I have spent the last day or so trolling through research into Lane Keep Assist (LKA). (Unfortunately, I have lost access to my university’s library database, and so I am stuck with open text articles via Google Scholar.)
I must admit that the more I read the literature the more I am left with a sinking feeling that ANCAP was too far ahead of the technology in mandating that ADAS is forced default-on at every vehicle start. We are going to see a few million cars sold in Australia the next couple years. With a sizable percentage fitted with ADAS systems that are going to seriously annoy its drivers.
Lane Keep Assist (LKA), Lane Departure Warning (LDW), Blind Spot Detection (BSD), Adaptive Cruise Control (ACC), Autonomous Emergency Braking (AEB), Electronic Stability Control (ESC) and the rest of the alphabet soup of acronyms fall under the umbrella term of Advanced Driver Assist Systems (ADAS).
Of the literature that I have read through, most research has concentrated on Adaptive Cruise Control (ACC) / Autonomous Emergency Braking (AEB) as one subgroup and Lane Keep Assist (LKA) / Lane Departure Warning (LDW) as the other subgroup of ADAS.
All the literature that I read acknowledged that Autonomous Emergency Braking and Lane Keep Assist was prone to erroneous misfires and on occasion failed to operate when presented with the correct triggers. Further, the literature also acknowledged these issues were distracting to the driver and resulted in a general loss of confidence in the vehicle. That said, international journals stated that of the driver assist technologies Adaptive Cruise Control (ACC) was the most accepted assist system by drivers and contributed most to reduction in accidents.
Australian research by Monash, “
The Potential Benefits of Lane Keep Assist Systems in Australian Light Vehicles, REPORT NO. 365”, stated “
… if all light passenger vehicles in vehicles in Australia were fitted with AEB, the estimated reduction in total injuries would be 10% for fatalities and serious injuries and 9% for minor injuries…”
In a separate publication by the same authors, “
The Potential Benefits of Autonomous Emergency Braking Systems in Australia, REPORT NO. 339”, it was stated “
… statistically significant 16% reduction in the risk of involvement in all casualty crashes of these types and a 22% reduction estimated for fatal and serious injury crashes was associated with LKA fitment to Australian light vehicle was estimated…”
However, these conclusions by Monash Research was challenged in this report; “
Quantifying the Lost Safety Benefits of ADAS Technologies Due to Inadequate Supporting Road Infrastructure”, Special Issue Current and Future Issues in Transportation Safety and Sustainability, 2022. If you want to dig out a copy (it is open source) the title is
https://doi.org/10.3390/su14042234
For example, the authors of this study noted: “
For example, AEB is unlikely to be effective on gravel or unsealed roads and lane keep assist technologies are unlikely to function at all on roads that have poor or no delineation. Therefore, the ability of these technologies to perform optimally will most likely be compromised in locations where supportive road infrastructure is absent … Research that quantifies the predicted benefits of AEB and LKA often fail to acknowledge assumptions made about roads on which cars fitted with these technologies will be driven … Studies that have previously estimated the benefits of advanced driver assist technologies have been conducted on the basis that road conditions are always adequate to support the technologies. The assumption that the entire road network is fully supportive of systems such as AEB, LKA and ISA, however, is severely flawed…”
“
… Across the three states, close to 50% of LKA-sensitive crashes have historically occurred on arterial, and sub-arterial (or equivalent roads), where delineation was found to be less than adequate, potentially compromising the ability of LKA function, once all vehicles are fitted with the technology. Based on the LKA-sensitive crashes in the historical crash data, up to 138 fatalities and 142 serious injuries may not be avoided annually across the three Australian states based on lack of delineation availability …”
A key word in all of the reports mentioned above is that the crash “
… is sensitive …” to AEB, LKA, or both. It does not say that AEB or LKA will prevent the accident. It just says that AEB and/or LKA may assist in reducing the accident severity, with an unknown reduction if fatal or serious injuries.
So now we have this interesting circular argument. A poor road surface and/or lack of marked lines is considered to contribute to an accident. So, the bureaucrats then suggest mandatory implementation of AEB and LKA to prevent accidents that nether system can mitigate (due to poor road surface).
Instead of dollars being spent on LKA, it makes be wonder why the government doesn't paint more white lines on the road. At the very least clean up the white lines they have painted. Seems like a simple way to reduce the road toll.
Still, there is one bottom line estimate for LKA Australian conditions. It was estimated that every vehicle had LKA then up to 138 fatalities and 142 serious injuries (50%)
may be avoided in Vic, SA, and Qld. And it was estimated 50% 138 fatalities and 142 serious injuries
will not be avoided because there are no lines on the road for LKA to detect.