Advanced driver-assistance systems (ADAS) are physics-based electronic systems that help the vehicle driver while driving or during parking to avoid collisions by offering technologies that alert the driver to potential problems, or by implementing safeguards and taking over the vehicle.
Common ADAS features:
- Forward Collision Warning (FCW)
- Lane Departure Warning (LDW)
- Automatic Emergency Braking (AEB)
- Adaptive Cruise Control (ACC)
- Lane Keeping Assist (LKW)
ADAS vs. AI
ADAS systems and other video-based telematics have tried to help fleets identify types of driver behavior and activity. However, these solutions are limited to detecting harsh maneuvers only, such as acceleration, braking, or cornering events. Once a harsh vehicle movement is detected, the video for that single event is reviewed—typically by a third-party human workforce—for additional labels, such as cell phones. In addition, vehicles with advanced driver-assistance systems can cost twice as much to repair following a collision, due to expensive sensors and calibration requirements.
Hard maneuvers detected by vehicle movement isn’t an indicator for driver risk. The use of AI sensors in the vehicle to detect driver movement, gaze direction/attention, vehicle activity, traffic conditions, and other contextual data to make real-time decisions about imminent risks can provide on average a 40%-60% reduction in collision frequency and collision-related costs. AI sensors and real-time AI intervention in the vehicle can accurately identify risks in driver behavior, including distracted driving and tailgating, and help 4 out of 5 drivers reduce distracted driving without manager involvement. AI safety systems do not require human review or video access at all and are perfectly compatible with ensuring driver privacy.
How Nauto AI improves driver behavior and fleet safety
Improving driver behavior and reducing fleet collisions requires more than safety policies, after-the fact driver training, and physics-based ADAS systems. Nauto’s collision avoidance system outperforms ADAS by predicting and preventing incidents at the source: driver behavior. Other AI solutions solely rely on AI processing in the cloud—Nauto’s edge processing responds to detected risks immediately, enabling In-Vehicle Alerts to help prevent collisions in real-time. Nauto AI-powered solutions continuously synthesize inputs from in and around the vehicle, including driver behavior, vehicle movement, traffic elements, and contextual data, in its multi-tasked Convolutional Neural Networks (CNN) model to determine levels of collision risk. As the detected risk intensifies, they signal the driver to take action with increasing levels of urgent alerts. This allows the driver to correct their behavior proactively, before an incident occurs instead of relying on retroactive coaching weeks after an event.