神经网络阻尼比模型及工业机器人导纳控制  

Neural Network Damping Ratio Model and Admittance Control of Industrial Robot

作　　者：党选举[1] 牛嘉晨 DANG Xuan-ju;NIU Jia-chen(School of Electronic and Automation,Guilin University of Electronic Technology,Guangxi Guilin 541004,China)

机构地区：[1]桂林电子科技大学电子工程与自动化学院,广西桂林541004

出　　处：《机械设计与制造》2024年第7期379-384,共6页Machinery Design & Manufacture

基　　金：国家自然科学基金项目(61863008);广西自然科学基金(2016 GXNSFDA380001)。

摘　　要：在工业机器人打磨过程中,环境刚度随未知环境的变化,将对力控制精度产生不利的影响,针对环境刚度变化的问题,该文提出一种基于神经网络阻尼比模型的自适应导纳控制方法。在导纳控制设计中,根据力误差与系统阻尼比之间的机理关系,设计激励函数,构造神经网络阻尼比模型;通过该模型使阻尼比在线调整,适应末端环境的刚度变化,实现力到位置自适应转换的导纳控制。与常规导纳控制进行仿真比较,结果表明所提出的力控制策略力误差更小,响应速度更快,能适应变刚度的未知打磨环境。In the grinding process of industrial robot,the environmental stiffness changes with the unknown environment will have a negative impact to the force control accuracy.In view of the problem of environmental stiffness changes,an adaptive admittance control approach based on damping ratio model of neural network is proposed.In the admittance control design,according to the mechanism relationship between force error and system damping,the activation function is designed,and the damping ratio model of neural network is constructed.Through this model,the damping ratio can be adjusted online to adapt to the stiffness change of the terminal environment,and the admittance control of force-to-position adaptive conversion is realized.Compared with the conventional admittance control,the simulation results show that the proposed force control scheme has lower force error,faster response speed,and can adapt to the unknown grinding environment with variable stiffness.

关 键 词：未知环境 导纳控制 神经网络阻尼比模型 自适应控制 工业机器人 

分 类 号：TH16[机械工程—机械制造及自动化] TP242.2[自动化与计算机技术—检测技术与自动化装置]