The race towards autonomous fleets has reached a tipping point with Aptiv, GM and Waymo planning to launch their self-driving robo-taxis in major cities of the United States by 2019. Though technology for autonomous vehicles is advancing rapidly, it is not without its fair share of challenges and, below, we take a look at some of the most important ones currently faced.
User Experience & Acceptance
Vehicles can drive themselves, but people are still going to compare their level of comfort and the whole driving experience in a human-driven vehicle against an automated one. Still being in the initial stages, people need to first start trusting autonomous vehicles in order to be ready to give up habituated driving controls. Although Aptiv, Volvo and Waymo already have dedicated UX projects with autonomous vehicles, the number of these projects across the automotive industry will rise significantly in the upcoming years. UX questions arise with every aspect of the autonomous vehicle. While GM is taking care of accessibility requirements in autonomous vehicles, Uber is trying to help riders get over motion sickness. Domino’s Pizza is testing autonomous pizza deliveries in partnership with Ford to understand how customers react to this new delivery process and, as an example, the tests indicated that customers preferred to use the last four digits of their phone number instead of their credit card number to validate their pizza. The interaction between the passenger and the autonomous vehicle will become the most important part of the UX design and passengers will want to feel a sense of familiarity and empathy, whether it is the dashboard or the control system of the vehicle.
Code of Ethics for Artificial Intelligence
Industry leaders have been long talking about this trolley problem. Though autonomous vehicles are designed to avoid accidents in the first place, these will eventually happen especially in the intersection of autonomous and conventional driven vehicles. No matter the unlikeness of an accident, the autonomous vehicle needs to make a split-second decision. Based on what set of morals or ethics will these decisions be made? Yes, Germany is already working on a code of ethics, but how long will it take for a universally accepted code to be constructed? Autonomous vehicles are being developed to mitigate the accidents that are 90% of the time caused by human error. But what about the other 10%? We need to make sure that autonomous vehicles will react appropriately to a big tree branch that suddenly falls right in front of it on the road or a sudden earthquake that makes the ground tremble. Safety is the number one priority of all automakers alike and it will be important to help users trust autonomous vehicles. We need to embed ethics into algorithms and the engineering design itself to achieve this.
Data Ownership and Privacy
Autonomous vehicles are literally data-generation engines with up to 4 terabytes per day, per vehicle of data collected and analyzed either on- or off-board. Apart from just capturing personal information and location data, the autonomous vehicle will be able to analyze and perceive user behavior among other things. The vehicle will know what the user’s purchasing habits are along with location preferences. This data will be extremely useful not only to automakers but to advertisers who want to know which people are interested in their products. Though automakers say they collect customer data only with explicit permission, the actual permission is often buried in lengthy service agreements which are hardly ever written in its full extent. Consumers of voice recognition technology have already started raising public concerns about the collection and transmission of private communications. While notice and consent from the customers are the basic necessities of data privacy, data from autonomous vehicles should also be anonymized. Additional steps should be taken to ensure that the data cannot be re-identified, taking into account technological developments and regulatory guidance. Under a “privacy by design” approach endorsed by the Federal Trade Commission, companies are encouraged to conduct a privacy or security risk assessment, minimize the data that is collected and retained, and test security measures before a product is launched.
The cause of concern for cybersecurity has been two-fold, most of which were injected through software security breaches while some were also related to hardware security lapses. In 2017, the world battled a cyber-attack every 39 seconds due to inefficient software security while the recent chip security flaws like Meltdown and Spectre rattled the industry with new fears. As an example of the financial impact, global ransomware damage costs increased 15X in two years from $325 million in 2015 to almost $5 billion in 2017 and are expected to rise up to $11.5 billion in 2019. The sophistication of technology and the ever-increasing degree of connectivity are opening up larger and larger attack surfaces to be maliciously exploited through cyber-attacks in many forms. When we talk about a world wherein there are thousands of connected autonomous vehicles on the road each having hundreds of subsystems, we are also talking about the potential risk of systems being hacked to endanger a large number of humans in the process. How well prepared are we to deal with that level of risk and to make sure that The Fate of the Furious does not go down in history as anything but number eight of an action-packed movie franchise?
New & Innovative Business Models
We will soon be having autonomous ride-hailing services and autonomous delivery services. New business models will be required to cater to the multi-modal transportation systems that will include both public transport and ride-hailing services. Mobility cloud services will be needed to cater to users’ needs whether it is safety or entertainment. A number of other requirements will come up with autonomous vehicles. Strategic partnerships are shaping up to repair and maintain autonomous vehicles, for example, Waymo’s partnership with Avis. But a couple of questions still remain. Where will these vehicles be parked when they are not in use? Will these vehicles increase traffic congestion even with platooning? It is going to be some time before people completely trust in autonomous vehicles and be able to afford them. In that duration, both autonomous and conventional vehicles will be driven on the road together. New regulations will be required for the operation and safety of autonomous vehicles. Usage-based insurance models are already replacing traditional automotive insurance models and automakers need to pursue technology and ethics in the framework of being technically feasible, ethically justifiable and legally defensible. It is also clear that the success of autonomous cars lies in the strategic partnerships between tech and automotive companies. While Daimler and Bosch have struck a partnership to launch self-driving fleets by 2023, Nvidia and Continental team up to create the complete artificial intelligence self-driving solution. To thrive in this rapidly evolving automotive industry, companies need to build an ecosystem of strategic partners and collaborate together to work on important issues such as cybersecurity and data privacy.
These are some important challenges that we need to think about as we bring autonomous vehicles to the roads for public use and try to sustain the transition of the automotive industry from conventional to fully automated vehicles. We need to make the autonomous vehicle not only safe but a great experience for the user.
As the auto industry is changed by technological and economic currents, OEMs and Tier-1 manufacturers will need to partner with technological specialists to thrive in the era of the software defined car. Movimento’s expertise is rooted in our background as an automotive company. This has allowed us to create the technological platform that underpins the future of the software driven and self-driven car. Connect with us today to learn more about how we can work together.