VANFIST: Versatile Autonomous Navigation for Flexible Industrial and Strategic Tasks

This project is a simulation of a self-driving car built at a 1/6 scale of a normal vehicle. It is an early test version designed to learn and practice new ideas that could later be used in full-size cars. The model uses only electric motors and an electric engine, which makes it simple and clean to run. The car will be controlled by a machine learning system running on an Nvidia Jetson Orin Nano processor. This system will take in information from a camera and a LiDAR sensor. It looks at the images and distance data to spot obstacles, read traffic signs, see the lines on the road, and check how far away things are. All this information will be turned into simple signals that help the car know what to do. For example, the system will tell the car when to speed up, slow down, brake, turn, or use its turn signals. In addition to handling immediate actions, the system will also be able to plan its route. In the short run, it can decide to change lanes, avoid obstacles, or stop at a stop sign. Over a longer period, it can calculate the best path to take from one place to another. This dual planning makes sure that the car not only reacts safely to nearby dangers but also travels efficiently from start to finish. The main goal of this project is to provide a hands-on learning experience that integrates the fields of mechanics, electronics, and machine learning. By working on this scaled-down self-driving car, students gain practical knowledge about how sensors, processors, and control systems interact in real-world applications. This approach helps build a solid foundation in both the theoretical and practical aspects of modern automotive technology.