基于深度学习的羽毛球后场击球线路规划方法

Method of Badminton Backcourt Shot Route Based on Deep Learning

为提高羽毛球后场击球线路的规划能力,本研究设计了基于深度学习的羽毛球后场击球线路规划方法。首先采用运动基元轨迹自适应学习方法设计击球线路节点和路径空间,从而构建击球线路最短分布网格结构模型。然后分析击球线路规划约束参数,再根据节点定位和最短路径寻优部署过来控制击球位置和运动员姿态。最后结合深度学习方法实现线路规划过程的自适应寻优。仿真结果表明:该方法的学习控制能力较好,且具有较好的收敛性能,提高了羽毛球后场击球线路规划的可靠性。

In order to improve the badminton backcourt shot route,a method was designed based on deep learning.Firstly,the path node and path space of the batting path were designed to construct the structure model of the shortest distribution grid of the batting path by using the adaptive learning method of the trajectory of the motion element.Then the constraint parameters of batting line planning were analyzed,and the batting position and player posture was controlled according to node positioning and shortest path optimization.Finally,the deep learning was used to adaptively optimize the shot route.The simulation results showed that this method had good learning control ability and good convergence performance,and improved the reliability of badminton backcourt shot route.