The July 2022 issue of IEEE Spectrum is here!

Close bar

Robotis and OSRF Announce TurtleBot 3: Smaller, Cheaper, and Modular

The newest version of the iconic ROS platform is designed to fit into your backpack

3 min read
Turtlebot 3
Photo: Evan Ackerman/IEEE Spectrum

Thousands of TurtleBots are out in the world right now, providing a (mostly) straightforward and (mostly) affordable way to get started with ROS. They’re (mostly) portable and (mostly) extendable, allowing you (with a limited amount of inconvenience) to modify the robot to keep up with your needs. TurtleBot 2 is a great platform (I certainly love mine), but its size and cost usually restrict it to people who already have some ROS experience, and know that a TurtleBot is something worth investing in. For people who want to get started with ROS but aren’t prepared to make as much of an investment, there just aren’t a lot of options with the same kind of community and support that you get with TurtleBot. 

At ROSCon this past weekend, the Open Source Robotics Foundation (OSRF) and South Korean robot maker ROBOTIS are tackling these problems by announcing a shiny new version of TurtleBot: TurtleBot 3. TB3 is small enough to fit into a backpack, and with a single-board computer instead of a netbook and just two Dynamixel motors driving a pair of wheels, it’s both simpler than previous TurtleBots and significantly cheaper. With tons of easy options for expandability (including sensors, computers, drive systems, and more) and the kind of software support that TurtleBots are known for, TB3 seems like the best intro to doing cool stuff with ROS yet. 

TurtleBot 3’s hardware is based around layered, 3D-printed plates that fit together like puzzle pieces. Or really, it’s just one plate that fits together with itself, and to make the robot bigger, you just get more plates and screw them together. The base version will include two rubber wheels driven by Dynamixel X series motors plus two casters, but if you want to upgrade to an awesome pair of tracks to turn your TB3 into a little all-terrain tank, it’s as easy as printing out two more wheel hubs, swapping some plates around, and adding a pair of servos and the tracks. 

Turtlebot 3Image: Robotis

Sensors and computers work the same way. Out of the box, TB3 will be able to run on anything from a Raspberry Pi to an Intel Joule. Of course, what hardware you want depends on what kind of perception you’re trying to do, and TB3 is set up for a variety of USB cameras, RGB-D sensors, lidars, and of course all of the low-cost IR and ultrasonic sensors as well. With a Joule and a RGB-D camera or lidar, TB3 is powerful enough to do SLAM and navigation.

But wait, there’s more: Want to do manipulation of small things very close to the floor? No problem, stick an arm on it. With a little creativity, TB3 can turn into just about anything you want:

TB3 also uses a custom controller board developed by ROBOTIS. It’s completely open source, meaning that you can really get into the guts of the hardware and interfaces later, if that’s your thing:

Turtlebot 3Image: Robotis

According to ROBOTIS, the TB3 will start at around $500 when it becomes available next year, although they’re still trying to decide on the final specs and pricing model, and fancier versions (say, with an Intel Joule board and RealSense camera) could run you up to $1000. The price would be for a full kit— if you’re comfortable buying the electronics and sensors (or you already have some), and you 3D print the frame parts yourself, it’ll be even cheaper. And since it’s all modular, you can always just start with the base version (probably Raspberry Pi with a camera) and upgrade as you start to run into hardware constraints on what you’re trying to do.

ROBOTIS hopes that TB3’s low price and flexiblity will help it find a home in undergraduate robotics programs, and perhaps even in high schools, which are currently dominated by LEGO Mindstorms and Vex Robotics. We don’t mean to suggest that there’s anything wrong with those systems, but there’s no reason why ROS can’t be an option for early robotics education as well.

All the TurtleBot 3s we saw at ROSCon were prototypes, so nothing’s quite ready to go just yet, but we’ll definitely be updating you as soon as we have more info.

[ ROSCon 2016 ]

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

“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