Imagine a world where you can sift through your social media or sip on the morning coffee while your car motors towards the destination. Or maybe you are a LiDAR enthusiast like me and are “driving” to work while reading one of the Blickfeld blogs on LiDARs and its unique applications. It might be only possible in simulations for now, but level 5 self-driving cars promise to make this a reality in not so distant future.
Self-driving cars, or Autonomous Vehicles (AVs), have captivated everyone’s imagination for years now. While ADAS (Advanced driver-assistance systems) has already automated our vehicles to a great extent, the prospect of level 5 autonomy making our automobiles completely self-driven is indeed exciting!
This enthusiasm around the idea was captured by a recent McKinsey and World Economic Forum global survey, which noticed a marked increase in consumer interest in autonomous vehicles. In North America alone, the willingness to take deliveries from AVs increased from 18% to 28% during the pandemic, as did the consumer’s readiness to use them for people transport.
Although these trends point towards a fascinating future, there is another side of the picture that begs our attention.
Transportation and the Environment
The transport sector is heavily reliant on fossil fuels, accounting for up to 45% of global oil demand according to a study by Morgan Stanley. And the U.S. Environmental Protection Agency notes that 90% of the fuel used for U.S. transportation is petroleum-based. Consequently, transportation generates the largest share of greenhouse gas emissions in the US, amounting to 29% of the total emissions. A similar trend can be seen in Europe, with a significant percentage of the fuel for transportation comprising of gasoline and diesel, leading to one-quarter of greenhouse gas (GHG) emissions. This is in addition to the generation of other sources of air pollution such as particulate matter (PM) and nitrogen dioxide (NO2).
Similarly, European Environment Agency reported emissions from new passenger cars increasing from 2017 to 2019, reaching 122.3 g CO2/km. Although this is below the 2015-2019 emissions target of 130 g CO2/km, it clearly surpasses the 2020-2024 target threshold of 95 g CO2/km.
With these trends in mind, it is imperative to expand the discussion from just safety and technological aspects of self-driving cars and also include the effect of transportation on the environment and the increasing share of AVs in the transportation mix. This blog will dissect the technological, economic, and behavioral aspects impacting the relationship between AVs and the environment.
Factors Affecting Self-Driving Cars and the Environment
Autonomous Vehicles and Energy
AVs offer significant environmental benefits in fuel usage, as noted by the Southwest Research Institute study claiming that AVs can lead to as much as 20% improvement in fuel consumption. The benefits of reduced consumption can be augmented with the choice of a suitable source of energy for the production and operation of AVs.
Shifting towards electricity as the primary source of powering AVs is key in guiding net reduction in GHG emissions, leading to the unique model of A-EVs (Autonomous Electric Vehicles).
A study by ICCT (The International Council on Clean Transportation) concluded that using electricity as the engine power source could reduce GHG emissions in the range of 28% to 72%.
This would require a faster adoption of Electric Vehicles (only 2.5% of total sales globally in 2019), and improvements in battery technology, charging infrastructure, and cost-competitiveness, as well as supportive government policies.
Besides shifting to electricity, changing the primary source of power in the charging grid to renewable or carbon-neutral sources must also be prioritized. EVs charged on a grid running on fossil fuels put a significant dent into any potential reduction in the carbon footprint promised by EVs and future A-EVs.
Challenges for Autonomous-Electric Vehicles
The biggest concern about AVs is their high energy requirements for powering the heavy computational needs and a large number of devices, as well as the air drag created by the large and protruding sensors. Experts have been apprehensive about how these power needs might be too large for any modern-day electric vehicle technologies and would raise the vehicle’s energy demands beyond what’s technically possible for an electric vehicle, as discussed by this research at the University of Michigan.
One solution is to use larger batteries, but that also adds additional weight and therefore reduces fuel efficiency. Moving from Autonomous Electric Vehicles (A-EVs) to a hybrid engine could be another, but that again sounds like a step back.
On the other hand, there are studies, such as this research by Carnegie Mellon University in Nature Energy, claiming that electricity alone should be enough to power an autonomous vehicle without a significant decrease in range only if energy-efficient computing and aerodynamic sensor stacks are implemented.
Self-Driving Cars and Traffic Congestion
According to a study by the European Commission, drivers in some countries can lose up to as many as 45 hours in traffic congestion annually, cutting into hours where people can be productive or find time for leisure while racking up the fuel costs and vehicle emissions.
Fast forward to a future where all cars are fully autonomous, AVs could play an essential role in reducing traffic congestion.
For instance, if ride-hailing services were to adopt AVs, it could truly disrupt the industry and reduce the traffic density by several folds. According to a report by McKinsey, the combination of ride and carsharing could “clear four out of five cars from the road.” And a study conducted at the University of California estimated that one shared vehicle alone could remove 9 – 13 conventional vehicles from the road.
Motivated by unprecedented profits, many companies are already in hot pursuit of the idea. For example, last October, the Alphabet-backed driverless technology startup, Waymo launched a fully driverless commercial taxi service called Waymo One in the Greater Phoenix area. Before the launch, the company had already tested its technology in 6 U.S. states and 25 cities. Similarly, ride-hailing company Uber had invested more than $1 billion in developing driverless cabs for their Advanced Technologies Group (ATG) and even conducted test runs in San Francisco and Pittsburgh. ATG has since then been sold to another self-driving startup, Aurora, with the company itself valued at $10 billion.
Adoption of AVs would also imply that ride-hailing vehicles can remain on roads at all times. This would allow carpooling at all hours of the day and eliminate any unoccupied time, which is almost 90% of a vehicles’ lifetime while reducing the overall size of the vehicle fleet.
Self-driving vehicles could also be designed efficiently to be more spacious and accommodate more people than in a traditional vehicle by cutting out the driver and reducing the number of cars on the road.
But what does traffic congestion have to do with carbon emissions?
Reducing traffic congestion directly impacts fuel consumption and carbon emissions since both of these factors are strongly influenced by the vehicle’s speed. As a general trend, they are high at low speeds, such as in a traffic jam, flatten out at average speeds, and then rise again when vehicles move at higher speeds. Autonomous vehicles can reduce traffic congestion, allow vehicles to move at constant speeds, and thus eliminate the stop-and-go driving behavior to improve efficiency.
The automated acceleration and braking by AVs, also known as eco-driving, can also reduce fuel consumption by 20% in aggressive drivers and 15% for normal drivers, as noted in a study by the University of Maryland, thus cutting down emissions.
And courtesy of the increased safety features of AVs, crashes are less likely to occur, which allows for lighter and smaller cars, decreasing fuel consumption between 5-23%.
Autonomous Vehicles and Total Number of Miles Driven
The convenience promised by autonomous vehicles could act as a double-edged sword when considering Vehicle Miles Travelled (VMT).
As traveling from point A to B gets easier with the automation of driving and lowered opportunity costs, people might opt for commutes spanning longer distances. AVs also promise to enable better transport accessibility for the population segment who cannot drive, such as the elderly, disabled, and children, leading to increased total distance traveled.
At level 5 autonomy, the phenomenon of ghost AVs can also add to the total number of miles, as empty cars could be driving around to pick up passengers or be deployed by logistics companies for goods deliveries.
On the other hand, on-demand mobility and carsharing, combined with autonomous technology, can significantly reduce the GHG (Green House Gas) by combining trips that share the same time and destination while lowering the VMT.
Fully autonomous vehicles can also intelligently interact with each other and the road infrastructure while utilizing the road smartly through better route calculation and efficient road occupancy to reduce fuel consumption and cut down VMT.
For instance, platooning could improve the road occupancy efficiency, where a train of detached vehicles interact and travel closely together to reduce the energy consumption from 3% to 25%, depending on the number of vehicles, their separation, and characteristics.
Always driving more efficiently would allow people to arrive at their destinations quickly and more sustainably. Although not directly affecting VMT, AVs’ ability to smartly interact with the surroundings can help in this regard. As explained by the engineers at the Southwest Research Institute,
“AVs can connect with the roadway infrastructure such as traffic lights to “see” up to half a kilometer ahead and alert drivers to drive more efficiently. For example, if a stoplight is red a few blocks ahead, the app will tell the driver to speed up or slow down by a few miles per hour to avoid stopping. For heavy-duty trucks, making a full stop at a red light and re-engaging the engine when the light turns green is a significant source of fuel consumption.”
While technology is pivotal in deciding whether the total VMT increases or decreases, governmental policies will also play a key part. Pricing mechanisms to encourage car sharing while reducing ghost AVs could offset the risk factor of an increase in the net VMT. In essence, a combination of social, behavioral, and economic factors would ultimately dictate the effect on total VMT.
Self-Driving Cars and Costs
Some experts claim that the cost per mile of an autonomous vehicle would be significantly more than owning a traditional car, and the cost estimations vary from $1.58-$6.01 per mile to $0.55-$0.70 per mile. Recently, there have been very promising projections, such as market leaders Waymo quoting operational costs of an autonomous vehicle as low as 30 cents per mile. In comparison, traditional ride-hailing services cost around $2 to 3 per mile, meaning AVs are offering considerable savings on average.
As complementary technologies, such as LiDARs, become cheaper and easier to produce, manufacturing costs will also drop, making the technology more economically feasible and help in bringing the cost per mile even lower.
The affordability of AVs will be the deciding factor in determining the adoption, frequency, distance (long commutes versus short trips), and type (solo versus pooled) usage, and hence their impact on the environment.
Are self driving cars better for the environment?
While the effect of autonomous vehicles in reducing traffic congestion, and thus emissions, through efficient design, eco-driving, platooning, and carpooling is clear, other factors such as VMT, source of energy, and total costs are still up for debate.
Truly standing at a crossroads, we must opt for a carefully measured blend of technology, infrastructure adjustment, transportation models, policy development, and perception about the technology to ensure a future that is both autonomous mobility friendly and sustainable.