Autonomous cars are smart cars that are expected to be driverless, efficient, and crash-free ideal urban vehicles in the near future. The future of autonomous vehicles will be extensive network guided systems working in conjunction with vision guided features and other sensory systems. In modern vehicles, we can see varying levels of automation where the vehicle can drive itself, while others can only have some level of automation. AIVs (Automated Intelligent Vehicles) guarantee a safe, comfortable, and efficient driving experience. However, fatalities involving AIV-equipped vehicles are on the rise. The full potential of autonomous intelligent vehicles will not be realised unless the robustness of current technology is improved further.
For this project, we want the AIV to choose his path based solely on the destination. We created this using a Raspberry Pi 3 and the Ubuntu operating system. Our project is based on creating an autonomous intelligent vehicle that can decide for itself which path to take and avoid any obstacles that may be in its path. A map of the room in which the vehicle is present would be stored on the single board computer (Raspberry Pi). The user will provide the bot's starting coordinates, and the Raspberry Pi will collect and perform all necessary calculations before proceeding to the destination.
This paper discusses unresolved issues and provides a technical overview of automated driving. Current challenges, high-level system architectures, emerging methodologies, and core functions such as localization, mapping, perception, planning, and human-machine interfaces were thoroughly examined. Furthermore,many cutting-edge algorithms were implemented and tested in a real-world driving environment on our own platform. The paper concludes with a discussion of the datasets and tools available for AIV development.