This past month, thousands of attendees converged at AUVSI XPONENTIAL, the prominent unmanned vehicle conference and exhibit. Leonardo’s Jean Michel Maillard presented a technical talk, Increasing Field of View and Reducing Cost for Laser Sources for Lidar, and met with system integrators in both automotive and UAV to better understand emerging demands for laser systems.
Nearly 50 people attended the technical presentation, from radar, lidar and UAV backgrounds, with a wide range of questions:
Are laser diodes able to achieve higher-power, long-range lidar for UAV?
Leonardo is already producing edge-emitter direct diode laser sources for illumination of targets for the military, but these are too heavy for use in UAVs. High power fiber coupled lasers being designed for directed energy applications will be used for low weight light sources for UAV lidar in the future.
Do you see the answer as radar or lidar, or both as complementary technologies?
To those who’ve studied the fatal Tesla car crash in 2016, in which the driverless car used cameras, radar, and ultrasonic sensors, but lacked lidar, it is clear that both radar and lidar improve detection and safety. Lidar more readily discriminates between a white truck and clear sky, the issue in the Tesla crash, while radar offers advantages in poor weather conditions.
| Ultrasonic | Radar | Camera | Lidar | |
| FOV / Range | - | Good | Good | Excellent |
| Spatial Resolution | Good | Good | Poor | Excellent |
| Angular Resolution | - | Poor | Excellent | Good |
| Poor Weather (Rain, Snow, Fog) | Good | Excellent | - | Good |
| Night / Direct Sunlight Disturbance | Excellent | Excellent | Poor | Excellent |
| Object Discrimination | - | Poor | Excellent | Good |
A comparison of imaging and sensing sources, such as ultrasonic, radar, camera, and lidar.
We don’t see the answer as an “either/or.” We see it as our challenge to work with system integrators to drive down costs for lidar with technologies like VCSEL (vertical cavity surface-emitting laser) diodes.
How will lidar systems discriminate signals?
Right now, they can’t. The detector on your car cannot tell if the signal is from your car or mine. Detectors must know which laser is which. They’ll require a unique sequence or coding. It’s one of the issues to be resolved before widespread adoption of lidar.
What approaches to lidar do you see prevailing?
A few of the lidar companies in attendance at XPONENTIAL are working with MEMs to scan the field for wide field-of-view. Moving to solid-state flash design would help them increase reliability without sacrificing performance.
The laser community as a whole is struggling with high-power, low-cost laser sources for large-scale commercial applications like automotive, because they’re used to more technical, higher-cost and lower- volume applications. It will take a real partnership to innovate and overcome market barriers to cost, performance, and regulation. I think our work in edge-emitter and VCSEL diodes mounted directly on the driver board has a lot to offer.
Interested in learning more about automotive lidar? Download our whitepaper, The Race to the Starting Line: Edge-Emitting Diode Lasers vs. VCSELs for the Automotive Lidar Market.
