A hierarchical planning and control framework for structured highway driving. Issue 1 (July 2017)

• Record Type: Journal Article
• Authors: Zhou, Shiying; Wang, Yizhou; Zheng, Minghui; Tomizuka, Masayoshi
• Is Part Of: IFAC-PapersOnLine. Volume 50:Issue 1(2017)
• Page Start: 9101
• Page End: 9107
• Publication Date: 2017-07
• Subjects: Vehicle dynamic systems -- Control architectures in automotive control -- Mission planning -- decision making -- Motion control -- Trajectory -- path planning -- Trajectory tracking -- path following; Automatic control -- Periodicals; 629.805
• DOI: 10.1016/j.ifacol.2017.08.1705
• Languages: English
• ISSNs: 2405-8963

Abstract

Abstract: Real-time planning and control play essential roles in autonomous driving under the structured environments. In this paper, a hierarchical planning and control framework is proposed to enable safe and robust self-driving of the ego vehicle. Instead of rule-based decision making, a Markov Decision Process (MDP) is employed for automated decision making by assessing dominant factors of driving stability, efficiency and safety. A real-time trajectory smoother is combined with versatile low level controllers. In addition, the sensor detection limitation is considered in the longitudinal control, which regulates vehicle cruising, following and collision avoidance with smooth speed profile. The proposed framework alleviates the high level planning responsibility and increase the system reliability. Simulations are performed on a dynamic vehicle model in a two-lane environment based on OpenDRIVE standards. The efficiency and effectiveness of the proposed technique are verified.