Home / Features / Autonomous cars: more than just carbon emissions

Autonomous cars: more than just carbon emissions


Local Air Quality and Emissions
The emissions cheating scandal in 2015 brought vehicle emissions and local air quality to the attention of the general public and the popular national media in a way that had never happened before. This, in turn, provided a grandstand for the new Mayor of London and the various political parties to talk about the issues surrounding vehicle exhausts, such as test-cycles which are the means by which vehicles standards are maintained under European legislation.

Defra (Dept. for Environment Food and Rural Affairs) subsequently released a draft National Air Quality Plan, which resulted in the Government department being taken to court for a lack of committed actions to tackle air quality centrally. This draft plan was reissued for consultation in May 2017, and continues to focus on the introduction of Clean Air Zones, which would be geographical extents where certain ‘high polluting’ vehicles are not allowed to travel, or may be issued a financial penalty in a similar way as the London congestion charge.

The primary source of local pollution issues in the UK are internal combustion engines (ICE) in vehicles, with the emissions subject to the fuel type, engine speed and loading. Coupled to this is the level of maintenance a vehicle undergoes and how the vehicle is driven. Whilst the exhaust is the primary emission source of nitrous oxides (NOX) there are two sources of particulate matter (PM) one from the exhaust and one arising from tyre and brake wear, and road abrasion. The non-exhaust fraction is becoming the predominant source (by mass) as engines become progressively cleaner. So, as we move into the future and power trains develop into pure electric and or hydrogen the advent of autonomous vehicles will not specifically address the non-exhaust PM fraction. On the contrary, researchers suggest that pure electric vehicles are likely to be heavier than the equivalent ICE counterpart and hence friction based particles are likely to increase as a result. 1

Whilst much attention in the UK has focussed on the health concerns from PM less than 10 microns in diameter there is increasing evidence that the finer fractions <2.5 microns, and ultra-fine particles (<1 microns) carry a much higher health risk factor 2 and are attributed to significant mortality effects. Recent guidance published by Defra ‘A Briefing for Directors of Public Health’ in March 2017 (Defra, 2017) advises that health outcomes from PM2.5 should be considered in the assessment and planning process, and to specifically consider the WHO threshold values of 20 μg/m³, although it should be noted that in 2011 city dwellers in the EU-28 were on average exposed to exposed to 27 μg/m3 of PM10. 3

The Growth of Autonomous Transport
“An autonomous car is a vehicle that can guide itself without human conduction”

It’s inevitable that these vehicles will be common place on UK roads however the pathway to autonomy is not likely to be complete until 2030 4 and beyond. The transition pathway will include a number of key technologies such as cruise control, park assist and adaptive cruise control which are now mainstream additions to new cars. By 2020 we might expect to see traffic jam assist being fitted to vehicles which combines adaptive cruise control and lane departure warning to allow easy slow driving in traffic jams. Moving forward to 2025 highway pilot may well be fitted as a means of traversing roads using intelligent camera and radar technology as well as valet parking. The later enabling the passenger to leave the latter vehicle to find a parking space by itself, only to be retrieved by means of a communicator. By 2030, it is likely that physical driving will be a thing of the past. As we progress along this transitional phase and the management of the vehicle becomes more automated in theory this should result in greater optimisation of the drive train. In other words, if we assume carbon-based fuels and combustion engines are here to stay (at least in the short to medium term) the drive towards autonomy may well result in reducing carbon dioxide emissions.

There has been a plethora of editorials written about the benefits of reduced carbon emissions that may accrue through more efficient driving patterns (often referred to as driving cycles) and the introduction of new technologies. However for many local air quality specialists, the elephant in the room is the lack of discussion about the changes we might expect from autonomous vehicles in terms of improving air quality in urban areas. Nitrogen oxide and PM are the pollutants of most concern to human health.

Far more will need to be considered as our detailed understanding of these driving cycles and systems grows, although a connection has yet to be quantified in terms of air quality and emissions. A positive step being undertaken by AECOM is to begin recording driving behaviour via instrumented vehicles and to use these data to inform microsimulation traffic and instantaneous emission models. The result is a second by second account of the emissions profile from roads under various traffic management regimes (see Figure 1)

Figure 1. Driving cycle of all vehicles on any given road section

Figure 1. Driving cycle of all vehicles on any given road section

Conjecture about Autonomous Transport
We can be confident that the fuel-mix in the UK fleet will become highly heterogeneous in the next few years, with increasing use of alternative fuels such as LPG, CNG, electric and hybrid combinations, plus continuing use of diesel and petrol combustion engines.  These fuels and technologies will have a range of emission characteristics, and may naturally aggregate around specific uses that provide the greatest economic benefits and flexibility.

We do not know how these vehicles will operate independently of a driver, although we do anticipate that urban and inter urban driving patterns of all vehicles when AVs are included will produce a smoother overall traffic response with far less braking and accelerating 5. As a result the real-world emissions may be very different when compared to human driving styles (either greater or less in different situations).

An obvious effect on emissions will be the fact that autonomous vehicles may increasingly operate without a passenger load and this is bound to have a reducing effect on emissions. Park assist and valet parking in urban areas may increase the dynamic of vehicle operations causing an increase in exhaust emissions and equally non-exhaust PM. Observations of park assist shows that there is no attempt to reduce the tyre road friction by gently moving backwards and forwards as full lock is activated.  Finally autonomous vehicles will use automatic gearboxes and so whilst this should enable engine load to be managed more effectively in urban driving situations automatics tend to consume more energy per kilometre driven. What effect this will have on NOx and PM is not fully understood.

What is clear is the level of communication that will be required to realise the autonomous vision. Vehicles will need to communicate with other vehicles, understand traffic management systems and navigate around obstacles in the road. Conservation of fuel and the management of exhaust emissions have just as much to do with the way the vehicle is driven as the vehicle itself. Individuals will have to make that choice between having the freedom to drive as they see fit or devolve that role to the vehicle manufacturer. At the moment most car drivers except that they are part of the air quality problem and the altruistic of us attempt to lessen our impact on emissions exploring low emission options. In the future, autonomous vehicles may widen the gap between perceptions of cause and effect as the vehicle becomes more of a utility than an experience. People may travel in cars who had previously not considered that to be a viable option (e.g. the very young or the very old). Under these circumstances road traffic may increase. That’s why it’s important that driver behaviour and subsequent driving cycles are understood in order to counter the effect this may have on increasing emissions and reducing air quality.

Conclusions and Summary
Autonomous vehicles will be seen on our roads but not in the near future. The transition will be stepwise as people get used to the various technologies. How this technology will affect emissions and air quality is uncertain.

Most of the research around emissions has focused on fuel consumption and hence the carbon benefits these vehicles may accrue. Very little research has focused on NOx and PM emissions and air quality. To gain a better understanding requires detailed analysis of driving behaviour using instrumented vehicles which can record second by second vehicle parameters. Using these data, comparisons between conventional and autonomous behaviour can be modelled to determine the impact on air quality and health. Analysis will also account for evidence emerging relating to real world vehicle emissions.

As more research is conducted the evidence concerning AV operation will become clearer and the uncertainty surrounding emissions estimates will continue to improve. AECOM will continue to follow the debate on AV emissions as part of a wider programme of developing suitable low emission solutions backed by a wave of technology. ◆

1     http://www.telegraph.co.uk/cars/news/study-says-electric-cars-could-emit-almost-as-many-particulates/
2     http://www.euro.who.int/__data/assets/pdf_file/0006/189051/Health-effects-of-particulate-matter-final-Eng.pdf
3     http://ec.europa.eu/eurostat/statistics-explained/index.php/Quality_of_life_indicators_-_natural_and_living_environment#Urban_population_exposure_to_air_pollution_by_particulate_matter
4     https://www.smmt.co.uk/wp-content/uploads/sites/2/AutoAnalysis-report-the-future-of-UK-automotive-manufactuing-October-2015.pdf
5     http://www.its-ukreview.org/autonomous-vehicles-to-bring-early-congestion-benefits/

Duncan Urquhart,  BSc (Hons) MSc CSci MIAQM, Principal Environmental Scientist, AECOM

Duncan Urquhart,
BSc (Hons) MSc CSci MIAQM, Principal Environmental Scientist, AECOM

Kevin Turpin, PhD, MIEnvSc, MIAQM, Principal Environmental Scientist, AECOM

Kevin Turpin, PhD, MIEnvSc, MIAQM, Principal Environmental Scientist, AECOM