Road transport is an essential service in society, but the burden of traffic crashes and Pollution is immense. European data show that road accidents in the EU countries cause about 250,000 seriously injured people every year, with 28,000 fatalities in 2012 (Kearns & Kidd, 2013). US data in addition, illustrate that automobile crashes led to 34,080 fatalities in 2012 (National Highway Traffic Safety Administration (NHTSA), 2013), where about 90% of the cases were attributed at least partially to driver error (Smith, 2013).
Therefore, car manufacturers have introduced a range of vehicle systems to assist drivers in their driving task, and subsequently enhance driving safety, comfort and traffic efficiency (Golias et al., 2002). Such systems are broadly known as Advanced Driver Assistant Systems (ADAS) and can be described as “systems developed to automate/adapt/enhance vehicle systems for safety and better driving” (Himabindu & Yasmeen, 2014). The ADAS, depending on their functions and design, are either primarily addressed to support the driver, and thuscalled driver support systems, e.g. navigation systems, vision enhancement, automated transactions and driver vigilance monitoring, or the vehicle and subsequently referred to as vehicle support systems, e.g. speed control, lane departure collision avoidance, obstacle detection (Golias et al., 2002). Building upon this generic classification more detailed ADAS categorisations can be found in the literature, e.g. (Golias et al., 2002; Shladover, 1995; Vahidi & Eskandarian, 2003).
Many studies have shown the positive impact of ADAS on traffic accidents, e.g. (Golias et al.,2002; Jermakian, 2011; Kahane & Dang, 2009; Kuehn et al., 2009; National Highway Traffic Safety Administration (NHTSA), 2000, 2005). Electronic Stability Control (ESC) and Antilock Brake Systems (ABS), which prevent a large proportion of both fatal and nonfatal crashes (Kahane & Dang, 2009), are now mandatory in new vehicles (iMobility Forum, 2013). This paper focuses on further ADAS, which are optional in a wide range of vehicles. Among them, the Implementation Road Map Working Group of the European eSafety Forum has selected those expected to reduce road fatalities and emissions in the short- and medium-termin Europe (eSafety Forum, 2008; van Calker & Flemming, 2012). These systems are known as Priority Systems, and are classified into:
- Blind Spot Monitoring systems, which use camera techniques with image processing or radar sensors to provide better vision into the blind spot area of a vehicle or supplemental information regarding an obstacle being in that area.
- Adaptive Headlights systems, which comprise electromechanically controlled headlights to ensure optimum illumination of the lane in bends.
- Obstacle and Collision Warning systems, which detect obstacles and warn the Drivers about an imminent collision.
- Lane Keeping Support systems, which indicate or warn the drivers when leaving the lane unintentionally.
- Emergency Braking systems, which detect obstacles and notify the drivers about animminent collision; in addition, in the case of a unavoidable collision the system then brakes automatically and forcefully.
- Eco Driving Support systems, which assist and encourage drivers to keep driving in amore environmentally friendly manner by providing them with information about the fuel consumption, energy-use efficiency and appropriate gear selection.
A large number of field operational tests, driving simulator studies and accident analysesstatistics worldwide, have shown and quantified the safety, economic and societal benefits derived from the implementation of the priority systems. Jermakian (2011), for example, has analysed US passenger cars crash records for the period 2004-2008 derived from the US National Automotive Sampling System General Estimates System (NASS GES) and Fatality Analysis Reporting System (FARS). Results showed that forward collision warning/mitigation system could prevent up to 20% of the total 5.8 million reported crashes in the US per year, including 66,000 serious and moderate injury crashes and 879 fatal crashes.
In addition, adaptive headlights were found to have the potential to prevent or mitigate 2.45% of all crashes, including about 2,480 fatal accidents and 29,000 injury accidents per annum, while the blind spot monitoring systems could prevent 6.8%, of all crashes, and 393 fatal accidents and 20,000 injury accidents respectively. Finally, it was shown that the lane departure warning system could prevent or mitigate up to 3% of allcrashes, including 7,529 fatal and 37,000 injury accidents per year. In addition, data collected from insurance companies showed that the US fleet of two passenger car manufacturers equipped with different types of ADAS have reduced the overall number of insurance claims (Highway Loss Data Institute, 2012a, 2012b). In particular, for the first manufacturer it was found that the frequency of bodily injury liability, medical payments and personal injury protection dropped by 3.6%, 26.5%, and 7.2% respectively, when vehicles were equipped with blind spot monitoring systems compared to identical vehicles not equipped with the system (Highway Loss Data Institute, 2012a). For the second manufacturer, the frequency of property damage liability claims was reduced by 2.4%. On the other hand, it also indicated an increase by 1.3% concerning the frequency of collision insurance claims, but also an average reduction of 159$ (or 11.4% in medical payments) in the average severity of the claimed costs (Highway Loss Data Institute, 2012b). The same data also showed the positive impact of adaptive headlights on property damage liability claims, the bodily injury liability claims and the medical payments claims (Highway LossData Institute, 2012a, 2012b).
Accident analysis in Europe for the period 2005-2006 (Wilmink et al., 2008) showed that the safety benefits of lane keeping support, if fully deployed in the vehicle fleet, would reduce thenumber of fatalities by 15.2% and injuries by 8.9% in EU25. The authors also found thatemergency braking has the potential to reduce both fatalities and injuries due torear-end-collisions and collisions against fixed obstacles by 7%.
The European New Car Assessment Programme (Euro NCAP) in collaboration with the Australasian New Car Assessment Programme (ANCAP) have also recently released a studyon the impact of low speed Autonomous Emergency Braking (AEB) on rear-end collisions (Fildes et al., 2015). For this study the authors used data collected from Australia, and five European countries. Results revealed a 38% overall reduction in rear-end crashes for vehicles fitted with low speed AEB compared to those without. Furthermore, results demonstrated nostatistical difference on the AEB effect between urban (above 60 km/h) and rural (below 60km/h) speed zones.
Anderson et al., (2011) analysed crash data derived from police reports in New South Wales, Australia between 1999-2008. Findings showed that adaptive cruise control in conjunctionwith automatic braking could reduce the number of vehicle fatal accidents by 7% and the number of injury accidents by 4%, at speeds larger than 60 km/h. Moreover, the authors found that lane departure warning would reduce the number of fatalities in all crashes by 7% and the number of injuries by approximately 9% in a year. In addition, a UK study by Robinson et al., (2011) showed that lane departure warning, when penetration rate reaches 100%, would reduce the number of fatalities in a year by 15-60, while the number of prevented serious injuries was estimated to be between 578 and 1,581. Moreover, the authors found that automatic emergency braking would reduce the number of fatalities of accidents involving thefront of a car impacting another vehicle by 30%.
A field operational study by Najm et al., (2006) showed that automatic collision avoidance systems, integrating forward collision warning and adaptive cruise control, have the potential to avoid about 6 to 15% of all rear-end crashes or between 133,000 and 687,000 rear-endcrashes in the US per year. Furthermore, Benmimoun et al., (2013) found among 100 passenger vehicles that a combined system of adaptive cruise control and forward collision warning increased the average time-headways between vehicles on motorways by 16%. In addition, the number of critical time-headways (less than 0.5 s) is reduced by 73%, while the study also showed that when adaptive cruise control is active, fuel consumption on motorways is reduced by 2.77%. Finally, Vagg et al., (2013), and Strömberg & Karlsson, (2013) investigated the impact of eco-driving features in two different cases. The former, based on a field test of 15 light commercial vehicles in the UK, showed that an eco-driving assistance system reduced fuel consumption by 7.61% on average with a reduction forindividual drivers between 0.43-12.03%. Results were obtained by advising drivers toaccelerate more smoothly and enforcing the gear shift indicator advice. The latter, based on astudy carried out by a Swedish public transport operator on a bus line, involving 54 busdrivers, found that a combination of eco-driving training and eco-driving advice system could reduce fuel consumption by 6.8%.
Positive results were also derived by several driving simulator studies. In this respect it shall be mentioned that driving simulator studies, due to their controlled conditions, provide more specific information on human behaviour, but behaviour and accident risks in Simulators deviate substantially from real world conditions. Bueno et al. (2013), confirmed that the forward collision warning system in passenger vehicles can reduce the reaction time of the driver when braking, due to reduction of the time that the drivers need to process the visual target at higher cognitive level. Moreover, Muhrer et al., (2012), found that forward collision warning in combination with emergency braking system resulted in significantly fewer accidents in critical situations. However, they also found that test drivers maintained higher mean speeds with the forward collision warning (with emergency braking) when noun expected events (such as the lead car braking hard) occurred. Finally, Dogan et al., (2011) showed that drivers would be less able to use eco-driving when the traffic environment is highly demanding, e.g. residential areas, and during critical situations. Results also suggest that performance on the fuel saving decreases under time pressure. On the other hand,concerns have been raised by simulator studies, regarding the impact of ADAS on drivingperformance. Jamson et al., (2008) for instance, found that the type of a system may have an impact on its adaptation from drivers. In particular, they found that the benefits of adaptiveforward collision warning systems were demonstrated only for aggressive drivers, wherea smarginal effects were found for non-aggressive drivers.Thus, field operational tests, driving simulator studies and accident statistics indicate substantial benefits of the priority systems, which are commercially available already sincethe early/mid 2000s. Table 1 illustrates the ADAS benefits based on the literature.
The potential of increasing safety by ADAS is so significant, that it is remarkable that theactual market uptake of the majority of the systems 10 years later is still limited. Specifically,within the EU 28 countries none of the systems was installed in more than 1/3 of thepassenger vehicles first registered in 2012, with most of the systems to be installed in lessthan 5% of the vehicles, as shown in Table 2.
Therefore this paper aims to analyse the deployment of the ADAS on the European roads and to investigate the reasons of the limited deployment rates. Subsequently, the main objective ofthis study is to quantify and compare ADAS deployment among European countries. Due to the limited availability of worldwide vehicle-specific data, we focus on recent sales (2012-2014) obtained from the European iCar database and two leasing companies in the Netherlands. In the discussion we interpret the findings and explore whether deployment rates could be affected by any governmental initiatives, e.g. incentives to consumers. In addition, we account for underlying causes responsible for the limited deployment rates of the ADAS. Inturn, we provide directions for future research, which should be focusing on relating deployment rates with overall road safety, i.e. the number of accidents, injuries and fatalities,and exploring any associations between the number of sales and the EuroNCAP results. Finally, we discuss the potential impact of the ADAS low deployment rates on the deployment of the forthcoming automated and connected vehicles technology.
Two sets of data were used to study the ADAS deployment in European countries. The first was extracted from the 2013 iCar implementation status survey (van Calker & Flemming, 2013). The second set of data was obtained from of a Dutch leasing company. The 2013 iCar implementation status survey includes the deployment rates of ADAS forpassenger cars first registered in 2012 in EU28. This data was derived from original equipment manufacturers (OEMs), and contains details for both the priority systems and their assigned relevant systems, as listed in Table 3.
Data represents passenger vehicles first registered in EU 28 in 2012. Data coverage varies per country and priority system due to various reporting methods and indicates the number of vehicles with available information, referred to all vehicles first registered within the regardedyear. Data coverage for 2012 is estimated to be about 76.6% for blind spot Monitoring systems, 53.4% for adaptive headlights, 66.3% for obstacle and collision warning, 66.5% forlane keeping support, 75.8% for emergency braking and 50.4% for eco driving support.
Based on this data, van Calker & Flemming (2013) calculated the direct deployment rates. The direct deployment rates are estimated by using only the vehicles of OEMs with available information and represent the amount of vehicles equipped with a regarded system, referred to all vehicles first registered within the regarded year. However, most of the OEMs nowadays provide only German-wide information about the number of vehicles equipped with a specific safety system (van Calker & Flemming, 2012). According to German regulations the OEMs must provide information about the safety systems equipped in their vehicles for use in the periodical technical inspections in Germany. In addition, the almost complete available European database on ADAS deployment rates, covers the built-in safety systems of passenger vehicles that first registered post April 2006 in Germany (van Calker & Flemming,2012).
Subsequently, among European countries, only the German direct deployment rates can be considered reliable. Hence, an alternative to direct deployment rates was adopted, referred to as corrected estimation deployment rates. These rates are calculated based on the reliably known German deployment rates and a system- and country specific correction factor, which represents the deviation between German and European direct deployment rates for each country and system. The deployment rates are then estimated by diving the sum number of vehicles equipped with the regarded system by the sum number of vehicles with available information (van Calker & Flemming, 2012).
The relatively small second dataset provides additional information on brand, type, price, and mass and was used in particular to investigate the relation between vehicle price & mass and ADAS deployment. This dataset comprises information on the Dutch and German market of a leasing company for the period 2013-2014 (until 31.08.2014). On the one hand, for the Dutch market, data provide details on (i) the amount of new vehicle orders per month, (ii) the amount of cars equipped with at least one type of ADAS, (iii) the orders per type of vehicle and OEM, and finally (iv) the type of the installed ADAS per vehicle. For the German market, on the other hand, data describes (i) the orders per OEM and year for those vehicles equipped with at least one ADAS, and (ii) the exact type of the installed ADAS per vehicle. For this data set, no information was available regarding the total amount of orders.
Analyses at international level
The deployment of the priority systems and assigned relevant systems among the EU28 countries were compared, based on the findings of the first dataset (van Calker & Flemming,2013). This dataset presents only the ADAS deployment, but provides no further analyses. Therefore, based on this data we firstly explored differences or similarities on the rates among the systems.
In turn, we further analysed the data conducting correlation analyses. Specifically, Spearman correlation coefficients (criterion for statistical significance at p < 0.05) were calculated between the nationwide ADAS deployments rates (corrected estimations) and (i) countries GDP per capita, (ii) countries’ GDP/capita in Purchasing Power Standards (PPS) to account for differences in the cost of living between countries, with GDP taken from (Eurostat, 2014a,2014b), and (iii) countries’ number of fatal road traffic accidents per 100,000 vehicles per year, with the number of fatalities derived from (UNECE, 2014), with all variables to be referring to 2012 data.
Analyses at national level
To consolidate and further understand and interpret the results at international level, the datasets from the Dutch leasing companies were analysed. First, the types of vehicles ordered by the two leasing companies were classified according to their mass and price. Then, for every vehicle, the installed ADAS were documented, and compared to the findings of the international study. This step aimed at identifying any differences or similarities significant changes in the rates between 2012-2014. Finally, the following hypotheses were tested:
- H1: The type of ADAS sold is correlated with the price of the vehicle
- H2: The type of ADAS sold is correlated with the mass of the vehicle
which could subsequently relate the ADAS market deployment to the type and cost of vehicles sold per country, and availability of ADAS per type of vehicle.
Results at international level
Figures 1 and 2 illustrate the deployment rates per priority system and EU country, with the countries sorted by GDP per capita in PPS. Figure 1 shows the rates for the safety related ADAS, while Figure 2 presents the rates for the Eco Driving systems. Three main observations can be made from the Figures 1 and 2. At first, in the Figure 1 it can be observed that overall the aggregated deployment rates for the safety related ADAS across Europe is relatively low. Only Luxemburg and Germany demonstrate fairly higher rates compared to the other countries. Further to that, it can also be seen that not only the countries with low number of registrations (Tables A.6, in Appendix) and/or GDPs but also the countries with higher number of registrations and/or reasonably high GDPs, indicate low deployment rates for the safety related ADAS compared to the average 2 EU 28 rates. Secondly, Figure 2 Shows that the eco driving support systems are overall the most commonly installed ADAS in mostof the EU countries. Thirdly, data indicate that only Portugal deviates strongly from the overall picture that countries with GDPs per capita in PPS higher than the EU 28 average (EU28avrg = 100) tend to demonstrate higher ADAS deployment.
Table 4 summarises the correlation matrix of the priority and relevant systems in relation to the testing variables. All the required for the calculations data are given in the Appendix.
Firstly, with respect to the priority systems (bold font), it can be seen that in general most of the correlations are moderate or strong, except those referring to Adaptive Headlights. Clear correlations of the blind spot monitoring systems were identified in relation to number of fatalities, and GDP per capita in PPS.
In particular, results show that higher market deployment of blind spot monitoring systems are found in countries with higher GDPs per capita in PPS. Data also show that in countries with lower numbers of fatalities the deployment of blind spot monitoring systems is higher. On the other hand, although nominal GDPs are also positively related to the deployment of the system, these correlations are not significant.
Weak and non-significant correlations were found between the adaptive headlights and testing variables. Nonetheless, data again show that higher number of new registrations and GDPs, imply higher deployment of the system. On the contrary, it is also shown that higher deployment rates are correlated with lower number of fatalities.
For the remaining priority systems data indicate significant correlations for most of the testing variables. For instance, for all the systems significantly higher deployment rates were found in countries with higher GDPs. Furthermore for all, but the emergency braking system, it was found that higher deployment is significantly correlated to lower number of fatalities. Nonetheless, results for the emergency braking systems, even non-statistically significant, show that higher deployment is associated to lower number of fatalities.
Consistent with the results about the priority systems, findings for the relevant systems provide evidence of significant correlations between ADAS deployment rates and testing variables. Specifically, data show that the great majority of the relevant systems are positively correlated with the GDPs per capita and negatively correlated with the number of road fatalities. Amongst them, the strongest correlations can be seen for the adaptive cruise control and start-stop ADAS systems. Lane departure warning and emergency braking systems indicate also significant positive correlations with the GDPs. Yet, the negative correlations between those systems with the number of fatalities are not significant.
In addition, results reveal very strong negative correlations between GDPs and the number of road fatalities per 100,000 passenger cars. In particular, the correlation between nominal GDP per capita and number of road fatalities was ρ = -.836 (p < .001, N = 25), and the correlation between GDP per capita in PPS and number of road fatalities was ρ = -.832 (p < .001, N = 25). Both findings highlight that in higher income countries the number of road fatalities is lower.
Analyses at national level
The first set of data at international level did not contain any information on the type of passenger cars equipped with ADAS systems for 2012. This information is presented in this section, using the second set of data. Vehicles were first grouped according to Euro NCAP (2014) classification, into: superminis, small family, large family, and executive cars , small MPVs, large MPVs, small off-road 4x4, and large off-road 4x4, roadster sports cars and finally business and family vans . Then, the installed ADAS for each vehicle were allotted. Figure 3 presents the overall information on the Dutch orders. In total 30,813 new vehicles were ordered, with only 3,061 (about 10%) of them equipped with at least one of the relevant safety priority systems (PS1-PS5). Out of those vehicles approximately 60% were equipped with multiple ADAS.
In addition, data for the German market indicate 10,188 new orders. 23% of the orders referred to vehicles with one ADAS on board, while the remaining 77% to vehicles with multiple ADAS. However, the total number of the German orders was not disclosed.
Figure 4 illustrates the ADAS deployment per type of vehicle, out of all vehicles equipped with at least one ADAS, for the German and Dutch market (the sum of ADAS deployment rates for all type of vehicles per country equals 100%). Data clearly show that most of the vehicles, regardless their type, are equipped with more than one ADAS. The different ADAS packages are given in Table A.5 in the Appendix, with 40% of them including ACC. Subsequently, data show that ACC is installed in about 50% of the vehicles equipped with at least one ADAS.
Figure 4 does not include information on superminis and roadster vehicles due to their small sample. Yet, for the former, data showed that Collision Warning Systems were the mostly installed ADAS, while for the latter, the majority was equipped with ACC.
Statistics for the Netherlands, specifically, showed that the more expensive brands, i.e. premium vehicles, are equipped with more than one ADAS, regardless the vehicle category as classified by Euro NCAP (ρ = -.179, p < .001). In addition, results confirmed our first hypothesis, i.e. “H1: The type of ADAS sold is correlated with the price of the vehicle”. In particular, results showed that premium vehicles are significantly more equipped with ACC or EB or ACC and EB, compared to the mass market vehicles, with Spearman correlations ρ = .585 (p < .001) and ρ = .558 (p < .001) respectively.
Among only the premium vehicles, no significant correlation was found between the mass of the vehicles and the installed ADAS. Spearman correlation for the vehicles equipped with ACC or EB compared to those with no ACC or EB was found ρ = -.013 (p < .693), and for the vehicles installed with ACC and EB compared to those with no ACC or EB ρ = -.031 (p < .351). Finally, among only the mass market vehicles findings indicate significant correlations between the mass of the vehicles and types of ADAS installed. Specifically it was found that bigger vehicles are equipped with less ACC, EB or ACC and EB systems compared to smaller vehicles, with Spearman correlations ρ = -.187 (p < .001) and ρ = -.072 (p < .001).
Subsequently, our second hypothesis, i.e. “H2: The type of ADAS sold is correlated with the mass of the vehicle”, is valid for the mass market vehicles but should be rejected for the premium vehicles.
Field operational tests, driving simulator studies and accident statistics indicate substantial benefits of ADAS in terms of safety and fuel consumption. In this study we explored the deployment of ADAS in EU 28, and their relation to the countries GDPs and fatalities. In addition, we investigated the relation between ADAS sold and the price and mass of vehicles. Overall, the ADAS deployment in the EU 28 is limited. Especially for the safety related ADAS, i.e. Blind Spot Monitoring, Adaptive Headlights, Obstacle and Collision Warning, Lane Keeping Support, and Emergency Braking, data showed they are typically installed in less than 5% of the vehicles registered for first time in 2012 (with the exception of adaptive headlights with 12.6% deployment). The Eco Driving systems, on the other hand, are installed in about . of the total vehicles first registered in 2012 in EU 28. This could be explained by national taxation policies, the relatively small cost of those systems and their availability in all passenger cars ranges. In particular national environmental policies may impact the deployment of Eco Driving systems. This, for instance, applies to the Netherlands tax policy, which effectively stimulates the purchase of smaller and eco-friendly cars.
Our findings show that ADAS deployment relates to the countries GDPs per capita, as well as the number of road fatalities. These findings are in agreement with the results by De Winter et al., (2015). Specifically, we found that significantly more ADAS were sold in richer countries and countries with lower numbers of road fatalities per 100,000 passenger cars. Subsequently, results suggest that safe countries become even safer.
Data from the leasing companies, confirm the limited deployment of the safety related ADAS. In particular, in 2013 – 2014, of 30,813 new vehicles ordered in the Netherlands, only 3,061 (about 10%) of them were equipped with at least one of the relevant safety ADAS. Although this data may not be fully representative of the fleet driven in the Netherlands, since people who lease may tend to lease more luxurious vehicles than those they would purchase privately, yet they display a clear correlation between the price of vehicles and types of ADAS bought. That is, the premium vehicles are significantly more equipped with Adaptive Cruise Control and Emergency Braking compared to the mass market vehicles. We also found that the availability of the different safety systems is significantly higher in more expensive cars. On the other hand, our results showed that the installation of ADAS depends also on the brand of the vehicle. There are specific brands that have already installed a large number of ADAS to a broad range of their fleet.
The limited ADAS deployment could be explained by the fact that a significant amount of people are afraid of technological interventions, or may not want to spend a lot of money to buy such features, as established in an on-line survey by Kyriakidis et al., (2015). In addition, deployment rates have been held back due to lack of legislative clarity within the European Commission on its plans on mandating car manufacturers to equip private cars with ADAS (SDB, 2011). Finally, deployment rates may have also been hindered as neither the Euro NCAP nor the U.S. National Highway Traffic Safety Administration (NHTSA) till recently included the testing of ADAS as part of their five star rating. The Euro NCAP, in particular, for its five star rating accounts only for the testing of two priority systems, the Lane Support and Autonomous Emergency Braking, with the latter being introduced in 2014 (Euro NCAP, 2015).
However, it is now expected that in Europe over 70% of vehicle models sold by 2020 will be fitted with at least one type of ADAS (Rangarajan & Dunoyer). Subsequently, the penetration of radars and cameras is expected to rise by 2020 from 16% and 11% to 63% and 69% respectively (Rangarajan & Dunoyer). In addition, on March 31, 2014, NHTSA announced that it will require all new vehicles under 10,000 pounds to have rear view cameras by May 2018. Thus, OEMs should work towards the installation of the ADAS on all cars regardless of the size and price. In turn, EU administrations could build upon Euro NCAP’s decision and force OEM’s to install ADAS in all types of cars. Furthermore, governments could also provide (tax) incentives for the consumers to buy vehicles equipped with ADAS. Incentives for eco driving should be continued, and should, given the still excessive cost of road accidents, be extended with incentives for safety related ADAS.
The results provide a comprehensive overview of the current ADAS deployment in EU 28. However, generalization of the findings derived from the analysis of the leasing companies dataset at large, requires caution, statistical weighting and consistency in data collection. Thus, it is desired to establish a central data repository and standard collection process about the ADAS systems installed in all vehicles across Europe.
Our findings give important reference information on current ADAS deployment. Moreover, they could be used to explore whether the limited ADAS deployment rates may reflect the deployment rates of the forthcoming automated and connected vehicle technology. It could be argued, for instance, that similar to the ADAS, the public perception, acceptance and trust in technological interventions, as well as the concerns about automated systems reliability and safety, in addition to their price, will impact substantially the deployment rates of automated and connected vehicles (Kyriakidis et al., 2015). However, towards the deployment of automated and connected vehicles additional challenges should be resolved, including human factors questions and liability uncertainties.
Building upon this study, future studies should be focusing on comparing the frequency and severity of the accidents involve vehicles with ADASs, to the frequency and severity of the accidents involve the same type of vehicles equipped with no ADASs. Subsequently, we could in detail measure the safety benefits of the ADAS on public roads. Finally, future research should investigate and quantify the financial long-term benefits of ADAS extensive deployment, in addition to the deployment of highly automated vehicles on public roads.
The authors are most grateful to Mr Risto Kulmala from the Implementation Road Map working group of the iMobility Forum and Mr. Hans Tonneijk from De Lage Landen International for providing the data used in this study.
1. Anderson, R. W. G., Hutchinson, T. P., Linke, B., & Ponte, G. (2011). Analysis of crash data to estimate the benefits of emerging vehicle technology. Adelaida, Australia: The University of Adelaide. Retrieved from http://casr.adelaide.edu.au/casrpubfile/1081/CASR094.pdf
2. Benmimoun, M., Pütz, A., Zlocki, A., & Eckstein, L. (2013). euroFOT: Field Operational Test and Impact Assessment of Advanced Driver Assistance Systems: Final Results Proceedings of the FISITA 2012 World Automotive Congress (Vol. 197, pp. 537-547): Springer Berlin Heidelberg. Retrieved from http://dx.doi.org/10.1007/978-3-642-33805-2_43
3. Bueno, M., Fort, A., Francois, M., Ndiaye, D., Deleurence, P., & Fabrigoule, C. (2013). Effectiveness of a Forward Collision Warning System in simple and in dual task from an electrophysiological perspective. Neuroscience Letters, 541 (0), 219-223. doi: http://dx.doi.org/10.1016/j.neulet.2013.02.030
4. De Winter, J. C. F., Kyriakidis, M., Dodou, D., & Happee, R. (2015). Using CrowdFlower for international survey research: A study on traffic violations. Paper accepted at the 6th International Conference on Applied Human Factors and Ergonomics. Las Vegas, NV.
5. Dogan, E., Steg, L., & Delhomme, P. (2011). The influence of multiple goals on driving behavior: The case of safety, time saving, and fuel saving. Accident Analysis & Prevention, 43 (5), 1635-1643. doi: http://dx.doi.org/10.1016/j.aap.2011.03.002
6. eSafety Forum. (2008). Final Report and Recommendations of the Implementation Road Map Working Group.
7. Euro NCAP. (2014). European car classification. Euro NCAP. Retrieved from http://www.euroncap.com/home.aspx
8. Eurostat. (2014a). GDP per capita - annual Data. European Commission. Retrieved from http://appsso.eurostat.ec.europa.eu/nui/show.do?dataset=nama_aux_gph&lang=en
9. Eurostat. (2014b). GDP per capita, consumption per capita and price level indices. European Commission. Retrieved from epp.eurostat.ec.europa.eu/statistics_explained/index.php/GDP_per_capita,_consumpt ion_per_capita_and_price_level_indices
10. Fildes, B., Keall, M., Bos, N., Lie, A., Page, Y., Pastor, C., Pennisi, L., Rizzi, M., Thomas, P., & Tingvall, C. (2015). Effectiveness of low speed autonomous emergency braking in real-world rear-end crashes. Accident Analysis & Prevention, 81 (0), 24-29. doi: http://dx.doi.org/10.1016/j.aap.2015.03.029
11. Golias, J., Yannis, G., & Antoniou, C. (2002). Classification of driver assistance systems according to their impact on road safety and traffic efficiency. Transport reviews., 22 (2), 179-196.
12. Highway Loss Data Institute. (2012a). Mercedes-Benz collision avoidance features: initial results. Arlington, VA: Highway Loss Data Institute.
13. Highway Loss Data Institute. (2012b). Volvo collision avoidance features: initial results. Arlington, VA: Highway Loss Data Institute.
14. Himabindu, S., & Yasmeen, N. (2014). Driver Assistance System - Shared Steering Between Human and Automation. International Journal of Advances in Applied Science and Engineering (IJAEAS), 1 (3), 199-202.
15. iMobility Forum. (2013). Monitoring Priority Systems. 05 November, 2014. Retrieved from http:// www.imobilitysupport.eu/imobility-support/work-packages/wp3-support-to-its-deplo yment/monitoring-priority-systems
16. Jamson, A. H., Lai, F. C. H., & Carsten, O. M. J. (2008). Potential benefits of an adaptive forward collision warning system. Transportation Research Part C: Emerging Technologies, 16 (4), 471-484. doi: http://dx.doi.org/10.1016/j.trc.2007.09.003
17. Jermakian, J. S. (2011). Crash avoidance potential of four passenger vehicle technologies. Accident Analysis & Prevention, 43 (3), 732-740. doi: http://dx.doi.org/10.1016/j.aap.2010.10.020
18. Kahane, C. J., & Dang, J. N. (2009). The Long-Term Effect of ABS in Passenger Cars and LTVs. Washington, DC: National Highway Traffic Safety Administration.
19. Kearns, H., & Kidd, D. (2013). Road safety: EU reports lowest ever number of road deaths and takes first step towards an injuries strategy. Brussels, Belgium: E. Commission. Retrieved from http://europa.eu/rapid/press-release_IP-13-236_en.htm
20. Kuehn, M., Hummel, T., & Bende, J. (2009). Benefit estimation of advanced driver assistance systems for cars derived from real-life accidents. Presented at the 21st International Technical Conference on the Enhanced Safety of Vehicles Conference (ESV), Stuttgart, Germany. Retrieved from http://www-nrd.nhtsa.dot.gov/pdf/esv/esv21/09-0317.pdf
21. Kyriakidis, M., Happee, R., & De Winter, J. C. F. (2014). Public opinion on automated driving: Results of an international questionnaire among 5,000 respondents. Transportation Research Part F: Traffic Psychology and Behaviour, 32, 127-140. doi: http://dx.doi.org/10.1016/j.trf.2015.04.014
22. Muhrer, E., Reinprecht, K., & Vollrath, M. (2012). Driving With a Partially Autonomous Forward Collision Warning System: How Do Drivers React? Human Factors: The Journal of the Human Factors and Ergonomics Society, 54 (5), 698-708.
23. Najm, W. G., Stearns, M. D., Howarth, H., Koopmann, J., & Hitz, J. (2006). Evaluation of an Automotive Rear-End Collision Avoidance System. Cambridge, MA: National Highway Traffic Safety Administration. Retrieved from http://ntl.bts.gov/lib/jpodocs/repts_te/14303_files/14303.pdf
24. National Highway Traffic Safety Administration (NHTSA). (2000). Automotive Collision Avoidance System (ACAS) Program - Final Report. Washington, DC, US. Retrieved from http://babel.hathitrust.org/cgi/pt?id=pst.000044931445;view=1up;seq=3
25. National Highway Traffic Safety Administration (NHTSA). (2005). Automotive Collision Avoidance System Field Operational Test. Washington, DC, US. Retrieved from http://ntl.bts.gov/lib/jpodocs/repts_te/14244/pages/14244.pdf
26. National Highway Traffic Safety Administration (NHTSA). (2013). Early Estimate of Motor Vehicle Traffic Fatalities in 2012. Washington, DC. Retrieved from http://www-nrd.nhtsa.dot.gov/Pubs/811741.pdf
27. Rangarajan, D., & Dunoyer, A. The global market for ADAS will grow to €7.2 billion by 2020: SBD. Retrieved from http://www.sbd.co.uk/files/sbd/pdfs/538IB.pdf
28. Robinson, B., Hulshof, W., Cookson, R., Cuerden, R., Hutchins, R., & Delmonte, E. (2011). Cost Benefit Evaluation of Advanced Primary Safety Systems - Final Report. Leeds, UK: Transport Research Laboratory. Retrieved from http://www.trl.co.uk/reports-publications/trl-reports/road-safety/report/?reportid=6724
29. Shladover, S. E. (1995). Review of the State of Development of Advanced Vehicle Control Systems (AVCS). Vehicle System Dynamics, 24 (6-7), 551-595. doi: 10.1080/00423119508969108
30. Smith, B. W. (2013). Human Error as a Cause of Vehicle Crashes. Retrieved from http://cyberlaw.stanford.edu/blog/2013/12/human-error-cause-vehicle-crashes
31. Strömberg, H. K., & Karlsson, I. C. M. (2013). Comparative effects of eco-driving initiatives aimed at urban bus drivers – Results from a field trial. Transportation Research Part D: Transport and Environment, 22 (0), 28-33. doi: http://dx.doi.org/10.1016/j.trd.2013.02.011
32. UNECE. (2014). Road Traffic Accidents - Number of fatalities per 100 000 passenger cars. United Nations Economic Commission for Europe. Retrieved from http:// w3.unece.org/pxweb/database/STAT/40-TRTRANS/01-TRACCIDENTS/?lang=1
33. Vagg, C., Brace, C. J., Hari, D., Akehurst, S., Poxon, J., & Ash, L. (2013). Development and Field Trial of a Driver Assistance System to Encourage Eco-Driving in Light Commercial Vehicle Fleets. Intelligent Transportation Systems, IEEE Transactions on, 14 (2), 796-805. doi: 10.1109/TITS.2013.2239642
34. Vahidi, A., & Eskandarian, A. (2003). Research advances in intelligent collision avoidance and adaptive cruise control. Intelligent Transportation Systems, IEEE Transactions on, 4 (3), 143-153. doi: 10.1109/TITS.2003.821292
35. van Calker, J. r., & Flemming, A. (2012). iCar - implementation status survey by use of OEM data 2012 - Deployment state in EU member states - Final report: iCar.
36. van Calker, J. r., & Flemming, A. (2013). iCar - implementation status survey by use of OEM data 2013 - Deployment state in EU member states - Final report: iCar.
37. Wilmink, I., Janssen, W., Jonkers, E., Malone, K., van Noort, M., Klunder, G., Rämä, P., Sihvola, N., Kulmala, R., Schirokoff, A., Lind, G., Benz, T., Peters, H., & Schönebeck, S. (2008). Impact Assessment of Intelligent Vehicle Safety Systems: eIMPACT Consortium. Retrieved from http://www.eimpact.info/download/eIMPACT_D4_v2.0.pdf