3D Search Based Hybrid Optimal Trajectory Planning for Autonomous LHD in Turning Maneuvers

Abstract

Load-haul-dump (LHD) is the main equipment for mining and transportation in underground metal mines. Autonomous driving represents an essential role in the daily operation of LHD. Nowadays, LHD autonomous driving is based on the human driver’s artificial teaching. However, it is not easy to obtain high quality teaching information for all the underground roadways in a mine, which not only affects the commissioning and deployment time of autonomous driving systems for LHD vehicles, but also not conducive to the full play of automation equipment performance, and finally impacts the production efficiency. Planning a high-quality target reference trajectory for LHD in the underground roadway is an effective way to overcome the above issues. This paper proposes an optimal trajectory planning method for LHD in turning maneuvers. This planning method is based on longitudinal and lateral decomposing, integrated sampling method, and optimization method. The contribution of this work lies in the 3D key factors search strategy design, the key parameters value range exploration and the reasonable key factor initial values design, which can improve the online execution ability of the algorithm and reduce the driving time of the LHD. The 3D key factors refer to the terminal state of the trajectory, which are longitudinal driving distance, the lateral driving distance, and travel time. The proposed method is simple and easy to implement. The initial values of the key factors are given, and the effectiveness of the proposed method is proved. The feasibility and effectiveness of the proposed method are verified by the comparative analysis of a series of examples.