Joseph Notaro is Chief Commercial Officer at Baraja, a global lidar technology company headquartered in Australia. He is an accomplished technology industry executive with 30+ years of experience in the automotive semiconductor market, and has held a range of management and executive positions at Onsemi; Fairchild, and STMicroelectronics.
He has successfully led R&D; business and market development; strategy, and worldwide sales organisations, and has extensive international experience across the Americas; Asia-Pacific, and Europe.
DVN-L: Joseph, will you tell us about Baraja?
Joseph Notaro: In 2016, our cofounders Federico Collarte (CEO) and Cibby Pulikkaseril (CTO) saw an issue in the lidar industry that they had observed from their experience in optical engineering from the telecommunications industry. They thought of a solution no one had tried before, five years before anyone else caught on to it. Baraja was founded to reinvent lidar from the ground up. Since then, we’ve set out to develop a lidar system that eliminates the most expensive; bulky, and failure-prone components of legacy lidar systems: mechanical scanning in the fast axis.
Our ambition is to make our cars and roads safer. There are way too many road fatalities every year. We are on the cusp of a revolution in mobility; a future where our roads are free of fatalities and accidents, and our vehicles are also safer while providing the driver more free time. But that’s only possible if we get lidar right. To get there, we need lidar to be as ubiquitous and reliable as the three-point seatbelt; airbags, or antilock braking systems. We need to deliver the same performance and scalability that carmakers demand of their own vehicles.
Successfully creating a safer and more autonomous experience is incredibly complex, and requires several different sensing modalities working in unison. Key lidar challenges in the past have been around reliability and scanning resolution, and these are the exact challenges our Spectrum-Scan™ is solving—but not the only ones.
DVN-L: Tell us more about that.
JN: Spectrum-Scan™ technology is based on a simple concept we learned in our early physics classes: light traveling through a prism is refracted at different angles depending on its wavelength. It is built on proven optical and photonics technologies originating from the telecommunications industry. Spectrum-Scan uses wavelength-tunable lasers to scan the environment, simply by changing the colour (=wavelength) of light. This unique scanning mechanism sets us apart from other lidar technologies, and solves the single design challenge of traditional lidars: the ability to scan a laser in a wide field of view.
While legacy systems rely on fragile rotating lasers and/or oscillating mirrors, we simply change the colour of our laser light and let our prismlike optics do the work. Steering a laser beam, though, is only half of a lidar. Ranging is the other half, and a key lidar function. We have implemented RMCW (random modulation continuous-wave) as our ranging solution. The result is a lidar technology that has no fragile moving parts in the fast axis, and is compact and completely free of interference from other lasers. This results in unprecedented range, precision and reliability in even the most extreme environments.
DVN-L: What sets Baraja apart from other lidar suppliers?
JN: Technology, ecosystem integration, and financial backing.It’s a combination of all of those, starting with our technology which can deliver the end-game lidar required for seamless and invisible automotive integration.
We understand automotive needs and the price targets required to enable mass deployment of safe autonomous driving functionalities. We are building the best sensor to meet the specifications of major automakers so we can deliver a fully-integrated lidar solving the issues of legacy lidars.
DVN-L: What made Baraja decide to go for mobile applications, especially lidar sensors for ADAS and AVs?
JN: Our Spectrum-Scan system is a unique architecture using wavelength-tunable lasers within the 1,550-nanometre C-band to scan the environment. This design innovation means we’ve removed the most expensive; bulky, and failure-prone components in a lidar system—mechanical scanning in the fast axis. This makes it perfect for mobile applications, whether in a heavy industry environment or out on the open road.
We have a number of static and infrastructure-related test cases under way that can leverage our unique technical capabilities, such as foveation and doppler, to solve novel sensing challenges.
Very low power dissipation and audible noise enabled by Spectrum-Scan also allows our lidars to be mounted behind the windshield, making it the lidar of choice for mass deployment on passenger vehicles.
DVN-L: One of your lidar’s core strengths is suitability for harsh environments like mining. How does that inform your work on automotive applications?
JN: Certainly, the environment is a key difference between mining out in the Australian Outback and driving on city streets. However, both use cases are looking for the same product features of performance; reliability; shock resistance, and sensitivity control. The overarching robustness of the system is crucial. Testing our sensors in these harsh environments has shaped the way we think about these product characteristics. This has been a great learning experience allowing Baraja to design a lidar system that can withstand the harshest environmental conditions.
DVN-L: How have you lowered your system’s power consumption, and how important will that be in the automotive applications?
JN: Power consumption is a key performance parameter along with range, resolution, etc. It continues to be one of the key areas of improvement across the industry. With Spectrum-Scan’s true passive fast-axis scanning technique, Baraja have brought down power consumption to meet and exceed current automotive specifications allowing a seamless integration inside the vehicle, behind the windshield. Another important, but often overlooked, parameter is the audible noise generated by a lidar sensor. Even in this case, Spectrum-Scan allows for a significant reduction in the total system audible noise and will not disturb the driver or passengers when mounted inside the vehicle.
DVN-L: What other features do you think are needed for automotive lidar sensors?
JN: Baraja’s proprietary solid-state scanning platform was designed from the ground up to completely rethink how cars see the world around them to enable true autonomy. The lidar system enables the range, resolution and performance required for safe deployment of autonomous functionalities, without the traditional tradeoffs faced by other legacy lidar technologies. In addition, Spectrum HD 25 delivers the world’s first lidar system combining per-point doppler capability at the hardware level with a tuneable-wavelength laser and RMCW ranging to deliver unparalleled performance and accuracy at range and speed. Per-point doppler velocity is instantaneous; precise, and requires no additional computation power.
This allows to enhance the performance of the perception stack. Grouping points and objects by velocity helps segment and predict trajectory of objects, leading to better perception outcomes. The data can be used to not only detect movement at a distance, such as a pedestrian stepping out of a stationary crowd into the road, but also help determine prediction and intent of movement.
As more and more lidars are deployed on our roads, immunity to interference will also be as important as other parameters. In several studies, it has been proven that lidars implementing RMCW exhibit at least a 2× improvement versus traditional FMCW or ToF systems.
DVN-L: This past October, Baraja announced a development agreement with Veoneer to develop a lidar system for a major automaker. What can you tell us about that?
JN: Our partnership with Veoneer is critical to this agreement. Veoneer will help us scale the development of our Spectrum HD25 platform and enable future opportunities for high-volume automotive integration. The support for Spectrum HD25 from Veoneer; the automaker, and partners is a significant validation for Baraja and its path towards enabling more autonomous vehicles.
We are excited about the progress we are making and the market response for our Spectrum HD25. Stay tuned for further updates in 2023!
DVN-L: How do you forecast the automotive lidar market’s development? Do you expect L3 to be the biggest market segment, or will L1&2 ADAS applications to take the bigger share of lidar sensor applications?
JN: We see growth across each of these markets in the next five years, with advanced ADAS solutions presenting the largest opportunity for automakers to bring the latest safety technology to a wide range of their product line. While we do not have a crystal ball to reliably predict the distribution between L3 and L1&2, we are certain that lidar will be deployed on more and more vehicles.
We need to dramatically reduce the number of fatalities on our roads, and the existing sensor suite, while contributing to safer roads for everyone, is not sufficient to reach our goal of zero road fatalities. lidar is complementary to other sensors. Either making existing vehicles safer—moving from L1 or L2 to L2+ or L3, or enabling more autonomous driving functionalities—we are certain that lidar is a must in both cases. Speed of deployment regionally will vary based on the automakers’ appetite for innovation and legislation. We can expect mass deployment in China to lead, followed by Europe and USA, and then Japan.
DVN: Joseph, thanks so much for talking with us! We hope to see you and your Baraja colleagues at our upcoming DVN-Lidar events.
JN: Thank you very much for the opportunity. I am looking forward to working together again in 2023!