CMU's Autonomous Car Doesn't Look like a Robot

This self-driving Cadillac uses seemingly invisible, automotive-grade sensors

2 min read
CMU's Autonomous Car Doesn't Look like a Robot

The future of automobile autonomy isn't going to involve cars covered in cameras and radar and lasers. It's going to be all invisible, and CMU is already there.

The 2011 Cadillac SRX in the picture above is an autonomous car. Carnegie Mellon University had it drive itself 33 miles last week on public roads, from from Cranberry, Pa. to Pittsburgh International Airport. At first glance, you probably wouldn't be able to tell that the car is self-driving, because self-driving cars looked like this just five years ago:

That's CMU's BOSS competing in the DARPA Urban Challenge in 2007, with who knows how many sensors mounted all over it. And even Google's autonomous cars have that signature Velodyne LIDAR mounted on top of them:

By contrast, CMU's SRX relies entirely on automotive-grade radars, lidars, and cameras. You can see them if you look closely in the picture at the top of this article (there's one above the windshield, for example), but you do have to look closely. Inside, there are some extra buttons and screens, but all of the computers are stuffed under the floor in the trunk. And despite the lack of giant and complicated and expensive sensor systems, the car is still able to achieve the level of autonomy that we all want it to, as CMU's Raj Rajkumar explains:

"This car is the holy grail of autonomous driving because it can do it all — from changing lanes on highways, driving in congested suburban traffic and navigating traffic lights."

In addition to controlling the steering, speed and braking, the autonomous systems also detect and avoid obstacles in the road, including traffic cones and barrels, as well as pedestrians and bicyclists, pausing until they are safely out of the way. The systems provide audible warnings of obstacles and communicate vehicle status to its passengers using a human-like voice.

It's unfortunate that while the technology for all of this is arguably mostly ready, society (socially and legally) just isn't yet. You can buy cars with adaptive cruise control and lane departure warnings, which could hypothetically let the car drive itself, at least under some specific circumstances. And despite the fact that even a bad autonomous (or semi-autonomous) car would still save lives overall, there's no legal infrastructure in place to make it possible for manufacturers to implement such technology without undue risk of being sued into oblivion the first time something goes wrong.

Via [ CMU ]

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

“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.

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

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