Drone Artificial Intelligence To Prevent Collisions

By admin | Technology

Aug 26

Artificial intelligence is a subject that has interested many scientists and everyday folks alike, movies have long showed this technology being used for various uses. Now two Vancouver-based entrepreneurs are bringing artificial intelligence into the world of drones to prevent them from mid-air collisions.

The project deal with industrial drones and is aimed to help the UAVs avoid collisions with other drones, planes or bird hits. Without a technology in place to prevent collisions drones could be prone to higher risks.

The need for industrial drones for mining exploration, pipeline inspection, agricultural surveying, forestry, or even package delivery, has created the need for this technology.

“We are building collision avoidance for industrial drones,” said Alexander Harmsen, CEO and co-founder of Iris Automation said. His company’s artificial intelligence computer blends real-time images and 3D maps to track incoming objects.

“That’s when it hijacks the auto pilot on board and spears it out of the way, or tells it to go for an emergency landing,” Harmsen said.

Iris Automation was started by Harmsen, who graduated from UBC last spring, along with his graduate James Howard. The team also won a spot at seed accelerator Y Combinator.

The artificial intelligence industry itself is now surging ahead as the need for the services are growing. Iris Automation is only one of several companies which are working on similar technology.

As the drone market expands into various sectors, other services related to drones too are required to keep up with the pace of the expanding sector. Analysts have predicted the UAV market to reach $127 billion.

One of the main challenges the drone industry is now facing are the mid air collisions, it is essential to prove to regulators that the drones are safe for use and tested.

“Some people call it the ‘holy grail’ of the UAV world,” Paul Di Benedetto, chief technology officer of Drone Delivery Canada said. His company plans to begin automated door-to-door deliveries by 2018.

Regulations in the emerging drone industry are constantly changing, hence companies are just looking to create the best possible solutions and later deal with the regulations.

Drone Delivery Canada is now testing drones and establishing new customers as the new regulations come into effect. Right now Transport Canada only allows UAV flights within line of sight, thus making sure the drone operator can see the drone and ensure it will not collide with another flying object.However once new regulations come into place and drones are allowed to fly out of sight it is essential to ensure they don’t cause collisions and maintain safety. This is where artificial intelligence is necessary.

The UAV has to be able to interact and react to what’s going on out there

Paul Di Benedetto, Drone Delivery Canada

“The UAV has to be able to interact and react to what’s going on out there,” Di Benedetto recently told a major network.

Regulators across the world have been debating drone rules as the regulations for traditional aircrafts cannot govern them, yet they cannot be left un-governed either. New regulations from Transport Canada are expected in late 2017, but experts still feel it would not include permission for drones to fly out of sight.

Canadian drone experts and users feel the country is still lagging behind as some countries like Japan have drones delivering orders made online of various things including prescription drugs and bottles of liquor.

Close here, the Federal Aviation Authority in the U.S. brought out a set of regulations dealing with commercial drones recently.

Companies that have been chosen as part of U.S. Pathfinder program meant to test commercial drone technology are allowed to fly out of sight. Those companies however don’t use artificial intelligence to prevent collisions.

They employ a combination of geofencing, which is a digital wall that blocks drones from designated airspace, and tapping into an existing aviation network to receive information on the drone’s flight and whether another aircraft is approaching its airspace.

However that solution may not work for everyone and could provide to be expensive and impractical, experts say, adding that artificial intelligence is just one of the solutions to the problem of mid air collisions.

Let us wait and see what drone technology makers have in store for us next.

Leave a Comment:

Leave a Comment: