UC Riverside

The needs for traveling and freight shipping keep rocketing in our modern society. As a result, the size of the transportation networks is rapidly increasing, and the number of vehicles is also fast growing. The pertinent safety, mobility, environmental sustainability problems raise public concerns. Intelligent Transportation Systems (ITS), utilizing Intelligent Vehicles (IVs) that possess automation capabilities and/or the ability to communicate with other entities, offer a promising solution. This approach paves the way for traffic management without adding the cost of infrastructure expansion.In the foreseeable future, IVs are a primary focus within ITS, and they are expected to coexist with conventional vehicles in mixed traffic scenarios. Such a setting demands an in-depth understanding of complex inter-vehicle interactions during motion planning. The potential for aggressive or overly cautious behaviors from IVs requires an awareness and prediction of other road users’ behaviors. To understand other road users, behavior modeling is an indispensable topic. An intelligent vehicle needs to model not only its own driver’s behavior for a better user experience, but also other road users’ behavior. With the development of sensing, communication, and cloud computing technologies, there will be more and more available data. To take advantage of the abundant data, personalized behavior is comprehensively studied to build a more accurate model. However, the concern on current “black box” strategy impedes the acceptance and trust on IVs. Therefore, the need for developing data-driven but explainable algorithms emerge. This dissertation aims to develop a personalized behavior-aware motion planning framework for intelligent vehicles operation. The framework incorporates driver profile modeling, driving preference modeling, and interaction pattern recognition for the behavior aspect. Furthermore, the motion planning component of the framework integrates the behavior model, thereby adapting to diverse driving styles and preferences in mixed traffic environments. The end goal is a more safe, efficient, and environmentally friendly transportation system, aligning with the objectives of Intelligent Transportation Systems.