Design of an IoT based autonomous vehicle with the aid of computer vision

A web controlled and partially autonomous vehicle system is presented in this paper. It highlights the idea to develop a remote controlled car which can be driven from anywhere using internet over a secured server. This car will also have limited automation features like traffic light detection, obstacle avoidance system and lane detection system so that it can drive itself safely in case of connectivity failure. The main goal here is to minimize the risk of human life and ensure highest safety during driving. At the same time the car will assure comfort and convenience to the controller. A miniature car including the above features has been developed which showed optimum performance in a simulated environment. The system mainly consists of a Raspberry Pi, an Arduino, a Picamera, a sonar module, a web interface and internet modem. The Raspberry Pi was mainly used for the Computer Vision algorithms and for streaming video through internet. The proposed system is very cheap and very efficient in terms of automation.