The U.S. National Academy of Sciences’ Transportation Research Board held its 99th annual meeting last week in Washington, D.C. Over 14,000 researchers, academics, scientists, practitioners, regulators, standards developers, policymakers, consultants, and others converged on the giant 213,677m2 Walter E. Washington Convention Center to present and attend lectures, presentations, and poster sessions on virtually every subject related in some way to the transport of people and goods by any and every means.
Included in the program was a lecture session on evaluating and implementing advanced vehicle headlighting systems. DVN Chief Editor Daniel Stern presided over the session, which gathered four high-level expert presenters who described, each through their own lenses, the safety benefit potential of advanced vehicle lighting and how these systems can be evaluated to ensure they realise those benefits for road users.
• Ian Reagan, Senior Research Scientist with IIHS, the Insurance Institute for Highway Safety.
He has conducted research on human factors topics related to traffic safety, including crash avoidance technology, and driver distraction caused by technology both built and hand-carried into cars.
Dr. Reagan spoke on the challenges and impacts of the IIHS headlight safety performance evaluation programme. He described the motivation for IIHS’ headlighting tests: over 19,000 traffic-related deaths in other-than-daylight conditions in the U.S. just in 2018; research and data suggest better lighting would help, FMVSS № 108 fails to practically guarantee a minimum level of on-road performance, and headlight performance varies widely among different vehicles and headlamp systems. Then he discussed how the IIHS headlight test protocol was devised, how it’s been adjusted to better assess real-road seeing and glare performance, and how the IIHS Top Safety Pick and TSP+ award criteria have been strengthened to place progressively more pressure on automakers to provide effective headlamps.
The most recent change: for 2020, a car can’t get the Top Safety Pick+ award unless it has “good” or “acceptable” headlamps as standard equipment (optional at extra cost won’t do).
And he described the apparent trend: in 2016, the first year of testing, only three cars got a “good” grade and just 36 got “acceptable” for their headlamps; 43 got “marginal” and 142 got “poor”. Those figures have gradually improved; in 2019 68 cars got “good” and 103 got “acceptable”—though 114 got “marginal” and 183 got “poor”. Trending in the right direction, but still with a ways to go.
• John Bullough, Director of
Transportation and Safety Lighting Programs at the Lighting Research Center of
Rensselaer Polytechnic Institute in Troy, New York.
He conducts research on vehicle and roadway lighting, traffic signs and
signals, mesopic vision, glare, and other human factors topics. He gave a good
review of recent research on the safety benefits of adaptive headlighting,
described the increasingly-sturdy consensus among researchers, regulators, and
industry that adaptive headlighting systems can provide real safety benefits,
discussed methods of linking headlight performance to crash reductions, and
emphasised that human factors data, visual performance modelling, and
statistical crash analysis all converge toward aligned conclusions, which
provides ways of realistically predicting safety benefits from advanced
headlights, rather than simply looking to see if benefits were realised after a
system was deployed. This is especially significant because the safety effects
of crash-avoidance technology like lighting and signalling is notoriously
difficult to satisfactorily model in this manner, which is one reason why it is
not a favourite topic of American regulators who must work within a legislative
framwork that requires concrete cost/benefit calculations.
• Carol Flannagan, 29-year veteran
research associate professor at the University of Michigan Transportation
Research Institute.
She who makes a speciality of applying statistical methods to assessing
crash-avoidance technologies. She’s in charge of UMTRI’s Centre for the
Management of Information for Safe and Sustainable Transport, and has over two
decades’ experience with research and data analysis on traffic-related injury
risk. She developed an injury outcome model that allows for direct comparison
of public health-, vehicle-, roadway-, and post-crash-based safety measures.
She described a major UMTRI study, sponsored by
General Motors, looking at the safety benefits of systems including HID
headlamps, swivelling HID headlamps, and automatic high/low beam selection
systems—each independently—versus ordinary halogen headlamps with manual
high/low beam switching. Results: HID headlamps and automatic high/low beam
switching are associated with large, statistically significant safety benefits,
and if a vehicle has both systems their benefits are both present, though they
combine in a complicated manner; they don’t just sum or multiply. The
swivelling HID headlamps showed a benefit that was smaller, less certain, and
not statistically significant, probably because the sample size was too small;
there weren’t many equipped cars represented in the crash data obtained from a
variety of U.S. states. Other interesting results: AEB (Automatic Emergency
Braking) showed a statistically significant 46% reduction in system-relevant
crashes, but FPB (Front Pedestrian Braking) showed a
statistically-insignificant 13%. And all the systems relevant to reversing
(rear vision cameras, parking assistant systems, rear cross-traffic alert, and
reverse automatic braking) showed large to enormous crash reductions.
• Michael Larsen, General
Motors’ longstanding Global Exterior Lighting Technical Lead.
He’s very active in the SAE Lighting Systems Group, wherein he chairs the
Adaptive Driving Beam task force.
Larsen presented a thorough overview and explanation of ADB—what it is, what it
does, how it works—and a detailed description of the meticulously dogged
efforts, including 110 hours’ worth of 34 meetings he led to translate the
subjective provisions of UN Regulations 48 and 123 (such as “any obvious
malfunctions shall be contested”) into the objective terms required of
vehicle safety standards under U.S. law, while maintaining minimal divergence
from the effective technical requirements of the UN regulations so as to assure
the greatest possible harmonisation.
Larsen outlined the philosophical position taken by SAE: ADB is optional, so
any improvement it offers in a driver’s ability to see at night is a benefit,
and his task force therefore prioritised glare control rather than on seeing
light performance. And while a dynamic system like ADB means thousands of
different driving scenarios could be evaluated, the SAE task force strove to
boil them down to the minimum count crucial to demonstrating appropriate system
performance, and standardising as many aspects as possible of the testing (and
of safety non-critical aspects), keeping in mind that it’s not necessarily
needed or beneficial to write in controls for every possible aspect of a device
or system. Importantly, Larsen and the task force looked at requirements
present in regulations simply by dint of inertia imparted by past practices,
and pared those from their proposed standard if they weren’t helpful. Larsen
noted Transport Canada’s prompt adoption of SAE J3069, and detailed NHTSA’s
regrettable rejection of it and that agency’s ongoing opaque deliberations on
the matter.
Finally, Larsen said this work demonstrated the feasibility of devising
objective requirements from a feedstock of subjective UNECE regulatory
provisions, but it can be difficult and won’t necessarily result in full
harmonisation.
After each lecture there was ample time for Q&A, which facilitated a good exchange of ideas and helped ensure that all sessiongoers had the chance to fill any gaps in their understanding of the technology and its topography in the U.S. context. Attendees were advised of the periodic DVN Workshops where more research and innovation on topics of this nature can be seen and discussed.
The National Academy of Sciences, founded in 1863 by an act of the U.S. Congress is charged to provide scientific advice to the government whenever called upo’ by any government department. The Academy receives no compensation from the government for its services.
