Multi-Modal Driver Behavior Analysis and Speed Estimation using Fusion of Computer Vision and InCar Sensor Data

This research presents a thorough and cohesive approach for examining driver behavior and precisely determining vehicle speed by combining several sensory inputs. The system we propose utilizes cameras, LIDAR, RADAR, and inertial sensors to collect a wide variety of data, allowing for a comprehensive understanding of both driver behavior and the surrounding driving conditions. By utilizing computer vision and machine learning methods, the system analyzes facial expressions, hand gestures, and contextual elements to assess cognitive states and detect any distractions. An important breakthrough is our innovative technique for estimating speed, which cleverly integrates information from RADAR, LIDAR, and inertial sensors. By adopting this integrated technique, accurate speed measurements may be obtained under different driving conditions. The effectiveness of the system is thoroughly verified through a blend of simulations and real-world tests, highlighting its capacity to greatly improve the safety and intelligence of advanced driver assistance systems. This study enhances the progress of intelligent transportation systems by offering a strong and dependable structure for examining driver behavior and estimating vehicle speed.