Save test data in csv file for post processing using python

[sentry5588]

Currently the python script ./examples/Robot1/visualization.py reads the stdout from the pipe. It runs flawlessly, but it has two problems for me.

• It doesn't save data that can be used for other post-processing purposes. Such as verifying the estimation algorithm by doing statistical tests of the innovation.
• It doesn't allow debug print. I was trying to add a std::cout in the ExtendedKalmanFilter::update method, but it pollutes the std::out which causes the python script to fail:
  Traceback (most recent call last): File "./examples/Robot1/visualization.py", line 41, in <module> line_data = np.array([float(s) for s in line_string.split(",")]).reshape((12, 1)) File "./examples/Robot1/visualization.py", line 41, in <listcomp> line_data = np.array([float(s) for s in line_string.split(",")]).reshape((12, 1)) ValueError: could not convert string to float: 'fingggerprint'

One possible solution is to save the data to a file, such as "example_robot1.csv" in examples/Robot1/, then read it from the python script visualization.py. @ jwdinius , what is your thought? If you like the idea, I'll make the change and create a pull request. Thanks.