Stanford's Autonomous Racer Hits 120 MPH on the Track

Shelley posts track times up there with the best human drivers, and she's only going to get faster

2 min read
Stanford's Autonomous Racer Hits 120 MPH on the Track

We first met Stanford's robotic Audi TTS back in 2009, and while it's managed to climb Pike's Peak all by itself, it didn't do so in a time that was all that impressive. More recently, Stanford let Shelley off the leash at California's Thunderhill Raceway, where it almost (almost!) destroyed all human drivers with some aggressive laps around the track.

Unlike those Google autonomous cars, Shelley can't navigate in traffic or in unfamiliar environments. She just uses GPS to get where she's going, and she'll plow right through anything that gets in her way, which is why she's restricted to racetracks. Her talents lie in her ability to sense the limits of her own performance, which allows her to drive right on the edge of what's physically possible.

On Thunderhill, Shelley clocked times that were close to (but not better than) professional human drivers, probably because the exactly line that the car calculated and followed proved to be slightly less efficient than a "smoother," optimized path that a human might choose. To fix this, Stanford is sending professional drivers around racetracks while hooked up to brainwave sensors to see if they can figure out where on the track humans have to concentrate and think ahead.

The point of all of this, besides a concerted effort on the part of Stanford to render NASCAR obsolete, is to teach your car how to drive on the edge of control. Potentially, this could save your life: since it's probably safe to assume that you don't have the experience of a professional race car driver, if your car does, it can take over if necessary (like, in the event of an impending accident) and successfully steer you to safety.

Via [ Stanford ]

The Conversation (0)

How the U.S. Army Is Turning Robots Into Team Players

Engineers battle the limits of deep learning for battlefield bots

11 min read
Robot with threads near a fallen branch

RoMan, the Army Research Laboratory's robotic manipulator, considers the best way to grasp and move a tree branch at the Adelphi Laboratory Center, in Maryland.

Evan Ackerman
LightGreen

This article is part of our special report on AI, “The Great AI Reckoning.

"I should probably not be standing this close," I think to myself, as the robot slowly approaches a large tree branch on the floor in front of me. It's not the size of the branch that makes me nervous—it's that the robot is operating autonomously, and that while I know what it's supposed to do, I'm not entirely sure what it will do. If everything works the way the roboticists at the U.S. Army Research Laboratory (ARL) in Adelphi, Md., expect, the robot will identify the branch, grasp it, and drag it out of the way. These folks know what they're doing, but I've spent enough time around robots that I take a small step backwards anyway.

The robot, named RoMan, for Robotic Manipulator, is about the size of a large lawn mower, with a tracked base that helps it handle most kinds of terrain. At the front, it has a squat torso equipped with cameras and depth sensors, as well as a pair of arms that were harvested from a prototype disaster-response robot originally developed at NASA's Jet Propulsion Laboratory for a DARPA robotics competition. RoMan's job today is roadway clearing, a multistep task that ARL wants the robot to complete as autonomously as possible. Instead of instructing the robot to grasp specific objects in specific ways and move them to specific places, the operators tell RoMan to "go clear a path." It's then up to the robot to make all the decisions necessary to achieve that objective.

Keep Reading ↓ Show less