Driverless cars: the future of the automotive industry

Picture the scene: you step into your car, relax into a comfortable reclining chair and set your destination. As the car whisks you away you fall asleep to the sound of smooth jazz, arriving at your journey’s end relaxed and refreshed. This is a dream shared across the globe; but is it achievable, and how soon will these autonomous cars be available?

The advances made to date
The simultaneous development of a range of technologies has brought about the opportunity of driverless vehicles. For example, some current production vehicles now feature adaptive cruise control and lane keeping technologies which allow the automated control of acceleration, braking and steering for periods of time on motorways and in congested traffic. Advanced emergency braking systems automatically apply the brakes to help drivers avoid a collision. Self-parking systems allow a vehicle to parallel or reverse park completely hands free [1]. These developments have helped the industry to move closer to the goal of developing a vehicle, which is ultimately “driverless”.

The current state of driverless technology
The current state of driverless vehicle technology goes far beyond what is available on the market to the average consumer today. Research and development is being conducted actively by a number of different companies, all racing to perfect the technology and be the first to bring it to market. There are already many driverless vehicle systems in place today; however, these operate primarily in closed systems, such as on rails or within a particular facility. For example, many train networks, like the Docklands Light Railway in London [2], operate autonomously and have been a reliable and efficient use of the technology for several decades. These systems are strictly limited to rails and their autonomy is simple, where they travel from one station to another, sticking to a timetable. More complex systems, such as autonomous pallet handling equipment in the warehousing sector and robotic port cranes like those used at Maasvlakte II in the Port of Rotterdam [2], rely on vehicles that must know their position in two-dimensional space, as well as a system to distribute instructions, map routes and prevent collisions. Crucially, these systems are autonomous, in that they do not require human input to perform the majority of their functions, but are not independent: they rely on some central control and coordination. In order for a vehicle to be truly autonomous, it must operate entirely independently, given some simple user-defined instruction. This requires a vehicle to see and understand its environment and apply complex and uniform logic in order to achieve that initial instruction. Here is where those companies are applying themselves to developing a reliable solution. Arguably the most well-known example of this development is the Google Self-Driving Car Project [3] (see figure 1), however, an array of other systems are being researched, including Mercedes-Benz’s Intelligent Drive [4] and Audi’s Piloted Driving [5] systems.

Figure 1 - Google's Self-Driving Car Project in testing

What is the industry aiming for?
There are many advantages, both commercial and technological, related to the development of driverless vehicle technology. It will make driving easier, improve road safety, reduce emissions and ease congestion [1]. Driverless vehicles could create more free time for drivers, who would not need to concentrate on driving for the entire duration of a journey. With this technology, the driver would be able to choose whether they want to be in control of the vehicle or hand the task of driving over to the vehicle itself.

In terms of road safety, human error is involved in as many as 90% of all vehicle accidents. The two most common human-related factors that contribute to heavy truck accidents are failure to look properly and failure to judge another person’s path or speed [6]. Automated vehicles will have the ability to eliminate these mistakes, using a range of sensors, which will constantly monitor their surroundings. Current technologies, such as Anti-lock Braking Systems (ABS), cruise control or parking sensors have already reduced, or eliminated, driver input in some aspects of driving. Therefore, fully automated vehicles are a natural progression from today’s automated safety technologies.

By communicating with their environment and other vehicles, driverless vehicles would allow the reduction of congestion and provide more consistent journey times. This could be achieved through the use of “connected vehicle” technologies. “Connected vehicles” would communicate with each other and their surroundings in order to identify the optimum route, as well as communicate with roadside infrastructure such as traffic lights to minimise fuel consumption and emissions.

In the past decade, vehicles have integrated various technologies designed to make cars safer, more responsive, and more efficient. From the hands-free mobile phone, to satellite radio, to automated parking, the vehicle is undergoing an electronic overhaul that promises to transform its role in the market [7]. This means that in addition to being used for personal transportation, a vehicle is fast evolving into a new mobile device, merging with the digital world into an all-encompassing communications environment. This may eventually mean that vehicle manufacturers would need to consider how to merge advanced technologies into their traditional automotive systems.

What are the current difficulties for the industry?
In order to achieve an autonomous driving system that is both reliable and safe, there are many challenges to overcome. Perhaps the foremost difficulty is to ensure that the vehicle is aware of its surroundings and can interpret and differentiate between visual cues [8]. An autonomous car cannot rely on being told information about its environment, but must understand it in the same way that a human driver would. This is hoped to be achieved through the use of cameras and LiDAR, which uses laser light to survey an area in the same way that RADAR uses radio waves. On-road testing of autonomous vehicles is already underway in several countries, such as the USA and Germany, but these tests require detailed and extensive mapping prior to the journey in order for the car to interpret what it senses through cameras. If a temporary set of traffic lights were to be installed, and this information not relayed to the computer, the vehicle would not be aware that it would have to stop. The ability of a vehicle to detect an oncoming collision and take preventative measures, such as swerving or emergency stopping, is already tested and available on the market. However, if, for example, a vehicle sees a plastic bag billowing in the wind on the middle of the road, but interprets it as a rock in its path, it may take sudden action that could be dangerous to other road users. Figure 2 shows a representation of the information that Google’s autonomous car can sense.

Figure 2 - Google's autonomous car sensing its environment [9]

Issues with the concept
Aside from the operating challenges of the technology, there are other issues that must be overcome before these systems reach the mainstream. For example, how will the technology prevent malicious software from overriding the controls and endangering the occupants, or even stealing the vehicle itself? There will also be legal hurdles to overcome, since responsibility for any wrongdoing may no longer lie with a human driver, but with the company who developed the software.

Despite the potential advantages that this technology would bring, the weight of the challenges faced by the industry means that autonomous vehicles are a long way from arriving on our streets. Most car manufacturers claim that they expect to sell vehicles that can drive themselves at least part of the time by 2020 [10], however, expert members of the Institute of Electrical and Electronics Engineers predict that we will have to wait until at least 2040 before a majority of vehicles are autonomous [11]. So perhaps we have a long time to wait before we can safely fall asleep at the wheel, however, with technology constantly advancing, driving will become a more relaxing experience with every year that goes by.

Works cited
“The Pathway to Driverless Cars,” Department for Transport, 2015.
“A striking argument for autonomous train drivers?,” The Engineer, August 2015. [Online]. Available: [Accessed April 2016].
“Robots running things in Rotterdam,” Maersk, February 2015. [Online]. Available: [Accessed April 2016].
“Google Self-Driving Car Project,” [Online]. Available: [Accessed April 2016].
“Mercedes-Benz's autonomous driving features dominate the industry -- and will for years,” Automotive News, August 2014. [Online]. Available: [Accessed April 2016].
“Audi Piloted Driving,” [Online]. Available: [Accessed April 2016].
“European Accident Research and Safety Report 2013,” Volvo Trucks, [Online]. Available: [Accessed April 2016].
“The Connected Vehicle: Viewing the Road Ahead,” Accenture Digital.
“Hidden Obstacles for Google’s Self-Driving Cars,” MIT Technology Review, August 2014. [Online]. Available: [Accessed April 2016].
“Institute of Electrical and Electronics Engineers,” [Online]. Available: [Accessed April 2016].
“Automakers say self-driving cars are on the horizon,” Tampa Bay Times, March 2014. [Online]. Available: [Accessed April 2016].
“Look Ma, No Hands!,” Institute of Electrical and Electronics Engineers, 2012. [Online]. Available: [Accessed April 2016].

For more information, please contact Dr Tariq Ahmad, Laurence Kingdon or Stefanos Stefanidis.