Aerialtronics develops smart camera

By Ann Bailey | December 10, 2015

Aerialtronics has developed a smart dual camera which has both digital and thermal sensors. 

The Netherlands-based company is using a NVIDIA embedded platform in the project, Aerialtronics said in a news release.

The prototype can process both raw data streams from the day-to-night camera which enables Altura Zenith UAS to be used more effectively in a variety of onboard computing tasks. The tasks range from object detection and recognition to path planning for security tasks, tower inspections and 3D mapping in mining, the news release said.

“We have been developing dual cameras that allows users to switch between daylight and thermal video streams for some time,” Jan Wouter Kruyt, head of Aerialtronics research and development said. “Now we have put these data streams into the NVIDIA chip and it means users can do some very smart things with it."

NVIDIA, which develops chip technology, has delivered embedded molecules in the Tegra family. Kruyt anticipates that will lead to a new generation of learning device. An advantage of the open chip is that it can be injected into any algorithm in which Aerialtronics or the operator chooses to embed it.

Integrating a daylight and digital thermal camera with NVIDIA’s Tegra K1 chip means that the operator can analyze both data streams and use the information, Kruyt said.

“Right now it allows us to do smart things with the images,” he said. “Eventually, we will be connecting this to the flight computer and then the whole drone will become smart enough to respond to what it is seeing.”

The Altura Zenith will use the video streams and information it observes to adjust its flight plan, Kruyt said. Operators, for example, can fly the UAS over a parking lot and find the vehicles which have the hottest engines and then scan their license plates for payment or security. The video stream also has the capability to recognize faces, measure vehicles’ speed and locate leaks in oil and gas pipelines, Kruyt said.

UAS operators now use data from UAS for a variety of inspections, including power lines, bridges and oilfield and gas pipelines. With the addition of the camera, it is able to recognize detailed shape, type and the material used in those structures at the same time enhancing 2D and 3D mapping.  

Kruyt believes the Altura Zenith is a major step forward in the level of autonomy for UAS. Integrating machine vision on an embedded chipset creates something more intelligent than simply pre-programing a UAS and setting it on a flight path, because with the addition of the chipset, the UAS can identify what is around it, Kruyt said.

For example, if a UAS operator drives to a utility pole, mobile phone tower or house, they only need to press a button, Kruyt said. The Altura Zenith will fly out with the camera, see what it needs to inspect and perform the routine inspection. It can even design its own mission based on what it sees and verifies it has completed its mission before returning to base, Kruyt said.