Coffee Talk: James Jung
This week we sat down with AEye’s head of development engineering, James Jung, who leads the company’s activities at the module level, including the scanner/MEMS, laser, receiver and optics. James speaks to why the right optical architecture is critical to achieving the advantages of MEMS, and what it’s been like to take 4Sight from prototype to product during his tenure at AEye.
“Scanning beam laser systems have been shown to deliver long-range performance, but ours is a step above the rest. That’s because scanning systems are commonly what we call “monostatic”. But for AEye, our scanning lidar system’s “bistatic” architecture makes us unique.”
Tell me about your role and your background
I am the senior director and head of hardware development engineering at AEye, where I’m in charge of the development and productization of our core opto-mechanical lidar components and the system box design. I also lead the scanner, MEMS, optics, mechanical engineering, receiver/detector and laser teams. I have a BS/MS in physics and applied physics and an MS/PhD in Electrical Engineering, and my thesis work was on photonics applications using MEMS technology. Prior to AEye, I was a scientist at a Department of Energy Laboratory working on high impact scientific applications using MEMS technology.
You’ve been at AEye since its early days. What’s it been like to see this tech go from prototype to product? What are the advancements you’ve been most proud of?
It has been highly rewarding to see prototypes using such sophisticated technology become a product. The multifaceted maturity that has been achieved in terms of the hardware and software/firmware has been a testament to the ingenuity and hard work from AEye’s engineers. I’m most proud of the transmission module (Tx)/scanner, which is an in-house development due to its unique architecture and state-of-the art nature. Several generations of this scanner have been developed, with each iteration improving on performance, size, cost, robustness and manufacturability. But those advancements aren’t unique to the scanner. We have pushed the envelope of what can be done with detectors and lasers, with joint development with our partners, in terms of size, performance and cost.
There’s been a lot of talk about MEMS. Why did AEye choose MEMS, and what advantage do you feel it brings to AEye, and to your customers?
MEMS has the advantage of scalability in size and cost, as well as robustness. That’s why you see its widespread use in applications for accelerometers/gyros, pressure sensors and other automotive products. But in order to utilize these advantages, you must be in the correct design space for the application. That means selecting the right optical architecture for the lidar. AEye’s bistatic lidar architecture allows for the MEMS mirrors to be very small – on the order of ~1 millimeter – resulting in devices that are highly agile and robust to automotive shock and vibration.
What else do you feel is distinctive with your underlying technology?
Scanning beam laser systems have been shown to deliver long-range performance, but ours is a step above the rest. That’s because scanning systems are commonly what we call “monostatic”. But for AEye, our scanning lidar system’s “bistatic” architecture makes us unique. Let me walk through some of the advantages.
One is the ability, as I mentioned previously, of utilizing MEMS in the right design space to make the mirrors themselves very small, highly agile and robust. Our resulting system can deliver fully configurable shot patterns that can trade off multiple performance parameters depending upon our customer applications.
Another is that the Tx and Rx are independent – allowing them to be independently developed and optimized without being hindered by the constraints of the other. This allows for design improvements or customization for specific customers and applications with a very short cycle time and production lead times. For example, the lenses are easily changed as components to cater to customer specific applications for range, FOV, size, etc.
Finally, the manufacturing and assembly is simple, due to the modular nature of our system. The optical components (Tx, Rx) only require mechanical tolerancing during assembly and do not require physical alignment. This architectural design choice allows us to resolve hardware and manufacturing requirements with software / firmware that can be matured and updated continually without hardware changes.
What’s your favorite mode of transportation?
Cars are my favorite mode of transportation. I love to drive, but would love not to have to drive long trips or through bad traffic. I’m looking forward to the time when AEye’s sensors are deployed broadly in cars to make these autonomous driving capabilities available in a safe manner.