基于混合滤波器的水炮打击目标轨迹预测方法  

Water Cannon Target Track Prediction Method Based on Hybrid Filter

作　　者：李福成 陈姚节[1,2] LI Fucheng;CHEN Yaojie(College of Computer Science and Technology,Wuhan University of Science and Technology,Wuhan 430065;Metallurgical Industry Process National Virtual Simulation Experimental Teaching Center,Wuhan 430065)

机构地区：[1]武汉科技大学计算机科学与技术学院,武汉430065 [2]冶金工业过程国家级虚拟仿真实验教学中心,武汉430065

出　　处：《计算机与数字工程》2024年第3期886-891,共6页Computer & Digital Engineering

摘　　要：在智能水炮打击处于运动状态的目标船舶时,射流在空中运动所耗时间会导致打击脱靶问题。针对这种情况,提出一种基于混合滤波器的水炮打击目标轨迹预测方法。该方法首先采集目标跟踪设备的轨迹数据,然后使用轨迹数据训练混合滤波器模型,最后使用模型实时预测目标轨迹。实验结果表明,在搭建的智能水炮实验环境中,持续预测目标未来0~2s的轨迹,均方误差在0.3025~1.4832之间,预测效果良好,在实际应用中,该方法有效地解决了智能水炮打击运动目标时的脱靶问题。When the intelligent water cannon strikes the target ship in the moving state,the time spent by the jet movement in the air will lead to the attack miss problem.In order to solve this problem,a hybrid filter based on water cannon target trajectory pre-diction method is proposed.The method first collects the trajectory data of the target tracking device,then uses the trajectory data to train the hybrid filter model,and finally uses the model to predict the target trajectory in real time.The experimental results show that in the intelligent water cannon experimental environment,the prediction effect is good,and the mean square error is between 0.3025 to 1.4832.In practical application,the method effectively solves the miss problem when the intelligent water cannon strikes the moving target.

关 键 词：智能水炮 脱靶修正 混合滤波器 轨迹预测 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]