Lidar Startup AEye Develops an Independently Verified 1,000 Meter Range Lidar Sensor for Autonomous Vehicles

author: Eric Walz   

As more vehicles come equipped with advanced driver assist systems (ADAS) and higher levels of autonomous driving, there is a new crop of lidar startups vying to supply the technology to the auto industry, including California lidar developer AEye. 

AEye is among a handful of lidar startups, including Luminar and Innoviz, as well as lidar pioneer Velodyne Lidar Inc., that are working on advanced lidar systems for automotive ADAS as well as autonomous vehicles.

Lidar has become an essential sensor for developers of self-driving vehicles over the past decade. In a self-driving vehicle, the technology is often combined with other vehicle sensors, such as cameras and radar. For a vehicle capable of autonomous driving, the lidar acts as an extra set of eyes on the road.

However, one of the biggest challenges for lidar developers is designing a lidar sensor that is able to detect objects at a longer range without sacrificing resolution until recently. AEye announced on Wednesday that it has made a major breakthrough in long-range lidar performance. 

AEye's 4Sight M lidar sensor was able to detect objects greater than 1,000 meters away, while maintaining a 10Hz Scan Rate (10 frames per second). 

The 1,000 meter detection range was independently validated by VSI Labs, one of the nation's leading independent evaluators. The company is one of the auto industry's top advisors on autonomous vehicle (AV) technologies, supporting major automotive companies and suppliers worldwide.

VSI Labs reported that AEye's LiDAR system achieved "breakthrough capabilities in range, high-resolution performance and safety for automotive applications or autonomous trucks."

The tests were conducted last month using a standard version of the company's 4Sight M sensor on the runway of an airport in Byron, California in order to better isolate targets and determine AEye's lidar sensor performance.

For the VSI range test, the company placed two vehicles (a Chevy Bolt EV and Mercedes-Benz Sprinter van) at the end of the airport runway and the lidar sensor at the opposite end over 1,000 meters away.

At this long distance, AEye's lidar was able to put more than 50 lidar points on the van, and over 35 points on the Bolt, which is roughly 4 times the range of conventional lidar units, according to VSI.  AEye's perception system can even detect objects as small as a brick in the road at a distance of 160 meters. 

For developers of self-driving vehicles, this long of a range gives an autonomous vehicle's software more time to react to hazards.

"The tests delivered impressive results," said Phil Magney, founder and president at VSI Labs. "We monitored the LiDAR performance tests and verified that in the range test, two vehicles were clearly identifiable and visible at over 1000 meters with dozens of detections on each.

AEye was recently awarded 17 new patents for its lidar technology with more pending.

AEye's 4Sight M lidar sensor is entirely configurable with software, so it's adaptable to meet the specific requirements of different automakers or individual vehicle models. The software can be used to program different levels of resolution and range depending on the application, making it a versatile sensor for automakers.

Lidar sends out pulses of laser light which bounce off objects to create 3D images like this one. (Photo: AEye)

What is Lidar?

The term lidar is an acronym for "light detection and ranging" and it's an important part of most autonomous driving systems. Lidar is used to create a 3D rendering of a vehicle's surroundings by sending out pulses of laser light which reflect back off objects. The 3D representation of an environment created by the reflecting laser light is known as a lidar "point cloud", which shows the road ahead including other vehicles, pedestrians, buildings or road hazards.

By measuring differences in return times and wavelengths in the light that's reflected back off objects, AEye's 4Sight M lidar sensor has the capability to detect an object, as well as determining its velocity, path and whether it will pose a threat.

"Intelligent" Scanning Using Lidar

However, what makes AEye's lidar technology unique is its ability to "intelligently" scan each scene. Unlike typical lidar that simply scans scene passively and relies on other sensors and software to determine what objects are in the field of view, AEye's active LiDAR scans the entire scene, using AI to intelligently focus on only what matters, such as a pedestrian or vehicle in the distance. This leads to and faster decision making in complex scenarios. It also requires less compute power, since the irrelevant data in each scene can be disregarded.

AEye's lidar sensor achieved a scan rate that is nearly 9 times greater than a human eye, which scans at the equivalent of 27Hz. It also scans 10-20 times faster than competing LiDAR systems to better track, determine and predict where an object is going. 

This intelligent scanning ability enables higher levels of autonomous functionality from SAE levels 2 through 5, according to AEye.

"At AEye, we set out to create low cost and high-performance sensing systems, driven by artificial intelligence, that would exceed the capabilities of the human visual cortex," said Luis Dussan founder, president and CTO of AEye. "By leveraging principles from automated targeting systems and biomimicry, we designed deterministic artificial intelligence sensing systems that redefine traditional LiDAR performance metrics, while adhering to automotive functional safety standards and delivering the advanced performance required to address the most challenging situations." 

By delivering more than 4X the distance of conventional LiDAR, AEye's system provides unique stopping and avoidance capabilities at highway speeds. AEye's lidar technology is well suited for collision avoidance systems, such as Automatic Emergency Braking (AEB) features. But it can also support commercial long-haul autonomous trucks currently in development that require much greater stopping distances.

"As impressive as these performance metrics are, what they enable is even more impressive," said automated trucking industry technology expert Richard Bishop. "The performance of AEye's 4Sight sensor will help address critical challenges, such as identifying potholes, stationary road debris or stalled vehicles in time to change lanes, or gaining clear visibility to on/off ramps beyond 300 meters to ensure safe ingress and egress on freeways. These are game changers for the trucking industry."

VSI Labs also conducted a windshield test in order to evaluate whether an AEye sensor mounted behind a glass windshield would have any impact on performance. 

For the test, a piece of windshield glass was placed in front of the 4Sight M sensor, with a target placed at 100 meters distance and from various angles. The same scene was analyzed before and after the windshield glass was placed in front, with little impact on detection performance, according to AEye.

The compact 4Sight lidar sensor can also be mounted inside the vehicle behind the windshield with less than a 10% degradation in performance, AEye said. This provides much greater flexibility for automakers in integrating the lidar into their vehicle designs, instead of having to install lidar on the roof of a vehicle or in the front bumper.

In Oct 2020, AEye announced an investment from Continental AG, one of the world's biggest auto suppliers. Germany-based Continental has accumulated over 20 years of expertise in LiDAR sensors. The amount of the investment was not disclosed.

Continental will utilize AEye's lidar technology to create an automotive-grade lidar unit for automakers. The first series production of a compact lidar system from Continental incorporating AEye's advanced technology is scheduled for the end of 2024.

AEye is also backed by leading investors Kleiner Perkins Caufield & Byers, Taiwania Capital, Hella Ventures, LG Electronics, Aisin, Intel Capital, Airbus Ventures, and others.

Eric Walz

Originally hailing from New Jersey, Eric is a automotive & technology reporter covering the high-tech industry here in Silicon Valley. He has over 15 years of automotive experience and a bachelors degree in computer science. These skills, combined with technical writing and news reporting, allows him to fully understand and identify new and innovative technologies in the auto industry and beyond. He has worked at Uber on self-driving cars and as a technical writer, helping people to understand and work with technology.