Evaluate the Driving Safety of the User Based on the Trajectory Collected: User-Specific Behavior for Safety Analysis
This study aims to evaluate the driving ability of users by collecting data from the CARLA simulator. Various types of data, such as collisions, line crossings, and traffic rule violations, are ordered to analyze the user's driving behavior. Initially, the User will provide information, such as age and driving experience, to evaluate their impact on driving ability. Now, The collected data will analyzed using Maximum Entropy Inverse Reinforcement Learning. The results provide insights into the user's driving behavior, including their driving preferences and patterns. Using information from can improve driving skills and reduce accidents in the real world if they follow the program interaction. Moreover, the study contributes to developing autonomous driving systems by identifying common driving patterns and behaviors. The study can help to design more personalized autonomous driving systems that mimic the user's driving style. In summary, this study provides insights into driver behavior analysis and the driving ability of users based on data collected during the CARLA simulation. The findings can improve driving safety by identifying areas where users need to improve their driving. Additionally, this project can use reinforcement learning to develop autonomous driving systems by providing an understanding of user behavior and preferences.