ROS Resources: Documentation | Support | Discussion Forum | Index | Service Status | ros @ Robotics Stack Exchange
Ask Your Question

Alternate Turtlebot Configurations

asked 2013-02-09 11:27:14 -0500

Cerin gravatar image

Does anyone know of methods for powering and recharging all components from the iRobot Create on a Turtlebot? I'd like to get a Turtlebot, but it's hacky power system and inability to auto-recharge, despite the Create's builtin auto-recharge mechanism, is a huge non-starter.

Specifically, would it be practical to replace the laptop/netbook with a more efficient Raspberry Pi or OLinuXino, which only need 5V?

Similarly, has anyone tried replacing the Microsoft Kinect (requiring 12V) with the much smaller and more energy efficient Asus Xtion, which only needs 5V over USB?

edit retag flag offensive close merge delete

1 Answer

Sort by ยป oldest newest most voted

answered 2013-02-09 14:11:08 -0500

Ryan gravatar image

If it helps, the TurtleBot2 has a much better power system and can auto-recharge.

Other than that:

1) Yes, it is possible to replace the netbook with a single-board computer. We do not ship with one by default because it's a higher barrier for entry for new Linux & ROS users, but it's quite possible. However, if you're working with Kinect data, we suggest something faster than a Raspberry Pi (even if you're only using pointcloud_to_laserscan and transmitting the scan). I know that @Tully (among others) got ROS running on a RasPi, but it can't do much on that board.

2) Yes, the Xtion is known to work. It's not the "default" sensor for TurtleBots because the Kinect is cheaper and we're really trying to make these robots as economical as possible.

edit flag offensive delete link more


On a side note, why is the Kinect sensor mounted to the rear of the platform where it has to "look through" the bot's interior? Wouldn't it have better visibility, and free up all that interior space, if mounted more towards the front?

Cerin gravatar image Cerin  ( 2013-02-14 11:10:33 -0500 )edit

The Kinect depth sensor has a minimum effective range of ~0.5 m and a limited horizontal field of view. Pulling it back means that you're able to get closer to objects and see a wider area in front of the robot.

Ryan gravatar image Ryan  ( 2013-02-14 14:00:29 -0500 )edit

Question Tools


Asked: 2013-02-09 11:27:14 -0500

Seen: 302 times

Last updated: Feb 09 '13