WORKSHOPLIST

WORKSHOP DETAILS

WS20: Parallel Planning for Autonomous Driving

Workshop Code: 3x21k

Organizers

  • Long Chen
    Affiliation: Sun Yat-sen University, China
    E-mail: chenl46@mail.sysu.edu.cn

  • Xuemin Hu
    Affiliation: Hubei University, China
    E-mail: huxuemin2012@hubu.edu.cn

  • Yun-xiao Shan
    Affiliation: Sun Yat-sen University,  China
    E-mail: yunxiao.shan@gmail.com

Scope and Goals

Planning is one of the most important parts for autonomous vehicle systems, and decision making in critical scenarios is the key to autonomy using planning algorithms. In order to make safe and collision-free decisions for an autonomous vehicle to move from the current location to its destination, planning methods need to be studied by researchers.

Parallel theory has been widely applied and achieved great progress in recent years. In the meanwhile, the parallel driving framework has been persistently developed. The parallel theory and parallel driving framework give a new research direction on the planning field. Parallel planning, where the parallel theory is introduced into planning field, has the potential to achieve a more effective, efficient and robust solution for planning problems in autonomous driving.

However, there are some challenges for studing parallel planning. How to introduce the parallel theory into tranditional path planning and motion planning algorithms? How to design architectures of parallel planning models? How to combine deep learning methods with the parallel theory in planning models? How to make vehicles drive the same way as human drivers? How to make vehicles own the ability of keeping learning from the environment?

The special session aims to provide up-to-date research that could help advance understanding of parallel planning problem in autonomous driving.

Topics of Interest

  • The reseach on the theory of parallel planning.
  • End-to-end motion planning for autonomous driving.
  • Imitation learning for driving a vehicle.
  • Reinforcement learning in motion planning for autonomous driving.
  • Motion planning using the input data from various kinds of sensors.
  • Reliable and efficient reseaches on planning algorithms for autonoumous driving.
  • Security issues regarding to parallel planning.