Computer Vision News - April 2021

351 Iris Automation Looking to the future, although they have tested on a huge variety of scenarios, Alejandro says they won’t stop there, and will keep generalizing so that they can guarantee performance in many different locations and conditions. He is keen to be able to fly the drones at night by modifying the system to use thermal camera instead of visual spectrum ones. Also, due to the use of geometry, there are certain conditions that present challenges he would like to solve, such as flying over water where the ground is moving . Iris Automation has around 50 employees, with 10 people in Alejandro’s perception team. Customers are already using one version of their detect-and-avoid solution, and they have other products coming out later this year, so watch this space. Finally, we ask Alejandro what he finds most exciting about his work. “ There are so many things that I could mention,” he smiles. “One of them is that the solution is currently being used. It’s so exciting that I’ve put something out into the world and our customers are using it in real operations. Also, this industry is less served by academia, so a lot of what we do is novel. We don’t just take a paper and implement it. A lot of things don’t work in the way they have been designed. There is a lot of discovering and a lot of inventing. That is extremely exciting!” “… now we’re at the point where we can generate data at a large scale.”