The human element in driving presents both a risk and a level of checks and balances that would disappear in an autonomous-only world of transportation.

From addressing safety concerns in human traffic-related accidents to working on hazard recognition in self-driving vehicles when there is no driver to react to immediate dangers, vehicle technology startup Aurora is working to deliver the benefits of self-driving technology realistically.

The Consumer Technology Association (CTA)® 2019 Innovation Entrepreneur Award Startup of the Year winner has been developing the Aurora Driver, a platform made up of software, hardware and data services to allow a vehicle to drive itself. The team is working on technology that will make the cars as safe as possible for occupants of the vehicle, surrounding vehicles and their passengers, and pedestrians.

“Safety isn’t just a theoretical concept or a talking point we put on our website,” said co-founder and CEO Chris Urmson in i3 magazine. “We have human lives in our hands, and that’s something we never forget.”

Nothing Overlooked

Aurora is scaling advancements in machine learning and artificial intelligence, engineering, and more to make sure self-driving vehicles offer the best and most secure passenger experience.

“We’re blending state-of-the-art machine learning and engineering in our perception system to help the Aurora Driver handle the unexpected,” Urmson said.

Among the technology applied are three that will help vehicles perceive and react to on-road situations on par with or better than their human counterparts.

Localization and maps: Detailed maps that are frequently updated help the Aurora Driver plan ahead for best routes and also determine where the vehicles are at any time, even in environments without GPS.

Perception: Sensor data enables a perception system that can detect all objects around the vehicle, such as pedestrians or unexpected obstacles, even if the vehicle is unable to recognize what the object is. Aurora calls the system no measurement left behind.

Planning and control: Machine learning software forecasts the intention and motion of objects in driving environments; determines what the Aurora Driver should do in response; and helps the vehicle throttle, brake and steer accordingly.

Tied together through the Aurora Cloud, the vehicles are easily managed, dispatchable for ride-sharing and business transports, and adjusted as tests and simulations continue.

Social Impact 

The Aurora team is on the road toward replacing error-prone driving with technology that never gets distracted. With the Aurora Driver system, Aurora hopes to make concrete differences, expanding transportation options to those living with a disability that make traveling difficult.

Shared self-driving vehicles will also free up garages and parking lots, creating opportunities to revitalize precious city real estate.

At the core, however, will always be Aurora’s dedication to making roads safer.

“We will deploy Aurora Driver when it’s safer than a human driver,” Urmson said.

Read more about Aurora and the Innovation Entrepreneur Awards in i3 magazine