For self-driving cars to evolve from futuristic possibility to widespread reality, the technology outside the car will be as critical as the technology inside it.
/ APRIL 7, 2017
Much of the excitement swirling around the emerging autonomous vehicle movement naturally involves the cars themselves.
These sophisticated, self-driving autos — now being trialed and modified on private test corridors and actual roadways across the world — are equipped with highly advanced and intricate sensors, inertial navigation systems and other incredible technological innovations.
But in order for autonomous driving to evolve from a sci-fi, futuristic possibility to widespread reality, the technology outside the car will be as critical as the technology inside the car.
Welcome to the emerging era of connected infrastructure, the linchpin to the digital, unified ecosystem required for the deployment — and ultimate success of — automated vehicles.
While perhaps not as glamorous as the vehicles themselves, a reliable connected infrastructure is no less vital to the future of this transportation transformation — which is estimated to one day enable 90 percent of all traffic fatalities to be avoided.
LEVEL 4, UP AHEAD
Just as human drivers require visual and auditory cues to ensure a safe journey, driverless cars will need vast amounts of vehicle-to-infrastructure (V2I) communications data to understand — and instantly act — if there’s a detour ahead, or a speeding ambulance emerging from behind.
The Society of Automotive Engineers has defined six Levels of Automation — ranging from driver-only control (Level 0) to full automation (Level 5). The conventional wisdom in the engineering community is that for vehicles to operate independently, the technology must advance to at least Level 4. At that stage, the vehicle can perform all safety-critical driving functions and monitor roadway conditions in a limited, defined-use scenario (like on a driverless-car-only roadway) for an entire trip.
We may never reach Level 5 — full automaton in every driving situation — but achieving Level 4 will enable us to truly deliver on the promise of self-driving vehicles. And that’s why researchers are working to develop the requisite roadway technologies that will transform the external environment into a truly connected infrastructure.
In recent years we’ve seen great strides in traffic and safety innovations, from increasing the brightness of road signs to ensuring pavement markings are visible even in wet conditions. Now looking at autonomous vehicles, continued advancement and updates in the infrastructure space are crucial to preparing roads for autonomous vehicles, which is why companies are working with municipalities and auto manufacturers to develop smart roadway technologies that connect with a car’s sensors. These updates will help achieve Level 4 (and beyond) automation.
The ultimate goal of these innovations is to build enough redundancy into the ecosystem so that autonomous vehicles have the intelligence — and the confidence — to make the right decision, instantly. Just as a driver makes a judgment in nanoseconds based on a variety of factors, even if weather conditions are not optimal or unexpected activities occur, so too must the “brains” of a driverless vehicle.
So when it comes to information to power this decision-making, the more the merrier.
COLLABORATION IS KEY
The road ahead for autonomous vehicles and the connected infrastructure to support it remains long. After all, the human brain is rather tough to replicate. But the destination, while distant, is in sight.
Public-private partnerships, such as the work by the University of Michigan Mobility and Transportation Center (MTC) and 3M, will allow for testing and research that can develop the full ecosystem required to drive autonomous vehicles. The challenges are many, but every day brings new discoveries and options that will, in the words of MTC Director Carrie Morton, “truly open the door to 21st-century mobility.”
Perhaps the biggest challenge that remains is the fact that so many disparate manufacturers, academicians and other organizations with a stake in the result are working independently on various aspects of the solution. Each is making progress on various technological systems, but for a true ecosystem to emerge, we must redouble our efforts to take an interdisciplinary approach to research and development.
While autonomy is the ultimate goal, we can’t be autonomous in the pursuit of it. The stakes are too high — in health-care cost reduction, economic enhancements and human lives saved — to pursue the goal alone.