基于MR混合现实的足球机器人目标跟踪仿真  

Simulation of Soccer Robot Target Tracking Based on MR Mixed Reality

作　　者：黄全高[1] 谢亦才[1] HUANG Quan-gao;XIE Yi-cai(Gannan Normal University,Ganzhou Jiangxi 341000,China)

机构地区：[1]赣南师范大学,江西赣州341000

出　　处：《计算机仿真》2024年第4期408-412,共5页Computer Simulation

基　　金：2021年江西省教育厅科技项目(GJJ211408,2021.06)。

摘　　要：机器人跟踪足球时,由于目标具有较强的运动随机性,且较小的目标尺度增大了跟踪框的自适应跟踪难度,目标跟踪时会将很多背景信息包含在内,导致目标模型的更新错误。为此,提出基于MR混合现实的足球机器人目标跟踪方法。构建MR混合现实平台,令足球机器人在虚拟现实场景下完成人机交互,在保证目标跟踪安全性的同时提升目标跟踪性能;采用帧差法获取目标模型,利用投影法对二值图像投影,并对差分图像滤波,以此获取足球机器人在目标模板区域及候选目标区域中的直方图。基于此,更新足球机器人目标模型,实现基于MR混合现实的足球机器人目标跟踪。实验结果表明,当跟踪误差阈值为0.8mm时,研究方法的目标跟踪准确率和成功率均可达95%以上,平均耗时为55.17ms。当足球目标出现遮挡问题时,该方法仍能完成高精度跟踪。When a robot tracks a football,strong motion randomness and the small scale of the target increase the adaptive tracking difficulty,resulting in the update error of the target model.Therefore,a target tracking method for the football robot based on MR was put forward.Firstly,we built a MR platform to enable the robot to complete human-computer interaction in VR scenes,and thus improve target tracking performance while ensuring target tracking security.Secondly,we used the frame difference method to obtain the target model.Meanwhile,we used the projection method to project the binary image onto it and filter the differential image,thus obtaining the histogram of the robot in the target template and candidate target.On this basis,the target model of the robot was updated to achieve target tracking of the football robot based on MR.Experimental results prove that when the tracking error threshold is O.8 mm,the tracking accuracy and success rate of the proposed method can reach more than 95%,and the average time consumption is 55.17 ms.When the target is occluded,the method can still achieve high-precision tracking.

关 键 词：混合现实 足球机器人 目标跟踪 人机交互 帧差法 目标模板区域 

分 类 号：TP242.62[自动化与计算机技术—检测技术与自动化装置]