摘要
提出了一种通过单目结构光相机测量Stewart载物平台位姿的方法,搭建了位姿误差测量试验台,对Stewart载物平台位姿误差进行了测量。建立了一种使用多元宇宙优化算法(multi-verse optimizer,MVO)优化反向传播(back propagation,BP)神经网络的Stewart载物平台误差估计模型,并利用测量得到的误差数据对模型进行了训练。通过测试数据对误差估计模型的有效性进行了验证。实验结果表明,模型可以更好地估计沿z轴方向的位移误差以及绕x轴和绕y轴的旋转误差,为Stewart平台的位姿测量研究提供了一种新方法。
A method of measuring Stewart platform position and orientation through a monocular structured light camera was proposed.A position and orientation error measurement test-bed was built,and the position and orientation error of Stewart platform was measured.A Stewart platform error estimation model using the multi-verse optimizer(MVO)to optimize the back propagation(BP)neural network was established and trained by the measured error data.The effectiveness of the error estimation model was verified through test data.The results show that the model can better estimate displacement errors along the z-axis direction and rotation errors around the x-axis and y-axis,providing a new method for the research of position and orientation measurement on the Stewart platform.
作者
林皓纯
陈秀梅
彭宝营
王鹏家
LIN Haochun;CHEN Xiumei;PENG Baoying;WANG Pengjia(Mechanical Electrical Engineering School,Beijing Information Science&Technology University,Beijing 100192,China)
出处
《北京信息科技大学学报(自然科学版)》
2023年第4期67-72,共6页
Journal of Beijing Information Science and Technology University
基金
国家重点研发计划子课题(2021YFC2400302)
北京市教委科研计划科技一般项目(KM202011232012)
宁夏自然科学基金资助项目(2022AAC03344)。
关键词
结构光相机
位姿误差
误差估计模型
多元宇宙优化算法
反向传播神经网络
structured light camera
position and orientation error
error estimation model
multi-verse optimizer
back propagation neural network