18 December 2023 in IEEE Transactions on Vehicular Technology
Motion planning is a crucial and challenging step in autonomous driving navigation tasks. Since traditional motion planning methods largely rely on complex hand-crafted designs, recent research has shifted focus to simple and effective learning-based methods. However, most of the existing learning-based methods primarily employ end-to-end networks to directly generate trajectories or control commands, which may not satisfy vehicle kinematic constraints. These kinematically infeasible outputs cause autonomous driving vehicles deviating from the intended trajectories, significantly impacting both ride comfort and task completion. In this paper, we propose a new Feasible And Smooth Tree (FAST) to concentrate on the generation and utilization of trajectory tree. This trajectory tree structure possesses curvature continuity and kinematic feasibility, making it a cluster of feasible and smooth trajectories that can represent various intentions for traffic participants. Moreover, by employing the trajectory tree as the outputs for both motion planning and auxiliary trajectory prediction tasks, FAST effectively integrates the trajectory tree into a multi-task framework to improve its performance. We evaluate FAST in the closed-loop urban driving setting using the different scenarios provided by CARLA simulator. Experimental results demonstrate that FAST outperforms other learning-based methods while maintaining high efficiency.
5 March 2025 in IEEE Transactions on Instrumentation and Measurement
Navigating in unknown environments with multiple and irregular obstacles is a crucial task for autonomous mobile systems, such as autonomous driving vehicles and mobile robots. However, most of the existing path planners can only effectively solve this problem under the small-scale precise perception conditions, and cannot handle the large-scale and high-dynamic environments. In this paper, we propose a new efficient path planner called IM Planner, which concentrates on the updating and utilization of Incremental patch Map (IM). IM adopts a hierarchical data structure, combining multiple adjacent girds into patches as the basic unit, which facilitates memory management and query operations, and provides accurate obstacle information. After partial sensor observations arrival, IM is continuously updated in patch blocks. At the same time, tightly coupled with IM, the non-optimized search algorithm utilizes a bidirectional strategy, and integrates a fast and accurate collision detection model, which can generate safe, kinematically feasible, and smooth trajectories without post optimization. Extensive experimental results show that IM Planner achieves enhanced efficiency and safety performance among the offline parking benchmark and the online CARLA simulation. The code are available at https://github.com/chenghuang66/IM-Planner.