A small prototype of the self-driving car using a Convolutional Neural Network (CNN). This is my course project to implement an end-to-end method for training convolutional neural networks for the autonomous navigation of a mobile robot. The proposed navigation system shows object detection. The vehicle it is used can directly output the linear velocity of the robot from an input image in a single step. The trained model gives wheel velocities for navigation as outputs in real-time making it possible to be implanted on mobile robots such as robotic vacuum cleaners. The experimental results show an average linear velocity with a maximum turning angle is 30 degrees. The proposed model built in python 3.0 and experimental tests based on small scale car Quanser latest self-driving car (QCar) state-of-the-art product equipped with Jetson TX2 has been conducted to verify the effectiveness of the proposed network.
