The current investigation’s goal is to develop a deep learning algorithm that are used to model the data required for autonomous vehicles and as a result, decisions are made that are appropriate for the driving situation. Autonomous driving presents a multitude of dynamic conditions for the car to operate in, unpredictable sensor input, and complex interactions with other road users, which makes planning and decision-making technologies even more challenging. These vehicles uses Neural Networks to detect various factors like line detection, environment segmentation and Navigation. Neural Networks also been used to understand how the vehicles can drive by itself. Over 25,000 photos make up the data collection, each of which is labeled with information on the type of driving and road behavior it shows, such as lane-keeping or turning. The simulation has a 320 × 160 pixel image resolution and operates at a frame rate of 10 frames per second. Before instruction begins, a picture will be normalized by eliminating the sky and other extraneous details to facilitate learning. Main Image, the pre-processing methods used by CNN (Driving Model) are also applied to the images derived from the gathered data. The cost of the center camera accuracy of prediction correction genuine worth CNN Right and left cameras are driving angles. Revision backward propagation Rotation, a vertical flip, a horizontal flip, and an RGB to YUV conversion are all given weights to increase the precision of the categorization outcomes. In the training approach, the first three convolution layers utilize a 5 × 5 kernel and remaining three convolution layers utilize a 3 × 3 kernel, with a maximum of 20 epochs and 2.000 samples per epoch. After two or three practice laps, the simulator enters autonomous mode. Eventually, the automobile will be able to predict changes in route and function autonomously thanks to a CNN model. With the utilization of with an i5-4200U CPU, an NVIDIA Geforce 740M GPU, and 12G of RAM the simulation is conducted on a laptop Innovation and uses: Industrial IoT and autonomous cars go hand in hand. The Internet of Things (IoT) provides the necessary technology for autonomous vehicles, working in concert with other techniques like artificial intelligence, deep and machine learning, and local computing and many more. A lot of people are curious about how these self-driving cars operate, what keeps drivers’ minds working properly when they’re not driving a car. It is well known that modern cars have a significant number of sensors, actuators, and controls. These end devices are powered by various functionally specific software running on Electronic Control Units (ECUs). This package comprises the machine learning software’s. Self-driving automobiles make use of the machine learning approaches since they continuously model and anticipate changes in the environment.