We’ve heard more and more about the self-driving car or, as it is also often called, autonomous car or autonomous vehicle (AV). Several carmakers claim they could have an AV in production by 2020, if not sooner. It remains to be seen, though, if we’ll be able to buy a car that drives itself in the next 3-4 years, but there is no question that technology is moving in that direction by leaps and bounds.
In the Beginning…
As recently as 20 years ago, an AV was almost as much science fiction as a Star Trek transporter or colonizing Mars. Visionaries at the time saw the actual science as hardware-based; somehow vehicles would take direction from some form of strip embedded in the roadway.
However, as technology followed a natural progression leading to such features as adaptive cruise control, rearview cameras and parking sensors, self-driving theory evolved from being an issue of hardware to one of software. Rather than a self-driving car sourcing control information from a strip in the roadway, radar and cameras could provide that information with on-board computers interpreting it.
Enter the Government
As it became more and more apparent that carmakers were moving from “what if?” to “here’s how” in the development of AVs, the National Highway Traffic Safety Administration (NHTSA), the federal safety-regulatory body, named five ascending levels of autonomous driving as a tool for establishing safety guidelines as driverless technology continues to progress. Providing a fairly clear picture of where the industry is at in the self-driving continuum, it also provides an accurate working definition of the self-driving car.
NHTSA’s 5 Levels
Level 0 — No Automation: Just as it sounds, the driver is in complete control and responsible for the primary vehicle functions: braking, steering, throttle and motive power. The driver has total responsibility for safe operation of the vehicle. Even vehicles with passive warning functions, such as lane-departure warning or a blind spot monitoring system, would fall into this category.
Level 1 — Function-Specific Automation: Level 1 vehicles have one or more specific function that is automated, but the functions are in no way connected to one another. Adaptive cruise control that provides automatic braking when engaged, as well as front-collision avoidance that brakes the car if the driver fails to do so, are examples of such independent systems.
Level 2 — Combined-Function Automation: At this level, two or more automated primary-control functions can work at the same time to relieve the driver in limited situations. The driver is still responsible for the vehicle’s safe operation and can reassume control in the blink of an eye. Such combinations could be adaptive cruise control and lane keeping that maintains a vehicle’s position within its lane’s lines. Here, the driver’s hands would be off the steering wheel and feet would be off the pedals at the same time.
Level 3 — Limited Self-Driving Automation: Level 3 vehicles allow the driver to surrender complete operation of the primary functions under certain traffic and environmental conditions. In this level, it’s up to the vehicle to monitor its surroundings and make the decision to transfer operation back to the driver. According to NHTSA, such a transfer would occur when the vehicle realizes it’s coming upon a construction zone and alerts the driver to reassume control.
Level 4 — Full Self-Driving Automation: This is total automation, with the vehicle responsible for the safe operation of all critical functions for an entire trip. Here, the driver’s sole responsibility would be to provide the destination — it’s the vehicle’s job to safely reach that destination.
What it means to you: Current technology available in the showroom has reached Level 2. We may not even see Level 3 vehicles because of the safety concerns of transferring control back and forth between driver and vehicle. That means technology’s next big jump could well be directly to Level 4: the self-driving car.