The villagers of Pampa Hermoza, Peru, gathered in the wooden bleachers on the local football field, their faces turned toward the sky and their legs dangling freely over the ground below.
On this late December afternoon, they hoped to witness the first cargo drone delivery of antivenom to their remote settlement in the Amazonian rainforest. At first they were excited, but as time passed and the drone didn't show, they grew anxious.
Even more frustrated was Patrick Meier, executive director and co-founder of WeRobotics, a global nonprofit employing drones and robotics for humanitarian causes. Meier worked for months with the Peruvian government, his own local Flying Labs team, Peru's aviation authority and a US drone startup to get this test up and running.
Getting the project to work could be a lifesaver. Local doctors in the region report up to 45 snake bites, many of which are life threatening, every month among thousands of indigenous communities. Delivering the right antivenom can take hours by boat and foot, which is why WeRobotics and the Peruvian Health Ministry were keen to find out if cargo drones could realistically complete the task in a fraction of the time.
It's yet another potential application for drones, which have gone from being military tools to Lady Gaga 's backdrop at the Super Bowl. In Japan, researchers are developing drones that pollinate flowers, augmenting a declining bee population.
Drones are everywhere, cheap enough for you to buy on Amazon for your kid, and smart enough that retailers like Amazon are testing them for deliveries. The drone market is expected to nearly double to $11.2 billion by 2020, according to Gartner.
Beyond antivenom, WeRobotics has research teams in Peru, Tanzania and Nepal to explore other uses for drones, including combating the spread of the Zika virus.
Some drones are already providing health care in remote regions, especially for disaster relief scenarios, according to Stelios Kotakis, an analyst from IHS Markit. The growth will largely depend on the outcome of projects currently in the "experimental stage," he said.
For WeRobotics, its field test in Peru can be chalked up as a success -- albeit in a slightly roundabout way.
Franky saves the day
Just as Meier began to lose hope, the villagers started calling his name. A speck appeared on the horizon. The drone finally arrived -- five minutes later than expected. The flight took 35 minutes.
Children clapped and cheered as it swooped in and skidded smoothly across the grass. "There's video of me jumping up and down a dozen times and yelling hooray, hooray, hooray," Meier said.
He had good reason to be relieved.
The few days prior were rife with frustrations as the original $40,000 vertical takeoff and landing drone that arrived from North America failed to complete even one of the many dozens of planned flights.
Fortunately, there was a plan B.
The team happened to bring an old $3,000 mapping drone along for the ride to the home base of Contamana, a six-hour riverboat journey from Pampa Hermoza. Battered, bruised and covered in gaffer tape, the mapping small fixed-wing drone, an Event 38 E384, was in such a state that Meier's staff nicknamed it Franky, after Frankenstein's monster.
For well over a year, Franky had been flying over the rainforest mapping and monitoring its environment. Fortuitously, the space housing the camera was exactly the right size to fit the mini cooler carrying the antivenom. With a few minor modifications, Franky suddenly had a new job.
There was only one problem: Franky had never flown more than 10 kilometers before, and it needed to go four times the distance.
"We had zero idea that this would work and zero guarantees," said Meier. The team decided to press on. "We might as well go all out."
By making it safely to Pampa Hermoza that day, Franky showed that there's hope for delivery drones to play a role in delivering health care in the Amazon.
"The use of the technology will be crucial in the future to improve access to health services, in particular highly specialized services," said Dr. Leonardo Rojas, former executive director of telemedicine at the Ministry of Health in Peru.
Shooting for the stars
But the plucky little engine that could hadn't finished demonstrating what it was capable of.
A return flight was planned. Franky was supposed to deliver a vial of blood for testing back at the hospital in Contamana. Meier and his colleagues swapped in a fresh battery for the return journey.
It was getting late. There were no records of any cargo drones being flown successfully at night over the Amazon. It didn't deter Meier and his team. "We figured we might as well go for gold here," he said.
They strapped a bike light to Franky's nose with yet more gaffer tape.
Just as the team was ready to launch, it realized Franky wasn't connecting to the base station back in Contamana, through which its flight path was programmed. A debate ensued. There was a small chance that if the drone got enough altitude it would manage to connect, said Juan Bergelund, Meier's counterpart back in Contamana.
Meier wasn't so sure.
"Even the other Peruvians were saying, 'Are you kidding me?'" said Meier. "We're going to lose the drone, we'll never recover it."
But Bergelund made the call and Franky flew into the darkness.
"All we see is this blinking drone fly off through the night." said Meier. "At this point it was beyond surreal, we were asking for trouble."
A bumpy landing
Bergelund's bet paid off, and Franky connected to the base station. But due to a slight miscalculation of the GPS coordinates, the drone ended up in a fight with a coconut tree near the field in Contamana.
If such an entanglement had occurred when testing the original $40,000 drone, it could have cost thousands of dollars to repair the damage, Meier said. For Franky, it meant -- yep -- another gaffer tape job and a $3 replacement part. The vial of blood stashed inside was unharmed.
Franky's success provided the whole team with a much-needed reality check on the approach it was taking to cargo drones.
"This awesome, humble Peruvian team had pulled off with just a few thousand dollars in a few hours what a major company had failed to pull off after a few months of planning with a drone that cost more than 10 times that price," Meier said
The report on the field tests, published Monday, concludes that for the price of the original $40,000 drone, WeRobotics could buy 11 E384 Franky-style drones, establishing a whole delivery network. Neither the report, nor Meier, would disclose the US startup that built the original drone.
But two successful flights are not enough on their own to justify building such a network. The team is out in Contamana conducting further tests, this time using E384 UAVs from the get-go. Presuming they are successful, they will conduct further tests in May that will see them push the range up to 100km.
Beyond the Amazon test, WeRobotics is exploring fighting Zika by using its drones to release millions of sterilized, male mosquitoes into an environment where they will compete with local populations to mate with females, reducing the numbers of insects in the subsequent generation.
WeRobotics will have to use larger drones, but it has learned its lesson about the kind of technology it employs.
"It doesn't have to be sexy, it doesn't have to cost $40,000," Meier said. "As long as it works, that's good enough."
Life, disrupted: In Europe, millions of refugees are still searching for a safe place to settle. Tech should be part of the solution. But is it?
Tech Enabled: CNET chronicles tech's role in providing new kinds of accessibility.