磁滞
粘弹性
前馈
控制理论(社会学)
材料科学
执行机构
非线性系统
介电弹性体
蠕动
电介质
补偿(心理学)
弹性体
计算机科学
复合材料
工程类
控制工程
物理
人工智能
控制(管理)
光电子学
精神分析
量子力学
心理学
出处
期刊:IEEE robotics and automation letters
日期:2019-07-01
卷期号:4 (3): 2340-2347
被引量:40
标识
DOI:10.1109/lra.2019.2902954
摘要
Dielectric elastomer actuators have shown promising applications in the field of soft robotics. However, due to their rate-dependent viscoelastic hysteresis nonlinearity, it is still challenging to achieve precision tracking control of dielectric elastomer actuators. In this letter, we propose a feedforward control approach that can compensate for the rate-dependent viscoelastic hysteresis nonlinearity with maximum tracking errors of 6.18% and root-meansquare errors of 2.96% when the frequency of the input voltage is between 0.05 Hz and 1.5 Hz. Our control approach consists of two feedforward compensators: 1) for the ease of the hysteresis compensation, a creep compensator is firstly developed to remove the viscoelastic creep nonlinearity; 2) based on a phenomenological mathematical model, an inverse hysteresis compensator is then developed to compensate for the rate-dependent viscoelastic hysteresis nonlinearity. Experimental results of tracking various periodic trajectories demonstrate that: the maximum tracking errors are reduced by 87.17% and the root-mean-square errors are decreased by 89.53%, by comparing the results without the viscoelastic compensation. It is the first time to successfully compensate for both the viscoelastic creep nonlinearity and rate-dependent hysteresis nonlinearity of dielectric elastomer actuators by a feedforward control approach, which may pave the way for further applications in dielectric-elastomer-actuators based soft robotics.
科研通智能强力驱动
Strongly Powered by AbleSci AI