Microsoft's Robotics IDE has been extended to cope with Kinect-based robots - which makes sense given how useful a depth sensor is. The first robotic hardware kit for combining a laptop and a Kinect has also been announced.
The Robotics Developer Studio 4 (RDS4) includes a simulator so you can create a virtual robot without having any hardware, but the main aim of the software is to give you the means to put together a robot made up of to four Kinect sensors, a laptop and some servos so your robot can move.
Your robot can then use the Kinect sensors to work out where it is in relation to other objects, and make use of skeleton tracking and speech. The raw sensor stream is available so you can build additional capabilities such as navigation algorithms, and there’s an obstacle avoidance service that shows how to carry out directional based navigation.
RDS includes a lightweight asynchronous services-oriented runtime, a set of visual authoring and simulation tools, along with templates, tutorials, and sample code to help you get started. Microsoft expects the toolkit to be used by commercial robotics manufacturers as well as home users, and the software includes support for Microsoft .NET Framework 4, XNA Game Studio 4.0, and Visual Studio 2010.
You can view an example of the type of results you can achieve in this video which shows the Kinect Follow Me robot using the sensors to follow its human ‘master’ around.
Microsoft has also announced a kit from Parallax, based on RDS, and expects other robotic hardware vendors to make kits and preassembled units available. The first kit is called Eddie and can be pre-ordered from Parallax for £1249.
The kit comes unassembled and includes all the robotic hardware that you need to combine with a laptop and a Kinect. The kit is based on an 8-core Propeller P8X32A microcontroller which connects to the laptop via USB. This is mounted on a control board and used to directly control two 12V motors and collect data from several sensors around the robot.
The Eddie Control Board provides access to additional I/O so you can expand Eddie with sensors and accessories.
Microsoft has also announced the 10 finalists for the Robotics @ Home competition they launched last September. Details are available on the Robotics @ Home Finalists page.
Where RDS sits in relation to the open source ROS - Robot Operating System - is an interesting question. At the moment it looks as if RDS is mostly considered to be an educational tool and as such it works well and products like the Eddie Platform can only help
Udacity has opened a new Nanodegree in Machine Learning. At advanced level it prepares you to apply predictive models to massive data sets in fields like education, finance, healthcare or robotic [ ... ]