Oculii’s software may one day help your car drive itself
Your car and millions of others have radar sensors installed to assist with emergency braking, blindspot detection and parking-assist systems, but those sensors could one day give your car the capability to drive itself thanks to Oculii. The Dayton-based startup is developing AI software that enhances the resolution and performance of your vehicle’s radar sensors up to 100 times their normal capabilities. Oculii’s technology is giving a boost to existing vehicle radar sensors, but the company plans commercial application as early as 2023 to pave the way as a leading software supplier for autonomous vehicles.
“Think of us as the ‘eyes’ for these autonomous systems that we’re going to have in the future,” said Steven Hong, CEO and co-founder of Oculii. “We’re building the ‘eyes’ so that the sensors, or the ‘brains,’ can see the world and react. No matter how smart sensors in an autonomous car are, if the car is ‘blind,’ I wouldn’t trust it to drive itself. We’re building software with high perception capabilities for existing hardware that will operate in these autonomous systems like self-driving cars, robots and drones.”
Traditional sensors use radar to detect objects by sending a wave of energy with a repeating signal reflected by objects in their path. More accurate object detection from a repetitive signal radar requires more antennae, which can take a lot of space and can be very costly to apply to commercial vehicles. Oculii’s software uses dynamic waveforms and signal patterns that learn from the environment and send different information at different times. Their software requires smaller and fewer antennas to detect objects and allows the sensors to generate significantly higher resolution.
“We see very accurate and high-performance radar sensors being used by the U.S. military, but their radar sensors are huge with many antennae and not something that could be cost-effectively applied to commercial use,” said Hong. “There has been a trade-off of radar design and radar accuracy in cars where the radar installed in your vehicle is more affordable, but has low resolution and low performance. It is good for safety applications, like emergency braking and collision warning, but you wouldn’t trust that system to drive itself. We are trying to solve that trade-off by increasing the accuracy of radar technology while preserving the cost of radars used commercially.”
A Dayton native himself, Hong grew up in Ohio but left for California to receive his PhD at Stanford. In 2015, Hong was working for a venture capital company that was investing in autonomous technology. He saw the potential of the automation industry, but wanted to lead in the growing automotive automation market. He founded Oculii as a partnership with his father, Lang, an experienced radar expert and former professor and dean of engineering at Wright State University. Hong decided to return to Ohio, and the two looked no further than Dayton to start their company.
“We wanted to build a team that could take Oculii technology further, and Ohio has a great concentration of engineering and technical talent that we wanted to leverage,” said Hong. “With Wright Patterson Air Force Base in Dayton, you couldn’t ask for a better place. All the ingredients are here and we are combining them in a way that allows us to capture the value and share it with everybody in Ohio. I lived in California for 15 years, and the quality of life here in Ohio is so much better. You can have your own space and things are much cheaper, so you have the opportunity to build a great company here.”
Oculii has worked with many large-scale robotics companies and has successfully implemented their software to develop autonomously driven robotic systems and drones, but their focus today is more geared towards the automotive market. They have raised over $75 million to date, and closed a $55 million Series B round of funding in February. Oculii‘s partners include several major global manufacturers, Great Wall Motors, the largest Chinese manufacturer of SUVs and trucks, Hella and Samsung and two Silicon partners, Texas Instruments and Infineon. With the support of global companies, Oculli is now preparing to launch their software for commercial use in vehicles as early as 2023.
“Our Series B close was a major validation and statement that this technology is ready, and hopefully soon it will benefit every single car,” said Steven Hong. “We want to provide people the ability to concentrate on the road and not have to worry about something in their blind spot. Our software is meant to complement the human driver, rather than replacing them, and to ensure safety, comfort and confidence while driving. Almost 2 million people worldwide die every year from car accidents, and 94% of these accidents are caused because of human error. If our software can provide fully autonomous capabilities that could eliminate human error, then we can save all those lives, and it would make our efforts worth it.”