Prediction of Efficient Driving Mode for Vehicle using Machine Learning

The field of autonomous driving cars is a rapidly growing area of research, machine learning and image recognition technology plays a pivotal role in ensuring their safe navigation on the roads. Different driving modes are offered by modern cars to accommodate various road conditions and driver preferences. However, manually selecting these modes can be cumbersome. Addressing the dynamic nature of mode switching in autonomous driving systems presents a multifaceted challenge. The system must continuously assess a multitude of variables such as traffic patterns, road conditions, weather conditions, and the wear and tear of vehicle components. Traffic conditions can change rapidly, requiring real-time analysis of traffic flow and congestion. Road conditions, including the presence of obstacles, construction zones, or slippery surfaces, demand instant adaptation for safe navigation. Weather conditions further complicate the scenario, as rain, snow, fog, or even intense sunlight can impact visibility and road grip. Addressing these challenges requires sophisticated algorithms capable of processing vast amounts of data from various sensors in real time. Machine learning models must be trained to recognize patterns in diverse conditions, enabling the system to make informed decisions swiftly and accurately. Moreover, the system must possess the ability to prioritize these factors based on their immediate impact on driving safety and efficiency. This can be a future enhancement of the paper. By leveraging machine learning algorithms, the mode selection process can be automated. The system constantly monitors driving conditions, learns from real-time camera sensor data and selects the most optimal mode to adapt to current situation. Different models such as Convolutional Neural Network (CNN), CNN with MobileNetV2 as base and decision trees are analyzed.