Home / Technology Policy & Ethics / July 2019 / Adaption of Autonomous Vehicles (via APIs) in Society

Adaption of Autonomous Vehicles (via APIs) in Society

Manish Rathore and Inderjit Rai, Department of Computer Science, Ryerson University, Canada, and Salah Sharieh, Ph.D, Royal Bank of Canada, Toronto, Canada

July 2019

Introduction

Autonomous Vehicles (AVs) were considered an interesting topic in a science fiction movie such as Knight Rider, however, AVs are becoming a reality and will change the way we get around. Not only the major automakers but software companies such as Google, Uber, and Apple are all developing  AV capabilities. Experts have categorized AVs into five different levels. In this work we will primarily focus on Level 5, which is defined as vehicles that “can do all the driving in all circumstances”.  The human occupants are just passengers and need never be involved in driving1. There has been quite a bit of discussion around the benefits of Level 5 autonomy, we would like to cover whether people are ready to embrace the AV future, or all these companies wasting their time and won’t see returns for the foreseeable future.

The research achievement of Level 5 will be detrimental to traditional car manufacturers, as it will increase their liability in accidents, as well as lower their revenues as fewer cars are sold, since one household may not need multiple cars.  The increased liabilities from accidents coupled with lower revenues is likely to reduce stock values and the number of employees at these firms.

The main stakeholders of this technology are the public who currently drive or use public transportation, car manufacturers, and technology companies.  A few of the technology companies, such as Alphabet with its self-driving car, Waymo, are leading the way in developing the next generation of self-driving cars and will play a key role in shaping the future of the field.  Many consider Waymo as the most advanced self-driving system in the world, and the car has driven more than 10 million miles on public roads in the United States without any major incidents2.

Car manufacturers also view self-driving cars as the future, and they have all started investing billions in the technology.  For example, GM has invested over seven billion USD into its self-driving division, Cruise Automation3.  Other stakeholders are the people who drive for a living or employees of car companies, because in the best-case scenario the nature of their work may change, and in the worst case they may find themselves out of work.

Challenges associated with AV Adaption 

Despite the above-mentioned concerns, we can expect fully autonomous vehicles from all major automakers in the near future. These vehicles solve various problems such as saving lives from car crashes, reducing the overhead of policing roads and highways, increasing the viability of on-demand Uber like services, increasing productivity, and reducing greenhouse gas emissions. At the same time, autonomous cars present major challenges to our society. One of those challenges is the ethics of the decision-making process for these vehicles.  A classic philosophical exercise known as the trolley problem, when applied to self-driving cars, poses a moral dilemma that humans face4.  For instance, if brakes fail, should the car try to save the people on the road or the driver in the car? What if the pedestrians on the road are children, doctors, or criminals? Germany has recently implemented a set of ethics rule that autonomous cars must follow5. Would these rules set a baseline that other countries can build from? Before autonomous cars become mainstream, there are many questions that we will need to be addressed.  In general, humans are less forgiving of machines making mistakes than they are towards humans, even after having evidence of the machine performing better than humans6.  Even extremely rare incidents of self-driving cars getting into minor accidents make headline news, which propagates people’s fears of the machine.  For example, even though Tesla’s semi-self-driving mode had been in use for some time, the one fatal incident in 2018 made international news7 and months of investigations and inquiries.  To put this in perspective, self-driving features have been around a few years now, and there have only been a handful of incidents, whereas 1.25 million people around the globe are killed in car crashes every year (W.H.O, 2013), with more than thirty thousand in the United States alone8.

With the advancements in communication technologies, there is myriad of applications and services that target consumers in different sectors ranging from finance, health, agriculture, smart industries, smart environment, and human well-being. The Internet of Things (IoT) is the best example of such applications and services realized through the interconnection of smart objects for different purposes such as, but not limited to, controlling cars remotely, monitoring car conditions, monitoring passengers, operating the car in difficult environments to automate, prioritizing traffic in jammed highways, and so on.

Furthermore, these services are also shared across different platforms with different consumers, vendors, government agencies, and other related entities. To make these services available for AVs, we need a unified mechanism to make the applications and services easily get accessed securely and interact with the heterogeneous consumer demands.

To this end, an Application Programming Interface (API) is a mechanism that makes it easy, affordable, and scalable for the services to integrate with AVs. APIs enable service integration, application development, and communication among different services and products without the need for developing new infrastructure for each service and product.  We believe integrating AVs with a broad network of services and distributed intelligence through an API will create a huge impact on the way society will function.

Adaption 

The attitudes towards the adaption of autonomous vehicles is mixed. In our research we found that people perceive AVs as the future but are not ready to be the early adopters of the technology. It is clear that most people wouldn’t want to pay for AV technology at its current projected price point. The technology would needs to mature, become safer and a lot cheaper in order for the masses to adopt it. Besides the willingness to pay extra for the technology, there were other factors that affected people’s attitudes towards autonomous vehicles. These factors focus on specific aspects that could influence a person’s attitude toward the technology, such as age, gender, educational background, profession, previous experience with autonomous vehicles, wealth and cultural differences.

Impact of age on people’s acceptance of AVs

It is generally believed that younger people are quicker to adopt new technologies than older generations.  A study by Jiang9 found that millennials led older Americans in their adoption and use of technology.   However, research by Anderson and Perrin10 discovered that while the older generation lags in technology adaptation as compared to the younger generation, their adoption of technology has increased over time.  For example, 43% of American adults ages 65 and older now own smartphones, up from just 18% in 2013, as shown in Figure 1.  It seems that the older generation is becoming more accepting of technology.

Figure 1. Average comfortability between age groups

Impact of income on people’s acceptance of AVs

The lower-income households generally tend to lag in technology adoption as compared to higher-income households10. We suspect that this lag in adoption will carry over to AVs as well. Not accounting for other factors, the cost of these new cars could lead to lower adoption rates for people with lower incomes. In our research, we found that people with higher income have a more accepting attitude towards AVs, as shown in Figure 2.

Figure 2. Count and comfort level of various income classes

Impact of profession on people’s acceptance of AVs

Certain professions tend to use technology more than others as shown in Figure 3.  Would professionals, such as technology specialists, be more accepting of autonomous vehicles than people in other jobs, such as labor? We have seen that automation in general is threatening a significant portion of lower-income jobs, which has made people in specific roles wary of autonomous technology. According to the estimate by the Department of Commerce, one in nine workers are in professions that will get impacted by introduction of AVs.  Naturally, we can expect taxi drivers, road freight drivers, and public transportation drivers to not have a favorable opinion of AVs.

Figure 3. Comfort level by profession

Conclusion

We found several correlations between age, profession, income group etc., and AVs adaption and studied the respective influence on the comfort levels with AVs.

Overall, we can conclude with high confidence that most people will be uncomfortable towards AVs. There are a variety of reasons that go into this such as fear of losing jobs, fear of accidents and lack of comfort with new technology.

Secondly, our hypothesis focused on income levels and their impact on comfort level towards AVs. On average, all income classes are uncomfortable with driving AVs. Specifically, the working class is the least comfortable with AVs with an average of 3.07 comfort level, whereas the middle class is the most comfortable with AVs with an average of 4.34 comfort level. We further looked at the regression model and found that none of the household income classes were over the 90% confidence interval and therefore we can conclude the income classes are statistically insignificant in determining comfort levels with AVs.

References 

  1. Hayes, A. (2018). Self-Driving Cars Could Change the Auto Industry. Investopedia.  https://www.investopedia.com/articles/personal-finance/031315/selfdriving-cars-could-change-auto-industry.asp
  2. Korosec, K. (2018). Waymo’s self-driving cars hit 10 million miles.  Tech Crunch.  https://techcrunch.com/2018/10/10/waymos-self-driving-cars-hit-10-million-miles/.
  3. Krisher, T. (2019).  GM Cruise autonomous vehicle unit gets $1.15B investment.  The Associated Press.  https://www.apnews.com/e697fb2f8c324302922991b2487da413
  4. Bakewell, S. (2013). Clang Went the Trolley. The New York Times.   https://www.nytimes.com/2013/11/24/books/review/would-you-kill-the-fat-man-and-the-trolley-problem.html?_r=0
  5. Schaft, P. (2018). Germany Creates Ethics Rules for Autonomous Vehicles. Robotics Business Review.   https://www.roboticsbusinessreview.com/unmanned/germany-creates-ethics-rules-autonomous-vehicles/
  6. Frick, W. (2013). When Your Boss Wears Metal Pants. Harvard Business Review.   https://hbr.org/2015/06/when-your-boss-wears-metal-pants.
  7. Levin, S. (2018). Tesla fatal crash: ‘autopilot’ mode sped up car before driver killed, report finds.  The Guardian.  https://www.theguardian.com/technology/2018/jun/07/tesla-fatal-crash-silicon-valley-autopilot-mode-report.
  8. Beltz, B. (2018.  100+ Car Accident Statistics for 2019. Safer America. https://safer-america.com/car-accident-statistics/
  9. Jiang, J. (2017). Millennials stand out for their technology use, but older generations also embrace digital life. Pew Research Centre.  https://www.pewresearch.org/fact-tank/2018/05/02/millennials-stand-out-for-their-technology-use-but-older-generations-also-embrace-digital-life/
  10. Anderson, M., Perrin, A. (2017) Tech Adoption Climbs Among Older Adults. Pew Research Centre.  https://www.pewinternet.org/2017/05/17/tech-adoption-climbs-among-older-adults/

Manish Rathore is a graduate student in the department of Computer Science at Ryerson University, Toronto currently conducting research on autonomous vehicles.

Inderjit Rai is a graduate student in the department of Computer Science at Ryerson University, Toronto currently conducting research on autonomous vehicles.

Dr. Salah Sharieh; Senior Director (Developer Experience and Open Innovation): Dr. Salah is a senior Director at RBC with extensive experience in business, technology and digital transformation. Salah holds the degree of Doctor of Philosophy from McMaster University.  He has more than twenty-five years of industry experience. He has several peer-reviewed publications and has contributed to several books. Salah is a member in the Yeates School of Graduate Studies at Ryerson University where he supervises Ph.D. and Master Students.

 

Editors:
Ali Kashif Bashir is a Senior Lecturer at School of Computing, Mathematics, and Digital Technology, Manchester Metropolitan University, United Kingdom. He is a senior member of IEEE and Distinguished Speaker of ACM. His past assignments include Associate Professor of Information and Communication Technologies, Faculty of Science and Technology, University of the Faroe Islands, Denmark; Osaka University, Japan; Nara National College of Technology, Japan; the National Fusion Research Institute, South Korea; Southern Power Company Ltd., South Korea, and the Seoul Metropolitan Government, South Korea.He received his Ph.D. in computer science and engineering from Korea University, South Korea. MS from Ajou University, South Korea and BS from University of Management and Technology, Pakistan. He is author of over 80 peer-reviewed articles. He is supervising/co-supervising several graduate (MS and PhD) students. His research interests include internet of things, wireless networks, distributed systems, network/cyber security, cloud/network function virtualization, etc. He is serving as the Editor-in-chief of the IEEE Future Directions Newsletter: Technology Policy and Ethics.He is editor of several journals and has served as a guest editor on several special issues in journals of IEEE, Elsevier, and Springer. He has served as chair (program, publicity, and track) chair on several conferences and workshops. He has delivered several invited and keynote talks, and reviewed the technology leading articles for journals like IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS, the IEEE Communication Magazine, the IEEE COMMUNICATION LETTERS, IEEE Internet of Things, and the IEICE Journals, and conferences, such as the IEEE Infocom, the IEEE ICC, the IEEE Globecom, and the IEEE Cloud of Things.

Dr. Fatima Hussain is working as Security Analyst in “API Security and Governance” squad, Royal Bank of Canada. She is leading the development and promotion of new API and API development learning curriculum along with API security and governance duties. Dr Hussain’s background includes number of distinguished professorships at Ryerson University and University of Guelph, where she has received awards for her research teaching and course development accomplishments within Wireless Telecommunication, Internet of Things, and Machine Learning. She has a long list of research publications in top tier conferences, books and journals.  Dr. Hussain holds Doctorate and Master of Science, degrees from Ryerson University in Electrical and Computer Engineering.