Advanced driver-assistance systems (ADAS) offer a range of features that are capable of increasing safety while reducing the amount of effort it takes to operate a vehicle. From backup cameras to lane departure warnings and adaptive cruise control, ADAS have already altered the way many people drive. But now that these systems are becoming commonplace, what’s next? Where do we go from here in the quest to decrease road accidents and improve the ease of driving?
With the ability to automate many core aspects of operating a car, advanced driver-assistance systems can be seen as a stepping stone towards fully autonomous vehicles. Although this technology and the surrounding regulatory considerations are still in development, most of the essential components required for self-driving cars are part of today’s ADAS. The expectation is for this technology to evolve into commercially-available vehicles that don’t need a human driver for any tasks.
The Society of Automotive Engineers describes the different levels of automated vehicles as follows:
Level 0: No automation — a human driver performs all functions.
Level 1: Driver assistance — ADAS provides alerts and some control over parking, steering, and throttle.
Level 2: Partial automation — although the driver must still monitor actions, an automated system controls braking, steering, and throttle.
Level 3: Conditional assistance — ADAS performs all operations, but a human driver must be available to take over control in particular circumstances.
Level 4: High automation — automated driving system performs all operations, although a human driver may still control the vehicle if needed, but in a smaller number of scenarios than level 3.
Level 5: Full automation — no human driver required, but an occupant can choose to intervene if desired.
In addition to complex software, several pieces of hardware are expected to be present for self-driving vehicles to operate, including video cameras, IR cameras, radar, ultrasonic sensors, and LIDAR. Although none of these sensors can safely guide a car on their own, collectively they provide a 360-degree view around the vehicle to detect nearby objects, identify them, and avoid collisions, while sending that data to the steering, throttle and braking controls.
Self-driving vehicles may be commercially-available as soon as two years from now, depending on who you ask. But regardless of exactly when the technology is ready to be widely deployed, virtually everyone agrees that fully autonomous vehicles are coming to our roads in the not-too-distant future, to reduce accidents and traffic while increasing the convenience of traveling in a car.
Written by Igor Ilunin, head of IoT at DataArt.