MIT researchers recently explored the potential energy consumption and related carbon emissions if autonomous vehicles (AV) are widely adopted.
The data centers that house the physical computing infrastructure used for running applications are widely known for their large carbon footprint. In fact, they currently account for about 0.3 percent of global greenhouse gas emissions — or about as much carbon as the country of Argentina produces annually, according to the International Energy Agency.
The researchers built a statistical model to study the problem and determined that 1 billion AV, each driving for an hour per day with a computer consuming 840 watts, would consume enough energy to generate about the same amount of emissions as data centers currently do.
In addition, they found that more than 90 percent of modeled scenarios, to keep AV emissions from zooming past current data center emissions, each vehicle must use less than 1.2 kilowatts of power for computing, which would require more efficient hardware.
“If we just keep the business-as-usual trends in decarbonization and the current rate of hardware efficiency improvements, it doesn’t seem like it is going to be enough to constrain the emissions from computing onboard autonomous vehicles,” said first author and grad student Soumya Sudhakar. “This has the potential to become an enormous problem. But if we get ahead of it, we could design more efficient AVs that have a smaller carbon footprint from the start.”
The team built a framework to explore the operational emissions from computers onboard a global fleet of fully autonomous electric vehicles. The model is a function of the number of vehicles in the global fleet, the power of each computer on each vehicle, the hours driven by each vehicle, and the carbon intensity of the electricity powering each computer.
“On its own, that looks like a deceptively simple equation,” said Sudhakar. “But each of those variables contains a lot of uncertainty because we are considering an emerging application that is not here yet.”
Research suggests that the amount of time driven in AVs might increase because people can multitask and the young and the elderly could drive more. However, other research suggests that time spent driving might decrease because algorithms could find quicker routes. In addition, the researchers needed to model advanced computing hardware and software that doesn’t yet exist.
So they modeled the workload of a popular algorithm for AVs and explored how much energy the deep neural network would consume if it were simultaneously processing many high-resolution inputs from many cameras with high frame rates. Sudhakar was surprised at how quickly the algorithms’ workload amassed.
If an AV has 10 deep neural networks processing images from 10 cameras, and that vehicle drives for one hour a day, it will make 21.6 million inferences each day. One billion vehicles would make 21.6 quadrillion inferences.
“After seeing the results, this makes a lot of sense, but it is not something that is on a lot of people’s radar,” said Associate Professor Sertac Karaman, Director, Laboratory for Information and Decision Systems. “These vehicles could actually be using a ton of computer power. They have a 360-degree view of the world, so while we have two eyes, they may have 20 eyes, looking all over the place and trying to understand all the things that are happening at the same time.”
To keep emissions in check, the researchers found that each AV needs to consume fewer than 1.2 kilowatts of computing energy. So, computing hardware must become more efficient — doubling in efficiency about every 1.1 years.
For more information, contact Abby Abazorius at