INVENTING A BETTER DRIVER

The biggest change to mobility in a century.

In 1908, the first Model-T rolled off the assembly line and the world changed overnight. Since then, how we work, play, travel, and even where we live have been shaped by the automobile. But despite all the advances, there are massive challenges. 1.2 million people die each year from human driver error. Commutes are longer and car prices are higher than ever. It’s time for another automotive revolution.

DARPA: Making the impossible, possible

In 2007, self-driving cars seemed like science fiction. But to the early pioneers of autonomous technologies, the concept was very real. Teams competed in multiple DARPA Challenges, failing at first but improving with each Challenge, culminating in the breakthrough DARPA Urban Challenge. Team VictorTango, a collaboration between Virginia Tech and Torc, was one of just three teams to successfully complete the challenge. Most teams were pure academic pursuits from respected universities, but Torc was founded with the purpose of refining and commercializing self-driving technologies. And we’ve been doing just that for over a decade.

Early adopters: mining + defense

In life, most decisions come down to a risk/reward choice. For the defense and mining industries, the risks of autonomy, even in its very early forms, were far outweighed by the potential benefits. In mining, huge haul trucks operating in dirty, dangerous environments were a major liability, so we worked with Caterpillar (CAT) to develop an autonomous hauling system that has removed drivers from those situations and improved operational efficiency by 20%.

Intelligently saving lives

Defense partners needed innovative ways to combat improvised explosive devices (IEDs), so we developed autonomous vehicles that could detect explosives while troops stayed at a safe distance. In addition to saving lives from IEDs, we also worked with defense partners to develop autonomous off-road systems that could carry equipment and act as scout and reconnaissance assets for troops. By reducing both the risk to troops and the workload they face, these systems have improved readiness and effectiveness drastically. Most importantly, these projects helped us form and refine the core of the IP that drives our systems today.

Build – Test – Learn

Rapid innovation isn’t about moving forward recklessly. It’s about following a build/test/learn cycle to quickly test advancements in the real world. For a decade, we’ve used this concept to continue to refine and expand our self-driving technologies in every environment imaginable, from off-road to highways to congested city streets. We believe in controlling the full software stack, always using raw, unprocessed sensor data and a mix of sensor technologies, including lidar, radar, vision (cameras), and others. Software and hardware needs drive each other in a cyclical fashion as we develop and test new systems. And we’re constantly researching and integrating emerging technologies such as deep and reinforced learning to help make our systems smarter, faster, and more reliable.

Project Asimov

In addition to working with commercial partners for the past decade, we’re continuing to test and improve our self-driving capabilities with Project Asimov. For initial tests, we often work with the Virginia Tech Transportation Institute (VTTI) and their Smart Road testing facility. Testing in controlled conditions lets us safely explore the boundaries of sensor and system capabilities.

Exploring corner cases

But for the ultimate real-world test, we’re taking our Asimov test vehicles out onto real roads to deal with the kinds of dangers human drivers face daily. From swerving drivers, traffic jams, and poorly-marked roadways to dismal weather conditions and extended driving sessions that would tire any normal driver, we’re continually putting our test vehicles into ever-more-challenging scenarios. In each test, we learn and invest that learning back into both our commercial and testing programs in a never-ending improvement cycle.

Changing the world through partnerships

Our goal at Torc isn’t just to develop some new technologies. To fundamentally shift how society moves, we need to partner with innovative companies who see the promise of the self-driving revolution. Together we can save lives and eliminate the drudgery of commuting. We can fundamentally improve our society, but it all starts with developing a smart technology roadmap and determining how Torc can accelerate your progress. Consult with us on your robotics and autonomous technologies needs. The revolution is coming. Are you ready?

NancyOur Work