One of the key components of the sensing system of autonomous vehicles is the LIDAR system that provides real-time and accurate 3D mapping of the environment. The LIDAR system collects an enormous amount of data every 30 to 100 milliseconds that is quite challenging to analyze in a timely manner. The objective of this study is to propose a new method to efficiently analyze the data collected from the LIDAR system and to extract the relevant information to control the autonomous vehicle. In this method, the LIDAR data points are clustered to identify and to locate the important objects such as vehicles, bikes, and pedestrians that directly affect the steering and brake, while ignoring the irrelevant information. The proposed algorithms borrow the concepts of conventional clustering algorithms, such as DBSCAN and Affinity Propagation, but in a purpose-oriented manner that improves the speed of the algorithm. The data collected from the actual driving is used to evaluate the performance of the proposed method in identifying cars, bikes, and pedestrians.

Connectivity and automation provide the opportunity to enhance safety and mitigate congestion in transportation systems. In fact, these technologies can enhance the efficiency of drivers/vehicles’ decision-making by managing and coordinating the interactions among human-driven and connected, automated vehicles. Such management and coordination can lead to developing a collaborative connected, automated driving environment. Game theory, as a methodology to model the outcome of the interactions among multiple players, is a perfect tool to characterize the interactions between these vehicles.

One of the most challenging maneuvers to model is drivers/vehicles’ tactical decisions at intersections. Focusing on unprotected left turn maneuvers, we aim at developing a game theory based framework to characterize driver behavior in unprotected left turn maneuvers in a connected, automated driving environment. A two-person non-zero-sum non-cooperative game under complete information is selected to model the underlying decision-making. NGSIM data is used to calibrate the payoff functions based on Maximum Likelihood Estimation. Validation results indicate that this framework can effectively capture vehicle interactions when performing conflicting turning movements while achieving a relatively high accuracy in predicting vehicles' real choice.