Predictive Risk Analysis and Safe Trajectory Planning for Intelligent and Connected Vehicles

The intelligent and connected vehicles(ICVs), aiming at providing safe, comfortable and efficient transportation experience for users, have drawn increasing research interest recently [1]. These vehicles leverage advanced technologies such as artificial 1 intelligence, machine learning, and real-time data communication to enhance their operational capabilities. Among all the research domains, studies focusing on the safety of ICVs hold the utmost significance, as safety is the fundamental necessity in transportation [2]. Ensuring the safety of ICVs not only protects passengers and pedestrians but also fosters public trust in autonomous driving technologies. Consequently, researchers are exploring various safety measures, including robust sensor systems, fail-safe mechanisms, and comprehensive risk assessment frameworks, to mitigate potential hazards associated with ICV operation. To guarantee the safety of ICVs, numerous risk assessment algorithms have emerged [3, 4]. The initial category of risk assessment algorithm calculates the driving risk based on time, such as time headway (THW) [5, 6] and time to collision (TTC) [7, 8].