The future of Autonomous Vehicles

Autonomous vehicles, or self-driving cars, are vehicles equipped with advanced technology that allow them to navigate and operate without a driver or any other human intervention. These vehicles use a combination of sensors, cameras, radar, and artificial intelligence to perceive their environment and make real-time decisions.

Autonomous vehicles rely on a range of technologies to function. Key components include:

• Sensors collect data about a vehicle’s environment and surroundings. These include cameras, lidar, radar, and ultrasonic sensors that provide a 360-degree view of the vehicle's surroundings.
• Artificial Intelligence (AI) algorithms process the data from sensors to understand the environment, identify objects, and predict the behavior of other road users to make driving decisions.
• Machine Learning models help the vehicle improve its decision-making over time by learning from past experiences and data.
• Lastly, Control systems execute driving actions such as steering, braking and accelerating.

Autonomous vehicles are also categorised into several levels based on their capabilities:

• Level 1: Basic driver assistance such as cruise control and lane-keeping.
• Level 2: Partial automation, where the vehicle controls steering and speed, but the driver must remain attentive.
• Level 3: Conditional automation, where the vehicle can manage most driving tasks but requires human intervention when needed.
• Level 4: High automation, where the vehicle can handle all driving tasks in specific conditions without human input.
• Level 5: Full automation, where the vehicle can drive in all conditions without any human involvement.

Read on to learn more about self-driving cars, their advantages and disadvantages, and discover how Shell Eco-marathon is involved in furthering technology for Autonomous Vehicles.

The idea of self-driving cars has been around for a very long time. They have populated sci-fi visions of the future since the 1930s. But it is only recently that the technology that could make them a reality has begun to be fully developed.

The first experimental (and extremely slow-moving) prototypes date back as far as the 1960s.
More recently, in 2004, Darpa (a US defense department, responsible for the development of emerging technologies) invited the whole world to build a vehicle that could drive across California’s Mojave Desert to win million-dollar prize. The most successful vehicle only went seven of the 142 miles. But the race fueled a belief that the robot cars were a possibility.

In the follow-up race in 2005 five vehicles finished the course. By the 2007 Urban Challenge, the vehicles were not just avoiding obstacles and sticking to trails but following traffic laws, merging, parking, even making safe, legal U-turns.

And by 2010, Google technicians had built a system that could handle some of California’s toughest roads (including the famously winding block of San Francisco’s Lombard Street) with minimal human involvement.

Autonomous vehicles rely on radar and high-resolution cameras, plus a wealth of computing power, to detect and identify the objects on a street and ensure the car is instructed to avoid them. Some of the more basic technologies that these systems use are already in place on today’s vehicles: automatic braking systems, lane sensors and parking sensors, for example.

The most widely-discussed benefit would come from improved safety. Last year there were 1,770 reported road deaths in the UK alone, with over 26,000 people seriously injured. In the USA the figures are even more alarming, with 36,750 pedestrian and bicyclist deaths. With the majority of accidents being the result of human error, even a 90% adoption rate of driverless cars could translate into 22,000 lives being saved there every year.

Another advantage would be to open up transportation to people previously excluded: children, the disabled and the elderly could, in theory, travel without a driver, increasing accessibility.

AVs could significantly improve fuel efficiency thanks to algorithms for more efficient driving and the possibility of deploying lighter weight vehicles due to reduced collision risk. One projection estimates that a transition to electric, driverless taxis for example, could reduce emissions per mile by as much as 94% by 2030.

If rather than individual ownership, vehicles are owned by a municipality or business and deployed like taxis, this could reduce the number of cars needed in urban areas by 80%. There would be less need for parking places, freeing up that land for other uses, such as bike lanes, parks, or pedestrian areas. And fleet owners are also more likely than individual owners to transition to electric vehicles, given the long-term cost savings.

It depends who you talk to. Some manufacturers believe their new self-driving system will enable them to start offering fully autonomous driving in 2020, but some experts remain sceptical.

Essentially, about 80% of the technology needed to put self-driving cars into routine use is in place. The hardware, to start, is mostly there. Radars are already cheap and robust enough to build into mass-market cars. Same goes for cameras, and the artificial intelligence that turns their 2D images into something a computer can understand is making impressive strides.

But the remaining 20%, including developing software that can reliably anticipate what other drivers, pedestrians and cyclists are going to do, will be much more difficult to perfect.

Although much uncertainty remains, autonomous vehicles have the potential to radically change how people move around, the shape of our cities, the amount of carbon pollution in the air, and the nature of work, among other things. Now is the time for policymakers, businesses, technologists, and citizens to start exploring the different possibilities in detail—and charting a path towards a better future.

Aimed at creating a challenging venue for student engineers to gain hands-on experience and explore autonomous technology, Shell Eco-marathon holds two Autonomous Competitions each season. Learn more about the on-track and virtual events here.