Not a tutorial but a video, but here's Anthony Cowley talking about using Haskell for machine vision, hardware, and robotics. In the first 5 minutes he writes Haskell code, proves it bug-free using a few tests via a Satisfiability Modulo Theorems prover, cross-compiles it to C, starts up a web server for debugging, then throws it on an Arduino.
Then he proceeds to run some basic vision filters in realtime using OpenGL, OpenCL, and OpenCV all together just to show off all in under 150 lines.
Then after that he talks about how his team built a robot doing realtime grasping and lifting of non-uniform, unpredictable, difficult to grasp, random dollar store objects that are coming at it on a conveyor belt moving at 33cm/sec. The visual processing all in Haskell from the noisy sensor of a Kinect.
The examples aren't immediately there to follow as it's not a tutorial, but it's a good look into the space of some real, complex hacking going on.
Then he proceeds to run some basic vision filters in realtime using OpenGL, OpenCL, and OpenCV all together just to show off all in under 150 lines.
Then after that he talks about how his team built a robot doing realtime grasping and lifting of non-uniform, unpredictable, difficult to grasp, random dollar store objects that are coming at it on a conveyor belt moving at 33cm/sec. The visual processing all in Haskell from the noisy sensor of a Kinect.
The examples aren't immediately there to follow as it's not a tutorial, but it's a good look into the space of some real, complex hacking going on.
http://vimeo.com/77164337