As human drivers' roles diminish with higher levels of driving automation (SAE L2-L4), understanding driver engagement and fatigue is crucial for improving safety. We developed an integrated hardware and software system to analyze driver interaction with automated vehicles, with a particular focus on cognitive load and fatigue assessment. The system includes three submodules; namely the Driver Behavior Measurement (DBM), Vehicle Dynamics Measurement (VDM), and the Driver Physiological Measurement (DPM). The DBM module uses electro-optical (EO) and infrared (IR) camera to track a number of facial features such as eye aspect ratio (EAR), mouth aspect ratio (MAR), pupil circularity (PUC), and mouth to eye aspect ratio (MOE). Although determining these metrics from images of the driver’s face in conditions such as low light or with sunglasses is challenging, the paper showed that fusion of EO and IR image analysis produces robust performance. The VDM module utilizes an Inertial Measurement Unit (IMU) to provide vehicular motion data such as speed, acceleration, braking and yaw rate to aid detection of fatigue-related irregularities. A wearable heart rate monitor was used in the DPM module to track driver heart rate as an indicator of stress and fatigue. Data from these modules is fused and processed using a previously published CNN-LSTM model, achieving 90.1% accuracy in detecting fatigue in preliminary tests performed with one driver. The test results show that the system is robust, scalable, and suitable for large-scale studies on driver engagement with highly automated vehicles.