Date of Award
8-2024
Document Type
Dissertation
Degree Name
Doctor of Philosophy (PhD)
Department
Automotive Engineering
Committee Chair/Advisor
Yunyi Jia
Committee Member
Beshash Ayalew
Committee Member
Johnell Brooks
Committee Member
Bing Li
Abstract
An autonomous vehicle (AV) is a promising engineering innovation that can operate without a human driver. Nonetheless, to ensure passenger safety and comfort in the absence of human drivers, the vehicle must continuously monitor and predict how the cabin’s state will evolve in the immediate future. The two major factors influencing the cabin’s state are the occupancy statuses of vehicle seats and the activities of human passengers occupying them. Therefore, this dissertation presents a system employing a capacitance-sensing mat and a machine learning unit to monitor vehicle seat occupancy status and track passenger activities non-intrusively. While the mat integrates with vehicle seats to continuously yield capacitance measurements indicative of seat occupancy status and passenger activities, the machine learning unit employs feature vectors extracted from these measurements and a collection of machine learning and deep learning models to classify seat occupancy statuses and the postures and actions of passengers. The respective models that classify seat occupancy statuses and passenger postures are a deep feedforward neural network and a k-nearest neighbor classifier. Both models accept a feature vector instance as the input to yield the class label for the respective objective. The model that classifies passenger actions is a long short-term memory neural network that accepts as input temporal sequences of feature vectors or capacitance measurements. The proposed system has demonstrated its robustness and effectiveness in achieving its objectives. The research’s future direction seeks to broaden the system’s scope by considering more complex human passenger activities and types of inanimate objects.
Recommended Citation
Prasanna Kumar, Rahul, "Monitoring Vehicle Seat Occupancy Status and Tracking Human Passenger Activities Inside Vehicle Passenger Cabins" (2024). All Dissertations. 3754.
https://open.clemson.edu/all_dissertations/3754