Intervention Needed Every 9 Minutes in Test of Advanced Driver Assistance Systems
Automobiles today come equipped with a range of advanced driver assistance systems (ADAS). These systems, often seen as a second set of eyes and ears, include features such as collision warnings, emergency braking, and blind spot monitoring. They are designed to alert drivers about potential accidents, and sometimes even step in to prevent them.
Some ADAS even offer comfort features like adaptive cruise control and lane keeping which help alleviate some of the driver's responsibilities on the road. One of the newer versions of these is the traffic jam assist, a low-speed, stop-start driving aid designed specifically for congested highways. It's crucial to note, however, that these are classified as "level 2" systems, meaning the driver still needs to be actively aware of their surroundings. More advanced "level 3" systems exist, which allow drivers to fully disengage from driving, but that's a topic for another day.
Testing ADAS Systems in Real Traffic
In a recent experiment, five different ADAS systems were put to the test in an area known for consistent heavy freeway traffic. The test was conducted during both morning and afternoon peak traffic hours, with the same routes being used for every test. On average, each vehicle was driven for 342 miles over a span of 16.2 hours, with each ADAS being operated according to the specific vehicle's user manual. All vehicles were equipped with cameras and GPS systems to record traffic conditions and the driver's behavior.
Hands-on vs Hands-off Systems
The aim of the test was to measure the performance of the systems, with particular attention paid to those that allow drivers to remove their hands from the wheel, and compare them to those that require constant contact.
The results showed that on average, these systems only afforded the driver less than 10 minutes of relaxation in heavy traffic. It was found that drivers needed to intervene every 9.1 minutes, or every 3.2 miles. A staggering 85% of these interventions were necessitated by "notable events" recorded by the data capture systems.
The most common incident that required driver intervention was when another vehicle cut into the driver's lane. This happened approximately every 24.4 minutes, or 8.6 miles, with a whopping 90% of these incidents requiring the driver's intervention.
Another common issue was inadequate lane centering, happening every 32.2 minutes or 11.3 miles and resulting in intervention 72% of the time. There were also instances where the car failed to resume after stopping, causing the driver to intervene. Additionally, there were several cases of the lane keeping or adaptive cruise control systems shutting down, or the test car not slowing down properly, both of which required the driver to hit the brakes.
The Findings
The less advanced, hands-on systems experienced notable events three times more often than the hands-off systems. The hands-off systems required intervention every 20.1 minutes or 7.2 miles, whereas the hands-on systems required intervention every 6.7 minutes or 2.3 miles. Interestingly, drivers were instructed to put their hands back on the wheel every 15.3 minutes or 5.5 miles when using the hands-off systems.
Recommendations
In light of these findings, some recommendations seem to be simply common sense. Regardless of any ADAS, drivers should always remain alert and engaged. These systems are not a replacement for an attentive driver. Avoid distractions, especially from smartphones, and always familiarize yourself with the car's user manual to understand how and when to best use its systems. Also, always maintain a safe following distance, even if this results in more vehicles cutting into your lane.
Based on the test results, improvements in ADAS performance, particularly in response to cut-ins and lane-centering, are greatly needed. These improvements will contribute significantly to the safety and convenience of drivers.