Machine learning is already having a huge impact on the automotive business. Only a decade ago, the idea of driverless cars was limited to the realms of science-fiction – but they no longer belong to the future.
In July 2017, Audi became the first car maker to launch a production vehicle on the open market – the Audi A8 – with level three autonomy.
There are four levels of autonomy: levels one and two are assistance programmes that help the driver. Level one autonomy takes one task away (either acceleration or steering) while level two takes two tasks away. Level three, however, is a giant leap forward for the automotive industry.
The Audi A8 is completely independent when the car is in slow-moving traffic at speeds of up to 60 kilometres per hour. It allows acceleration, braking, steering and starting from a dead-stop – all without the driver needing to do anything. If they have the companion app, the driver doesn’t even need to be in the car.
The possibilities are huge, and they have many uses on a user experience level too. Imagine a scenario where you’re driving down a street and you see a parking space that doesn’t allow enough space for you to open the door. The Audi A8’s smart learning means you can hop out, tell the car to park itself via the app, then go about your day.
Audi has said that it expects to have fully self-driving cars (level four) on the market by 2020. Ford and Toyota have the same target while BMW has a target of 2021. Peugeot, through a subsidiary start-up called NuTonomy, claims it will have autonomous cars on the road in Singapore by the end of 2018. Tesla has similar plans.
Machine learning has given manufacturers the tools to make rapid progress in a relatively small amount of time. Manufacturers have spent years mining data by putting prototypes on the road and monitoring every decision the driver makes. If a driver approaches a low bridge and presses the brakes, that information is fed into the algorithm. Naturally, the volume of information going into these algorithms is huge.
Google, through its Waymo business, is one of the main players in the driverless car ecosystem. At the beginning of 2017, its test fleet had clocked up more than 636,868 miles on the roads of California alone.
According to a recent report in The Atlantic, Google’s machine learning programme is modelling eight million miles per day. In 2016, a total of 2.5 billion miles were modelled – and that was on top of the three million miles that its team of human drivers completed.
Uber, which is also highly invested in the world of driverless cars, has collected so much data that it already knows how people travel. Unlike traditional car manufacturers, Uber reckons that once driverless cars reach their full potential, people won’t want to own personal cars at all. Traditional manufacturers, however, favour a model where drivers will swap their old cars for the autonomous equivalent.
Andrew Salzberg, Head of Transportation Policy and Research at Uber said, “with no driver to pay, rides will be so abundant and cheap that people will not want to own a car at all. The number of cars on the road could fall by 90 percent. With electric cars, the cost of journeys will be negligible. A 20-mile journey could cost less than $4.”
Salzberg goes on to explain that this revolution isn’t just going to impact car manufacturers. It could revolutionise the physical world too.
“Some 20 to 30 percent of city centres are devoted to parking,” Salzberg explains. “If you don’t need parking, buildings can change, streets can change, homes can change. We can have more park spaces, instead of parking spaces.”
Regardless of whether Uber or the traditional car manufacturers are right, machine learning is already taking hold in the automotive manufacturing industry.
With the help of AI, automotive assembly lines are becoming more efficient, productive and cost-effective. The use of smart robotics on manufacturing floors has transformed vehicle production, making the manufacturing process increasingly automated.
The area of predictive maintenance, whereby sensors on machinery prevent costly downtime by forecasting when maintenance is needed, is already commonplace within the industry.
A study by the US Department of Energy claimed that a, “functional predictive maintenance program can reduce maintenance cost by 30 percent, reduce downtime by 45 percent, and eliminate breakdowns by as much as 75 percent.”
By all accounts, those numbers are more than conclusive for arguing the case of machine learning in the automotive business.
Between saving costs on driving and manufacturing, it’s clear that machine learning has had a huge impact on the automotive industry. If you think your industry is ripe for machine-learning inspired change, get in touch today.
We offer a range of custom services to suit your needs: advanced data analysis and modelling, custom algorithm creation and implementation with a particular focus on Predictive Maintenance.
Our advanced data science consultancy can team up with you to interpret your data and make your business work more efficiently so get in touch today to unleash your business’ potential.
Want to make sense of your data? Download our comprehensive guide: The Pedictive Maintenance Cookbook.