Design and Development of Autonomous Driving Car Using NvidiaJetson Nano Developer Kit

The implementation of the steam engine sky-rocketed humanity’s progress towards the current age of technological marvel. A lot of modern electro-mechanical feats can be traced back to their inception. With the many obvious benefits of an autonomous driving system, the future of self-driving vehicles is looking very bright. A system like this could greatly benefit our environment as the usage of self-driving vehicles could reduce fuel consumption worldwide by over 10%. In this paper, we present a design of a system consisting of a self-driving car model which employs Jetracer, an AI framework for autonomous driving cars. We make use of AI and ML frameworks such as Pytorch, OpenCV, and TensorRT. We apply image recognition to capture traffic signs and classify them using the mentioned AI and ML frameworks and respond to them in real-time through the Jetson Nano’s interface.In our work, we have employed Jetracer using interactive web programming via a web browser. It allows high frame rate processing due to torch2trt (PyTorch to TensorRT translator) optimization, to achieve faster autonomous line driving using the jetson nano. Also, we have summarized the results highlighting the self-driving car model that uses AI, Machine Learning, and Neural Networks to autonomously drive on a track