Summary

In this chapter, we first went through detailed steps to set up Raspberry Pi with all the necessary accessories and operating system, as well as the GoPiGo toolkit that turns a Raspberry Pi board into a moving robot. We then covered how to install TensorFlow on Raspberry Pi and build the TensorFlow library, and how to integrate TTS with image classification, and audio command recognition with GoPiGO APIs, resulting in a Raspberry Pi robot that moves, sees, listens, and speaks, all without using Cloud APIs. Finally, we introduced the OpenAI Gym toolkit for reinforcement learning and showed you how to build and train a powerful reinforcement learning neural network model using TensorFlow to keep your robot in balance in a simulated environment.