There are two headlines used by our sector around the world. They are…. Electric vehicles and autonomous vehicles. Guess what? Neither will be as significant nor affect the collision repair business as much as advanced driver assistance systems (ADAS).
The increase in high voltage systems on cars (up from 12V to 48V and through to around 800V) is underway, but to think only of pure electric vehicles is a mistake except for those who choose to specialise in such a relatively small market. Equally, many large corporations have autonomous vehicles in fleet trials around the world, with not one fully autonomous vehicle for sale.
The steps towards full autonomy are underway with Level 2 systems requiring driver input every 15 seconds reached the global market during 2016, and the first Level 3 system which allows the vehicle to check the immediate road conditions – and if permitted – can drive itself at speeds up to 56 km/h is going on sale at the end of 2017. That means acceleration, lane change and braking.
Interesting, but not the big game
Think of a nest of newly hatched birds. One starts to cry for food, and within one bleat all of the birds are crying. That’s the situation around ADAS – a huge sea of mis-information, spin and facts all mixed together. The automotive industry repeatedly makes the mistake of treating every single item fitted to the vehicle as a limitless profit opportunity – putting high quality sound systems in exactly the same business strategy as primary safety.
ADAS is not a luxury. Neither is it a legal requirement for cars, yet. The U.S. will get there first by 2022. In Europe autobrake with lane departure warning has been a legal requirement on new trucks and coaches over 7 500 kg gross vehicle weight since September 2015, and will come to panel vans over 3 500 kg gross vehicle weight by September 2018.
Meanwhile, ADAS is seen as a great marketing device and is spreading quickly as build costs tumble and external bodies such as Global NCAP/Euro NCAP reward fitment of such systems.
What is ADAS?
Welcome to the first stage of the wild, wild West. ADAS covers any electronic system that can activate the vehicle braking or steering system. In the broader sense it can include warning systems such as parking sensors. Vehicle manufacturers have been quick to exploit this by fitting anything from one sensor through to more than a dozen, all under the single banner of ‘ADAS’. All items seek to defend the vehicle we are using from either hitting other vehicles, or mitigating the effects of being hit. The defence system can include any or all of the following elements:
Optical: A CMOS camera enabled with image processing capability, which can classify images from cars (560 variations), trucks (168 variations), pedestrians (503 variations), traffic signs (353 variations) and other objects (120 variations). The camera can track images in the transverse plane to the direction of travel, at a range of up to 500m.
If we add another camera unit close to the first one, the dual image processing allows triangulation to occur and hence calculation of how far ahead the potential obstacle is. The first stereo camera system was co-developed by Hitachi with Subaru in 2010, and has since been taken up by Bosch and ZF (which now includes TRW).
If we have three cameras each with their own focal range then the accuracy of the module over the 500m range increases dramatically. ZF are due to start fitment to production vehicles from early 2018 onwards.
Does this need to be aimed? Yes. The camera module has to know where the vehicle centre line is, and a view of the horizon. Typical locations are on the windscreen, usually in front of the rear view mirror. Rear view cameras sometimes may also need to be calibrated if they are linked to auto-braking. These typically are located on the exterior trim of the boot lid/tailgate.
Small cameras fitted to provide a 360 degree surround view don’t normally need to be calibrated.
RADAR: Electromagnetic wave chopped emission and reception module, using the Doppler effect to determine relative velocity/range. Range varies from 250m with a relatively narrow spread of sensing through to 60m – or less – with a much wider spread of sensing. The sensor modules can detect obstacles right down to a few metres, but the accuracy is not great. Thus, they are effectively blind to close range obstacles (sub 10m).
The revolution was to develop multi-range solid state RADAR modules – however, because electromagnetic waves do not ‘see’ plastic they have typically ended up at the very front or very rear of a vehicle, where minor low speed knocks can damage the mounting brackets. Volvo Car and Delphi changed that thinking by making a combined RADAR and CMOS camera module which sits behind the windscreen, fitted to XC90 II, S90, V90, XC60 II and now… XC40. That means the windscreen has been calibrated as part of the RADAR module system, which was a world first.
Does the RADAR module need to be aimed? Yes. The modules need to know the vehicle centre line position and a view of the horizon. The front module(s) can be located on the bumper beam or inside the front bumper skin. The rear modules can be fixed to the rear quarter panel or clipped to the rear bumper skin.
LiDAR: Infrared chopped beam light emission and reception module, using the Doppler effect to determine relative velocity/range for sub-10m range measurement. The first applications were used to ‘fill in’ the fuzzy near vision of typical RADAR systems, and used an expensive rotating drum internally to chop the transmission signal. However, in the past year solid state versions of the module have gone into production along with greatly enhanced capability – indeed the next generation high capability modules will appear during 2018, on each corner of a vehicle.
LiDAR: is used to map terrain and is typically found in military as well as experimental autonomous vehicle applications. The build cost of these ‘mapping’ sensors has fallen from more than $100 000 per unit in 2010 to around $5 000 in 2017. For road vehicles the terrain scanning capability is largely irrelevant due to the availability of 3D map data via WiFi.
Does the LiDAR module need to be aimed? Yes. It has to know where the vehicle centre line is. Typical locations are on the windscreen, usually in front of the rear view mirror.
Ultrasonic: Yes, parking sensors. Special ones with a narrower sensor width and extended range to all of 4.5m. These are used to enhance the near ‘vision’ of ADAS, and are usually present for automated parking. They don’t usually need calibration.
A vehicle can have any one of the above sensors, or multiples of each type. Further, it is not necessarily fitted as standard, but is frequently buried in an optional ‘pack’. This means the end user/owner may not know what type of systems are fitted to the vehicle. Neither will the insurer. Nor the repairer.
Let chaos commence…..
The aiming ‘systems’ fall into three categories:
Pre-2000, or ‘tool room’: Right now we can buy brand new vehicles with such calibration requirements. Essentially the vehicle is set up in a generous level floor space, ranging from two repair bays through to an area 6m long/5m wide at the front and 8m long/15m wide at the back of the vehicle.
The process involves completing four wheel alignment, and then repeating the process to establish the vehicle centre line relative to the sensor which needs aiming. The process involves specialised tooling which is available to buy or hire from the OEM, or from specialist suppliers.
The aiming process is achieved via some very basic mechanical adjustment, but the main process is completed via the diagnostics module.
After 2008, or ‘dynamic’: There is usually a basic mechanical adjustment to ensure the sensor is set level. After that the entire process is electronic, via the diagnostics module. The software updates not only the module calibration but also other modules on the vehicle network, and involves a road test lasting between 10 to 30 minutes with speeds between 64 and 100 km/h. The process means the vehicle must ‘see’ a range of hazards, road markings and more.
The diagnostic system requirement is for one technician to drive and another to monitor the calibration progress.
2017 on: The holy grail. The very first self-aiming modules came onto the market in 2106, with rear end RADAR on some facelifted Volkswagen MQB platform models. As ADAS fitment becomes standard and the number of sensors increases, so the time/cost of associated warranty will force more systems to self-calibrate.
Further, models can migrate from one calibration approach to another during a single model life span.
The key to success is to get OEM repair methods and access to OEM, or OEM quality diagnostic systems. Further, we can start to check what type of sensors could be fitted to the vehicle, via physical inspection and where possible – the OEM build specification. The latter will include the building blocks that were used for the specific vehicle, via VIN.
This does mean we approach each vehicle as a unique proposal – there are no generic solutions, and expertise in use of diagnostic systems is a vital requirement.
However, the ability to keep work inside the repair shop rather than automatically handing it over to an OEM franchise workshop has to be attractive. The flip side is the realisation that even within a single brand there are multiple ADAS applications which share little common ground, and that right now the fitment level is low.
Why bother with ADAS? Because the fitment level is increasing exponentially, and the next five years will become as common as airbags. Time to consider business strategy, with the associated planning for investment both in equipment as well as training. This does not mean rushing into purchasing decisions right now, but careful research to scope the likely take-up of models equipped with ADAS in your area, or where appropriate for more expensive vehicles, nationally.
This requires thought, but the future is clear: ADAS introduces us to the new revenue stream for collision repair: panel, paint, MET… and electronics/software. The latter is going to become more significant as collision repairers develop the capability to be a one-stop for all software/electronic issues, which will become far more common in the next decade.
Perfect market positioning. Time to build that capability.
Auto Industry Consulting is an independent provider of technical information to the global collision repair industry. Products include EziMethods, our online collision repair methods system and Auto Industry Insider, our collision repair industry technical information website. For more information please visit the websites: www.ezimethods.com and www.autoindustryinsider.com or contact firstname.lastname@example.org