says don't activate. If it sees a possum, then do."
So a cheap AI recognition device would already have many uses as part of the Predator Free 2050 effort.
The project's second major goal is to develop the technology of sound lures. Traditional control measures, whether a poison pellet or trap, rely on food odours to attract.
But Finlay-Smits says social sounds, like mating and distress calls, or food-like noises, such as the rustle of a nest of chicks or the scratching of wetas, could be used to draw animals towards a device. Even the curiosity value of a faint electronic beep, might work.
And immediately, that increases a device's range. "The maximum you'd get out of a food lure is 3 to 5m, whereas it'd be tens of metres or more out of sound," he says.
So the plan is to equip the Cacophonator with the capability to play sounds from a library of recordings. Finlay-Smits says Zip and others will be running field trials around New Zealand to discover what might work in the wild.
Finally, the killing bit. The ultimate goal. Finlay-Smits grins.
As project leader, he says his job is to ensure Cacophony moves ahead in sensible stages. And right now, less sexy stuff, like finding a reliable battery, is a more pressing matter for him.
But yes. They have done some Heath Robinson trials with a laser sighting system.
A heat pack – the kind you microwave for a sore back – strapped to the top of a toy remote-control car, makes for a good surrogate stoat. "We got the prototype to happily follow that around."
Fun is clearly being had in someone's backyard.
Then he says in about six months, the plan is to start looking at adapting a paint-ball gun. Technically, nothing seems too difficult as it is all existing hardware. Someone just has to put it together as a practical system.
Of course, he realises that the creation of an "autonomous kill-bot" must instantly raise health and safety fears. Even project members make the joke – what could possibly go wrong? The Cacophony Project will eventually face regulatory hurdles.
But think about it realistically. The heat signature of a human or other large animal is going to be easy to pick out. The devices will be set 1m high, pointed down towards the ground, and only operating after dark.
"Most of the risks can be mitigated. At worst, you're only going to hit people below the knees. It might break the skin, but as long as people don't lick themselves, they should be OK," Finlay-Smits answers with a laugh.
There will be also the chance of error in hitting a native bird by mistake. Or someone's beloved cat.
But he says compare that risk to what already happens with 1080 being rained from the sky by the truckload. The wrong things often get killed as it is.
Then there is the bang for buck factor – Ryan's elevator pitch that these devices could be 80,000 times more cost effective.
The argument is sound lures should represent a 100-fold improvement over regular traps in terms of labour hours at least. No need to ever reset baits, or clear dead bodies from traps.
And only a hundredth of the actual number of traps would be needed if sound lures can attract animals into killing range over a distance ten times greater.
The fact the system is universal – taking out possums, rats, mustelids and feral cats all at the same time – spells another factor of four cost reduction. Then the kill rate should be nearer 100 per cent rather than the 50 per cent of conventional methods.
Tot it up and you get Ryan's eye-catching figure.
And drones? Finlay-Smits says another cost saving is the Cacophonators would be deployed as a line marching across the landscape, clearing the bush as they go.
This would be cheaper than having to sprinkle them about. They would just sit in one spot for a week or so, then be moved along.
And a drone version would save on paying humans to do the moving. Finlay-Smits says you could have a system that even returns to base for recharging and reloading before flying off to its next location.
It sounds crazy. Even wilder. But he says the advantage of the open source approach is that if there are drone enthusiasts out there who think they could make this work, they are free to give it a go, add this tweak to the design.
A conventional approach to system development is always going to be conservative. Yet modern tech is evolving so fast that you can really try just about anything you can imagine.
"Because the project is of such national interest, and there's a lot of smart technologists in New Zealand, once you start showing some success, people are going to be saying, hey, I can help you with that."
By its nature, the Cacophony Project could still go off in surprising directions. It is not following some exact blueprint but making it up as it goes along. The final outcome may be different from what is currently envisaged.
But Devon McLean, the ex-forest industry boss who now chairs Zip and is a director of Predator Free NZ Ltd, says Cacophony's contribution should be significant.
"It could be the game-changer for us in the end. Especially if it allows us to chase down the last few percentage of the pests."
Yet also, says McLean, the incremental approach is looking like it can extract far more out of existing control technology than expected.
Working with DoC, Zip has been conducting landscape-scale trials on the more effective use of 1080 and trapping.
McLean says it might be thought New Zealand already must know how to do the old school stuff. However eradication programmes have often had one hand tied behind their backs by public opinion and uncertain government funding.
Knocking back the numbers was often enough. Only now is the research going into finding what kind of control regimes could lead to complete elimination of pests in an area.
Zip has been experimenting with more intense 1080 campaigns – doubling the coverage – while also taking advantage of "defensible" geographic features like a peninsula, or tract of bush isolated between two rivers.
McLean says they are calling it "1080 to zero". Once an area is cleared, it then needs to be maintained by a virtual border of traps along the boundaries or bridge crossings.
"It's more expensive. But not if it's also your last 1080 drop needed."
Then Cacophony-style technology is inevitably going to slash the labour costs of even conventional trapping. DoC is talking of 90 per cent reductions.
And in the long term, McLean says genetic engineering may be added to the arsenal.
Gene-editing – adding a gene for sterility to a population to let a species breed itself to death – looks about 10 years off.
Although, as that will be something developed internationally, it is likely only to be on offer for rats and mice – the kinds of animal considered pests elsewhere.
But generally, says McLean, the Predator Free 2050 promise is looking hopeful. And smart technology is going to have a lot to do with that.