To LiDAR or not to LiDAR

Bosch developing LiDAR sensors for autonomous vehicles
Many companies support use of LiDAR sensors for semi-autonomous vehicles, but Qualcomm unveiled its Snapdragon Ride Platform with a demonstration vehicle that didn't require LiDAR. (Bosch)

Qualcomm’s autonomous car demonstration in Las Vegas during CES 2020 relied on a Lincoln MKZ equipped with eight cameras and six radars, but no LiDAR.

RELATED: Qualcomm launches Snapdragon Ride for assisted and self-driving

The decision not to use LiDAR sensing surprised some analysts and other companies at CES, including Bosch engineers.

Sponsored by Digi-Key

Digi IX20 Secure LTE Router Available for Immediate Shipment from Digi-Key

The IX20 rugged, secure LTE router is a great choice for applications from basic connectivity to industrial-class and security solutions. Its high-performance architecture gives primary and backup WWAN over software selectable multi-carrier LTE.

“We see LiDAR as very critical to have a robust system combined with other sensors” including cameras and radars, said Jan Mertens, product manager for Bosch, in an interview. Bosch is active in a range of car-related sensing technologies and has actively been developing LiDAR technology for years.

“The vast majority of people in the industry would agree that LiDAR is critical” in autonomous vehicle technology, Jeremy Carlson, an analyst at IHS Markit, said in a separate interview.

Qualcomm announced the Snapdragon Ride Platform on Monday to support assisted and autonomous driving technology and decided to demonstrate it in its Vegas demonstration vehicle without LiDAR included. In a press conference, Qualcomm President Cristiano Amon said in response to a question that Qualcomm built its Ride platform partly to show LiDAR isn’t absolutely necessary.

“We’re looking for a number of technologies and the ability to use a connection depends on the application,” Amon said. “We also believe you can build a system without it,” referring to LiDAR.

Part of the reason for Qualcomm’s decision not to use a LiDAR sensor in its Vegas demonstration is mostly likely to help automakers reduce the cost of autonomous vehicles, as LiDAR is much more expensive than camera or radar sensors. A single LiDAR device could be 10 times or more the cost of a simple radar or camera sensor. Last year, Google Waymo officials said LiDAR sensor technology could cost as much as $7,500 per car.

Mike Feibus, an analyst at Feibustech, said that tech companies are working hard to reduce the cost of LiDAR and also to find ways to reduce the size and appearance of LiDAR sensors. Some LiDARS are large, sitting atop of a car with a tower that rotates rapidly and might be six to 15 inches in height. But some innovators are experimenting with splitting up a LiDAR sensor into smaller pieces. LiDAR works similar to radar but emits laser light to detect objects in the environment.

Feibus also believes LiDAR is a critical element in assuring safety alongside of camera, radar and even ultrasonic sensors. Tesla has not used LiDAR sensors in its cars, arguing that camera sensors are better. But LiDAR can complement a camera image, especially when there are opaque colors in the driving environment, such as a white truck set against a concrete overpass seen against a cloudy sky, Feibus said.

“LiDAR is a fool’s errand and anybody relying on LiDAR is doomed,” Tesla’s Elon Musk, said last April. He called LiDAR “expensive sensors that are unnecessary.”

Tesla is relying on a neural network and software to understand road conditions with camera and radar sensors instead.

Suggested Articles

Annual growth of 15% is expected for next decade according to Future Market Insights

New tech relies on time-of-flight sensor tech used in HoloLens combined with CMOS

With the rise of wearables, electronics are making their way into challenging environments unimaginable before, making protective coatings essential.