Automotive

Watch Stanford's self-drifting DeLorean spin around an entire circuit

Watch Stanford's self-drifting DeLorean spin around an entire circuit
Stanford's self-driving DeLorean in action
Stanford's self-driving DeLorean in action
View 3 Images
Stanford's self-driving DeLorean goes by the name of Marty (Multiple Actuator Research Test bed for Yaw)
1/3
Stanford's self-driving DeLorean goes by the name of Marty (Multiple Actuator Research Test bed for Yaw)
Stanford's self-driving DeLorean in action
2/3
Stanford's self-driving DeLorean in action
When engineers at Stanford University first unveiled their autonomous drifting DeLorean back in 2015, they did so with plenty of future upgrades in mind
3/3
When engineers at Stanford University first unveiled their autonomous drifting DeLorean back in 2015, they did so with plenty of future upgrades in mind
View gallery - 3 images

When engineers at Stanford University first unveiled their autonomous drifting DeLorean back in 2015, they did so with plenty of future upgrades in mind. After a few years of tinkering, they have now equipped the car with some serious driving chops, sending it sideways around a kilometer-long drifting course all on its own.

The car goes by the name of Marty (Multiple Actuator Research Test bed for Yaw) and its likeness to the iconic DeLorean from Back to Future basically ends at the exterior. Inside, the team has packed electric motors delivering 7,000 Nm of torque to each of the rear wheels and computer-controlled braking, throttling and steering systems designed to send the car traveling predominantly in a sideways orientation.

Among the upgrades is a new set of GPS antennas on the roof that track the car’s position to within an inch, while the mechanical controls for steering, braking and throttling have been switched over to electronic systems. The suspension has also been overhauled, to make it less “squishy” and more effective at holding a decent drift.

Once the team had fed Marty the course layout, which consisted of a kilometer-long (0.62 mi) circuit made up of traffic cones at California’s Thunderhill Raceway, the onboard computers needed just a few seconds to calculate the optimal route. This involved adjusting the braking, steering and throttling to have the car wind itself around the course mostly while skidding sideways with a billow of smoke trailing behind. The precision of all this, and how close it comes to the traffic cones without taking them any of them out, is a sight to behold.

Stanford's self-driving DeLorean goes by the name of Marty (Multiple Actuator Research Test bed for Yaw)
Stanford's self-driving DeLorean goes by the name of Marty (Multiple Actuator Research Test bed for Yaw)

“The results so far are rather outstanding,” says mechanical engineer Chris Gerdes. “The stability control systems of modern cars limit the driver’s control to a very narrow range of the car’s potential. With Marty we have been able to more broadly define the range of conditions in which we can safely operate, and we have the ability to stabilize the car in these unstable conditions.”

The team says that Marty’s performance was on par with a human drift driver, and it is hard to argue after watching the video. But there’s more at play than some stunt work here. The end game for the research team is to advance the safety of autonomous driving systems by enabling them to handle tougher conditions and quickly take evasive action, such as avoiding a pedestrian that runs on to the road.

When engineers at Stanford University first unveiled their autonomous drifting DeLorean back in 2015, they did so with plenty of future upgrades in mind
When engineers at Stanford University first unveiled their autonomous drifting DeLorean back in 2015, they did so with plenty of future upgrades in mind

“We’re trying to develop automated vehicles that can handle emergency maneuvers or slippery surfaces like ice or snow,” Gerdes said. “We’d like to develop automated vehicles that can use all of the friction between the tire and the road to get the car out of harm’s way. We want the car to be able to avoid any accident that’s avoidable within the laws of physics.”

You can see Marty in action in the video below.

Beyond the Limits: MARTYkhana

Source: Stanford University

View gallery - 3 images
2 comments
2 comments
vince
Astounding. Amazing. Kudo's !!
MQ
Awesome display of "closed loop" Machine control. (In more wayze than 1)

PS.. when you have a surplus of Torque, most of the input need only be "throttle control" ie. one-dimensional". "Lock-to lock time" - and not enough torque at the relevant speed ) is where many human driver/car - combinations fail to maintain optimal control (too-many turns of the wheel to get it done quickly enough,, not enough torque to "float the rear end") - NB. drifting is by no-means an efficient use of available energy, though it can improve manoeuvrability (Degrees per second of yaw authority) significantly.

Chuck in "bags of torque" and a quartic equivalent of the course, an Arduino could "control the path".. (stating that multiple GPS sensors were necessary adds the illusion of "required" high-tech. [A simple magnetometer - with adequate regional correction] could provide the "pointing-control necessary" to pull this off.). Maybe it is time to get rid of the steering wheel in cars and use speed proportional joysticks... Ok! we're not fit for that, we will have to be satisfied as passengers of the "Machine". joy...

remember when all steering and braking was mandatory to be 100% mechanical, then Lexus, Mercedes and Tesla came in and ignored the regulation so change the status Quo. Time for mass civil disobedience by all manufacturers.. Show the consumers what is possible.