The combination of machine learning (ML) and deep learning (DL) technologies is largely responsible for the fast progress being made in the car business in terms of connecting and autonomous vehicles. These improvements are turning cars into smart, self-learning machines that can make decisions in real time, do predictive analysis, and talk to their surroundings better. Vehicle connection includes interactions between vehicles (V2V), between vehicles and infrastructure (V2I), and between vehicles and everything (V2X). A lot of people use ML and DL to make communication paths safe, quick, and reliable. A car's ability to drive itself, on the other hand, relies more and more on complicated programs that use sensor fusion, sensing, decision-making, and path planning. This paper talks about how machine learning and deep learning can be used to connect cars better and let them drive themselves. It's mostly about important methods and designs that help make connected vehicle environments and cars that drive themselves. There are a lot of ways to improve real- time decision-making, communication systems, and instrument data handling.