This SAE Recommended Practice provides test procedures, performance requirements, and design guidelines for adaptive driving beam (ADB) and associated equipment.
Recent advances in vehicle technologies have enabled active control of road illumination to the point that portions of a beam can be dimmed or removed based on inputs from the vehicle and/or its surroundings. This, coupled with advances in technologies used for lane departure warning, automatic high beam activation, and other functions has enabled the identification and location of other vehicle road users at night and to actively limit potentially glaring light to those vehicle road users. This technology, generally referred to as Adaptive Driving Beam (ADB), can provide a driver with the benefits of forward illumination similar to that from a high beam without the glare that beam would cause for other motorists. This is because portions of the beam are adapted such that what would be high beam levels of glare towards another vehicle road user is dimmed to at or below levels that would be produced by low beam. While this document does not require ADB glare levels to be dimmed below low beam levels, it does allow that possibility. Although current low beams cause less glare than high beams, the residual glare that they produce can still impair the vision of oncoming drivers. Because of this, ADB technology offers the possibility of better glare protection than current low beams. By means of both increased seeing light and reduced glare light, ADB has the potential to greatly enhance the vision of drivers using ADB equipped vehicles as well as the vision of drivers of vehicles with conventional headlighting systems.
United Nations (UN) Regulations were updated to accommodate ADB under UN Regulation 48 because of the potential benefits of ADB. Numerous vehicles equipped with ADB are now available for sale in regions governed by UN Regulation 48. However, in the United States it is unclear how ADB would be treated under the current Federal Motor Vehicle Safety Standard (FMVSS) 108. Understanding the potential safety benefits, the National Highway Transport Safety Administration (NHTSA) has indicated a willingness to update FMVSS 108 to include ADB. NHTSA has indicated that an SAE Recommended Practice representing an industry consensus on the requirements for ADB is desired and that such a Recommended Practice would be considered when FMVSS 108 is updated.
The industry desired an expeditious update to FMVSS to allow ADB since these systems are already offered by vehicle manufacturers in other countries and because they appear to offer substantial safety and visibility benefits to consumers. To this end, the SAE Regulatory Cooperation Task Force first decided an update to SAE J565 Semi-Automatic Beam Switching could facilitate, adding the necessary provisions to allow ADB. Semi-automatic beam switching systems are currently allowed and referred to in FMVSS 108. These systems operate under the same principles as ADB. Since 2004, more than 6 million vehicles have been sold with camera based semi-automatic beam switching systems with few complaints or issues. The revision of SAE J565 was initially thought to be the most expeditious way to allow ADB within the current FMVSS 108 regulation; however, this direction was not pursued because there were no vehicle based performance requirements contained in SAE J565. The task force was advised objective vehicle based performance requirements would be needed for NHTSA to consider allowing ADB.
Therefore, it was decided to create a new SAE Recommended Practice, SAE J3069 Adaptive Driving Beam. The requirements and rationale contained in SAE J565 were considered in the development of SAE J3069. To expedite the completion of SAE J3069, the SAE Task Force decided to base its ADB requirements on existing requirements for low beam headlamps rather than creating new beam patterns and requirements which would not only delay adding the improvements of ADB to the national fleet, but were deemed non critical in accomplishing our objective. Any vehicle based requirements were drawn from NHTSA’s report NHTSA- 2011-0145 “Feasibility of New Approaches for the Regulation of Motor Vehicle Lighting Performance”. To the extent possible, the task force also attempted to harmonize J3069 with current UN regulations.
The test protocol and requirements were derived from information contained in NHTSA-2011-0145, UN Regulation No. 48 ADB requirements, FMVSS 108 requirements, University of Michigan Transportation Research Institute (UMTRI) reports and testing and Rensselaer Polytechnic Institute (RPI) testing. The requirements of SAE J3069 are intended to ensure that glare values from ADB lamps are at or below the glare values from the low beam lamps mounted on the same vehicle. Since glare levels from low beams are already defined, agreed upon, and a part of SAE J1383 and FMVSS 108, the goal of this document was not to exceed these values with an ADB equipped vehicle. However, as stated earlier, this document also allows lower glare values than allowed for low beams.
The original intent of the Task Force was to create a recommended practice that would only involve test track requirements applied to an equipped ADB vehicle. However because the glare requirements are based on the low beam, laboratory testing within the “dimmed” area of the ADB pattern was deemed essential to confirm compliance of each lamp to the low beam pattern glare values and maintain low beam minimums outside of the “dimmed” area. Including a lamp laboratory test also had the benefit of providing lamp manufacturers with photometric requirements to use in their process control, similar to the current requirements for low and high beam lamps. A laboratory test also provides a more controlled environment in which to evaluate an ADB than a road test. Within the “dimmed” area, only maximum intensity requirements were included in order to allow lower intensities near the horizon. This enables lower glare intensities than the low beam. If low beam minimums were required inside the “dimmed” area, the glare intensities with ADB would not be lower than the low beam. In the future, if whole vehicle testing, defined in NHTSA–2011–0145, for a low and high beam are incorporated into SAE and/or FMVSS, the need for laboratory photometric requirements may be reconsidered for removal.
During the development of SAE J3069, a key premise was that an ADB is a supplemental road illumination system, similar to front fog lamps, which can be easily deactivated by the driver. This is important because there are inevitably certain conditions and situations under which ADB may not perform to driver’s expectations. In these situations, the driver would need to be able to deactivate the system and either revert to low beam or activate their high beam. In that respect, ADB’s functionality is similar to current semi-automatic beam switching systems.
Having an indicator to alert the driver that ADB is enabled and that the system is emitting greater than low beam light levels was discussed at length. It was agreed that a lighted telltale indicating that ADB is enabled would be included. However, determining if there should be an indicator notifying the driver that the ADB is active, i.e. emitting light above low beam levels, was less straightforward. UN regulation requires that there be a telltale indicating that ADB is enabled and that the high beam telltale be activated when the system is active. Because UN regulation defines ADB as a mode of high beam, this is straightforward even though it appears that there have been different interpretations of when the high beam telltale is activated in practice. It was noted that both FMVSS 101 and FMVSS 108 require the high beam indicator to be activated in the case when the high beam is illuminated. The purpose of this telltale is to warn the driver that his/her vehicle may be glaring other drivers, unless high beams are switched off when such other drivers are present. However, ADB obviates the need to warn drivers that they may be glaring other drivers, and thus the telltale is arguably not needed. Also, it did not seem appropriate to follow UN Regulation 48 in requiring that the high beam telltale be used to indicate ADB activation, since the ADB is herein defined as an addition to or equivalent to the low beam. The group also did not agree that an additional indicator should be required to notify when ADB was active, i.e. emitting light above low beam levels, as this would place a substantial burden on the industry, create a new requirement above what is required under UN Regulations, and cause confusion for the driver all without providing any apparent benefit. It was therefore agreed that only a ‘system enabled’ telltale, and not an additional ‘ADB activated’ telltale would be required.
The road test requirements were also discussed at length. UN Regulation 48 requires vehicles with an ADB to be driven on public roadways following a prescribed set of conditions to verify that ADB performs appropriately. For SAE J3069 to be accepted into the regulatory framework of the Motor Vehicle Safety Standards, repeatable road test conditions were developed and defined, repeatable measurement criteria were defined, pass / fail values were established, and a reasonable set of tests were developed which adequately discriminate an acceptable ADB.
The pass/fail requirements were defined based on the whole vehicle testing methodology defined in NHTSA–2011–0145. These provided definite absolute maximum light levels representing illuminance levels on the opposing vehicle driver’s eyes and preceding vehicle side view and rear view mirrors. These values are represented in Table 1. In addition, an allowance was included that if the glare values exceeded these lux levels, they would be acceptable if the levels didn’t exceed 125% of the low beam levels measured from the ADB equipped vehicle under the same conditions. The rationale was to require a light level restriction that is no greater than the low beam on the vehicle being tested, while allowing slightly higher light levels than that vehicle’s low beam to accommodate test variation and small amounts of increased light due to beam pattern variations. It should be noted that the ADB light level comparison to the low beam is for the particular low beam for that vehicle, not to the regulatory maximum allowed light levels for the low beam. It is expected that low beam light levels for any particular vehicle will be below the maximum allowed, such that the 125% allowance does not represent 125% of the maximum allowed glare.
Test variation was seen during vehicle testing conducted by University of Michigan Transportation Research Institute (UMTRI) due to a slight dip in their test track. This slight dip created a repeatable peak in the measured opposing eye illumination levels when the ADB vehicle passed over the dip. These values would not have complied with the values in the table, however they would have complied with the 125% allowance. On the other hand, if the ADB vehicle being tested had met all the requirements in Table 1, then the track tests with only the low beam activated would not be required which would reduce the testing burden.
The track test would be conducted using a stationary fixture representing the preceding or opposing vehicle and the ADB test vehicle would drive toward this fixture. It was decided that the fixture would be stationary to reduce test variability. This was considered worst case since some camera systems utilize opposing or preceding vehicles’ movement within a scene to identify them as vehicles instead of other road objects, such as reflectors on the side of the road.
The light towards the opposing and preceding vehicle driver/rider would be measured using lux meters positioned at the driver’s eye point for an opposing vehicle and at the rear view and outside rear view (OSRV) mirror points for a preceding vehicle. The task force considered the driver/rider’s eyes the most critical area for glare prevention. However, the expectation is that ADB will reduce any glare producing light toward all locations on opposing and preceding vehicles, thereby also providing benefit to any passengers in the vehicle. The test fixture containing these meters would include representations of headlights or taillights indicative of a small vehicle and a motorcycle. The locations of the lux meters and lamp representations were taken from median location values provided by UMTRI. However, truck OSRV mirror horizontal locations shown in Figure 3 were simplified to be in line with the car OSRV mirrors. This was deemed acceptable because the portions of an ADB that are dimmed or removed are based on the relative location of the headlamps to the driver’s eyes and taillamps to the mirror locations. The horizontal width of the ADB dimmed area towards a preceding vehicle is proportional to the width of the taillamp’s illumination. If a truck’s taillamps are spaced at a greater distance, the width of the dimmed area of the ADB would become wider. If truck OSRV mirrors were spaced further apart in Figure 3, a separate fixture would have been required to accommodate the wider truck taillamp locations. The task force deemed this unnecessarily burdensome.
In discussing the various road test conditions that should be included, it became clear that there are hundreds of potential test conditions, and specifically covering them all would make the testing requirements excessively burdensome. It would also result in excessive test-to-test variation due to differing curve radii, hills, bumps, dips, etc. Through the discussions, it was determined that the ADB would function similarly over many of these test conditions, and therefore to include them all would provide no added value. A subgroup of the Task Force determined that testing on a straight level road over two lanes to the right and left of the ADB vehicle would cover the important conditions to adequately determine if an ADB functions in an appropriate and safe manner. Combined with a test of speed for response by the ADB, testing over these locations would ensure an adequate response to a wide variety of road geometries while allowing the test method to be simple enough to be objective and repeatable. For example, sensitivity to other vehicles encountered on curves is covered by the test requirement that, in a straight-line encounter, an ADB must continuously track the angular location of an opposing vehicle as that angular position becomes progressively further from the center of the camera’s field of view with decreasing distance to the opposing vehicle.
The lamps represented on the fixtures were also discussed. The task force agreed that the lamps on the preceding and oncoming fixtures should represent reasonable worst-case for intensity and location and should promote test repeatability.
There was concern that if the actual low beam headlamps were used on the opposing vehicle test fixture the large gradients present in typical low beam patterns would cause unnecessary test variability. Therefore, it was decided to use headlamp representations for the opposing vehicle test fixture emitting a diffused uniform light of 300 cd intensity. The 300 cd from each lamp is a representative of the lower range of intensities that ADB would encounter from the UMTRI 25th percentile beam pattern and so represents a reasonable worst-case stimulus for camera recognition systems. Representing only headlamps on the test fixture, rather than including parking lamps or other parts of an actual vehicle, can also be considered a reasonable worst-case condition. The lighted area of each lamp was specified to be representative of available headlamps. For the preceding vehicle fixture, it was agreed to incorporate only one pair taillamps (rear position lamps), with no other rear lamps or retroreflectors, again as a reasonable worst-case condition. The rear position lamps would each be required to output an intensity of 7 cd from a surface emitter of nearly uniform luminance. Originally, the task force specified an output of 13.5 cd, which was derived from SAE J565 when converted from the illuminance level specified in that document. However, this was considered too high when compared to the minimum requirements of 2 cd for taillamps in SAE J585 and FMVSS 108. However, 2 cd was considered too low when compared to typical lamps on the road. An informal survey was conducted by lighting suppliers of their low light output taillamps and 7 cd was a typical minimum value. In addition, simply due to the minor filament intensity, a taillamp using a bulb and a reflector will typically output 10 cd, well above the specified 7 cd. The lighted area of each lamp was specified to be representative of commonly available taillamps. The lighted area of each lamp was standardized to 50 cm2, representing a minimum optically combined stop / tail function.
The SAE Task Force understood the importance for an ADB to react within an appropriate time when confronted with the sudden appearance of an opposing or preceding vehicle due to cresting a hill, a vehicle entering a roadway, etc. To test that response time, the track procedures require that the ADB respond to the sudden appearance of the preceding or opposing test fixture lamps during the test drive. It is known that ADB cannot react instantaneously when confronted by the sudden appearance of a vehicle, so the opposing or preceding driver may experience high beam glare levels for a short period of time. The allowed period of time was debated. The current SAE J565 requires a reaction time of less than 4 seconds. Based on published literature, the un-alerted driver reaction time was considered to be approximately 2.5 seconds to dim from high beam to low beam. Based on this, the task force selected 2.5 seconds as the maximum reaction time of ADB for the sudden appearance of an opposing or preceding vehicle. The 2.5 second duration is also supported by the research summarized in SAE 890732 Perception/Reaction Time Values for Accident Reconstruction.
Horizontal Aim – It was realized that since ADB contains vertically oriented cutoff lines, with horizontal gradients or transitions from the full ADB pattern to the dimmed area, some type of horizontal adjustment would be required. This is not an issue if a separate ADB lamp is utilized. However, it is anticipated that most systems would use portions of the low beam or high beam or both to create the ADB. FMVSS 108 does not allow a horizontal adjustment mechanism for Visual/Optical Aim (VOA) headlamps unless equipped with a vehicle headlamp aiming device (VHAD). A VHAD system was not considered accurate enough to properly place the transition from dimmed area to full ADB. For this reason, horizontal adjustment requirements were incorporated into the document by automatic or manual means. However, this would require a change to FMVSS 108 to allow a horizontal adjustment mechanism on headlamps with ADB function.
A requirement was included specifying that ADB cannot be activated until the system has determined that the sensor is not blocked. This is to prevent blockage of the sensor from being falsely interpreted by the system as lack of other vehicle road users, resulting in ADB maintaining high beam when it should not.
A requirement was included which specifies that if a fault in ADB is detected, the system shall automatically activate headlamps until the fault is no longer present.
A color requirement was included equivalent to the low and high beam color requirements.
Environmental tests from SAE J575 Test Methods and Equipment for Lighting Devices for Use on Vehicles Less than 2032 mm in Overall Width and SAE J1383 Performance Requirements for Motor Vehicle Headlamps and SAE J2357 Application Guidelines for Electronically Driven and/or Controlled Exterior Automotive Lighting Equipment were included in this document.