Stanford’s Self-Driving, Drifting DeLorean Is Every Bit as Good as Doc Brown’s Flying Car

Today, as we’re sure you’re aware, is 21 October, 2015, the very day that Doc Brown and Marty McFly visit in Back to the Future II. As a consequence, we’ve seen plenty of news in the past week related to the hit 1989 film, including the launch of a special limited-edition version of the Pepsi Perfect soda featured in the film and the debut of those famous self-tying Nike Mag boots.

It doesn't have a flux capacitor, but this sucker can drift. By itself.

It doesn’t have a flux capacitor, but this sucker can drift. By itself.

But while Japanese automaker Toyota might be using its own Back to the Future connections to help launch the 2016 Toyota Mirai fuel cell sedan, there’s one thing sorely missing from the reality of 2015: hover cars. That’s because the technology needed to build such a vehicle — or let Goldie Wilson III’s Hover Conversion company retrofit any old road car into a skyway flier for $39,999.95 a reality — hasn’t yet been invented.

Luckily, some clever folks at Stanford University have come up with what we think is the next-best thing for Back to the Future day: a fully-electric, self-driving DeLorean that can even pull it’s own stunts.

Meet the Multiple Actuator Research Test bed for Yaw control, or MARTY for short.

The Brainchild of Chris Gerdes and his students, MARTY adds BTTF Style to proceedings.

The Brainchild of Chris Gerdes and his students, MARTY adds BTTF Style to proceedings.

The brainchild of Stanford professor of mechanical engineering Chris Gerdes and his students, who decided to transform a standard 1981 DeLorean DMC-12 into a research platform for discovering just where the physical limitations of autonomous driving lay. Working alongside the REVS program at Stanford and Silicon Valley electric car startup Renovo Motors, the team replaced the original gasoline engine in the DMC-12 with an all-electric drivetrain, as well as the necessary actuators and control mechanisms needed to give MARTY self-driving capabilities.

Unlike some of the other autonomous vehicles we’ve featured on Transport Evolved however, MARTY hasn’t been built primarily as an autonomous vehicle intended to take you to the shops and back without requiring you to touch the wheel. Instead, its purpose is one designed primarily to investigate just where the limitations of vehicular control systems lay. Or to put it another way, to see just where the line between safe car control and the laws of physics teaching you a lesson lay.

“We want to design automated vehicles that can take any action necessary to avoid an accident,” explains Gerdes. “The laws of physics will limit what the car can do, but we think the software should be capable of any possible maneuver within those limits. MARTY is another step in this direction, thanks to the passion and hard work of our students. Stanford builds great research by building great researchers.”

As Gerdes explains, the Electronic Stability Control (ESC) systems used in most modern cars are designed to ensure that the car never strays beyond the bounds of safe handling by taking over braking or power distribution to prevent the driver from losing all control of the vehicle. It helps to avoid the kind of kneejerk reactions which in a world before ESC often lead to an accident being far worse than perhaps it might have been had the person behind the wheel not panicked.

But ESC systems don’t get close to the limitations of what a car can actually do, as many generations of professional rally drivers and stunt professionals prove when they execute the kind of crazy maneuvers that would give most of us a cardiac arrest.  Instead, they focus on delivering a safe and comfortable ride.

“In our work developing autonomous driving algorithms, we’ve found that sometimes you need to sacrifice stability to turn sharply and avoid accidents,” said Gerdes. “The very best rally car drivers to this all the time, sacrificing stability so that they can use all of the car’s capabilities to avoid obstacles and negotiate tight turns at speed. Their confidence in their ability to control the car opens up new possibilities for the car’s motion.”

Explaining that current control systems, designed to help a human keep control of their car, aren’t designed with that in mind, Gerdes says autonomous cars are capable of far more precision and far quicker reactions than your average human driver. As a consequence, automated cars should be able to perform the kind of maneuvers that stunt and rally drivers perform on a daily basis in an attempt to reduce the number of accidents and keep their occupants safe (if a little stunned when the time for extreme maneuvering comes).

“We think that it is important to open up this design space to develop fully automated cars that are as safe as possible,” he continued.

And that, in a round about way, is how Stanford taught MARTY to drift. Thanks to Renovo’s brand-new drivetrain — given to the Stanford team on early-access — MARTY is powered by the same Twin Sequential Axial Flux motors and single-gear, on-motor gearboxes found in the Renovo Coupe. Capable of delivering around 4,000 pound -feet of torque to the rear wheels with each wheel being fully controllable in a fraction of the second, MARTY’s on-board software can blip the throttle like professional drifter Leona chin. Add in software designed to control the steering with millisecond accuracy, and MARTY can drift from dawn until dusk — or at least until its high-capacity battery pack runs flat.

It's all about understanding the limitations of a very cool way.

It’s all about understanding the limitations of physics…in a very cool way.

Why teach an autonomous car to drift? It’s all about better understanding the careful purposeful loss of traction that enables humans to drift, rally and autocross.

“When you watch a pro driver drift a car, you think to yourself that this person really knows how to precisely control the path and angle of the car, despite how different it is from normal driving,” said Jonathan Goh, a mechanical engineering grad student who has been key to the project’s successes thus far. “The wheels are pointed to the left even though the car is turning right, and you have to very quickly coordinate the throttle and steering in order to keep the car from spinning out or going the wrong way. Autonomous cars need to learn from this in order to truly be as good as the best drivers out there.”

So far, MARTY has learned to use the drifting techniques it has learned to negotiate the kind of tight turns and obstacle avoidance that many human (and autonomous) drivers would simply fail to clear. Even after a short period of development, MARTY’s current party trick — locking itself into a continuous precisely-controlled doughnut at a large drift angle — is impressive to watch.

“The sublime awesomeness of riding in a DeLorean that does perfect, smoke-filled doughnuts by itself is a mind-bending experience that helps you appreciate that we really are living in the future,” joked Goh.

Ultimately, the skills taught to MARTY will help Stanford develop algorithms that can be used in the real world to further enhance the capabilities and safety credentials of autonomous vehicles. But for now, there’s one more thing Gerdes would like to see MARTY accomplish: taking part in a full-blown drift competition alongside another car driven by a professional driver.

We're pretty sure Tesla could make its cars drift, too...

We’re pretty sure Tesla could make its cars drift, too…

In order for that to happen, MARTY will not only need to anticipate and follow the moves of the human drifting alongside it to avoid crashing, but also to try and out drift the competition, something that the Stanford team are eager to do.

“While we aren’t picturing a future where every cloud produces clouds of white tire smoke during the daily commute, we do want automated vehicles that can decipher the subtle cues drivers give when driving and incorporate this feedback when planning motion,” said Gerdes. “Drifting is a way to study these larger questions, with style.”

Whatever the outcome however, we’ve got one thing to say.

“Great Scott! A self-driving electric DeLorean? Now why didn’t I think of that?”


Want to keep up with the latest news in evolving transport? Don’t forget to follow Transport Evolved on Twitter, like us on Facebook and G+, and subscribe to our YouTube channel.

You can also support us directly as a monthly supporting member by visiting

Related News