For anyone questioning the strides being made in the world of autonomous vehicles, meet FLIR Systems. The California tech company has been developing its thermal sensor camera technology to enhance capabilities of self-driving cars.
- FLIR Systems is showing off its thermal sensor cameras for autonomous vehicles on the streets of downtown Los Angeles during the year’s L.A. Auto Show.
- Thermal sensor cameras have proven to be four times more reliable than other sensor systems at classifying pedestrians.
- FLIR says the cost of its thermal sensor technology makes it very feasible for self-driving vehicles.
FLIR has been developing thermal sensor technology that could impact the future of self-driving vehicles. (Photo: Marcus Amick)
Ride actually had an opportunity to experience FLIR’s thermal technology during a demo ride at the 2019 Los Angeles Auto Show.
Widely known for its military applications, thermal sensors focus on heat emanating from objects, which makes the technology more efficient in detecting obstacles on the road across various depths of field. The resolution of the images captured with thermal cameras is much more detailed than radar and lidar systems, especially when it comes to detecting pedestrians.
“Pedestrians all emanate heat. So, you’ll see pedestrians in the thermal quite easily, day or night, driving into the sun, in shadows, it doesn’t matter,” said FLIR Chief Engineer Chris Posch, who conducted our demo. “The thermal will provide you the best sensor at detecting pedestrians and making your car safer.”
The technology has proven extremely reliable in the classification of pedestrians, bicycles, and vehicles in total darkness at nearly four times the distance of a typical vehicle’s headlight range, along with other challenging lighting conditions such as fog, smoke, shadows, inclement weather and sun glare, according to FLIR.
FLIR’s thermal cameras (in the right-hand corner of the screen) have proven to be highly effective in detecting pedestrians. (Photo: Marcus Amick)
Cutting the cost
FLIR’s thermal sensors are currently being used for driver warning systems for a number of car companies, including General Motors, Volkswagen, Audi, Peugeot, BMW, and Mercedes-Benz. But one of the biggest challenges in adopting thermal sensors for autonomous vehicles has been finding a way to make the technology economically feasible for self-driving cars.
“The technology has advanced so much and the cost has come so far down that instead of thousands of dollars for a thermal camera, they are now hundreds of dollars,” said Posch. “That’s the key. This technology is now affordable for mass market.”
FLIR has been able to keep the cost of its thermal sensors relatively low for autonomous vehicle applications. (Photo: Marcus Amick)
As part of its development strategy, FLIR is currently collecting thermal data in a number of major metropolitan areas, with a focus on covering common seasonal driving conditions at all hours of the day. The company’s future strategy includes gathering data in several large U.S. and international cities with regional customization to offer accelerated thermal testing.
The overall goal, explains Posch, is to integrate thermal cameras in the overall sensor systems for autonomous vehicle companies, not as a replacement for lidar and radar technology. “All of the sensors have their own importance,” he says.
FLIR already has a major self-driving car company looking to adapt the technology for use in their autonomous vehicles and plans to make an official announcement in the near future.
WHY THIS MATTERS
Given that pedestrian detection continues to be one of the primary concerns with autonomous vehicles, FLIR’s advanced thermal camera technology is poised to play a pivotal role in helping to push the technology forward. The mass application of thermal sensor technology in the space could potentially anchor the future of self-driving vehicles.