optimization-based interactive speed planning of autonomous vehicles in structured environments

Safe and efficient interaction with other moving vehicles is crucial for autonomous vehicles. Speed planning affects the safety and smoothness of autonomous vehicles in dynamic environments, and is still an open problem. This paper focuses on how to avoid colliding with around moving vehicles by planning the speed of autonomous vehicles when its driving path is determined. The proposed method combines a heuristic search algorithm and a quadratic optimization algorithm to generate a safe and smooth speed profile for autonomous vehicles. In the interactions with other dynamic vehicles, the planning speed of the autonomous vehicles is adaptive to the changing environments. Finally, the feasibility and flexibility of the proposed method have been validated by the simulation results in two typical interaction scenarios by using Prescan Software.