In this chapter, we mainly discussed the various machine learning techniques and libraries that can be interfaced with ROS. We started with the basics of machine learning and deep learning. Then, we started working with TensorFlow, which is an open source Python library mainly for performing deep learning. We discussed basic code using TensorFlow and later combined those capabilities with ROS for an image recognition application.

After discussing TensorFlow and deep learning, we discussed another Python library called scikit-learn, which is used for machine learning applications. We saw what SVM is and examined how to implement it using scikit-learn. Later, we implemented a sample application using ROS and scikit-learn for classifying sensor data. This chapter provided us with an overview of the integration of Tensor Flow in ROS for deep learning applications. 

In the next chapter, we will look at how autonomous cars work and try simulating them in Gazebo.