基于末端触觉传感器的智能机器人压力补偿  

Pressure Compensation for Intelligent Robot Based on Terminal Touch Sensor

作　　者：朱慧珍[1] 王凯[1] 王丽君[2] ZHU Hui-zhen;WANG Kai;WANG Li-jun(Department of Computer Engineering,ShangQiu University,Shangqiu Henan 476000,China;School of Computer Science and Software Engineering,University of Science and Technology,Anshan Liaoning 114051,China)

机构地区：[1]商丘学院计算机工程学院,河南商丘476000 [2]辽宁科技大学计算机与软件工程学院,辽宁鞍山114051

出　　处：《计算机仿真》2024年第4期446-450,共5页Computer Simulation

基　　金：教育部高等教育司产学合作协同育人项目(201702134014);教育部高等教育司产学合作协同育人项目(201802070086);教育部高等教育司产学合作协同育人项目(202002269033);河南省虚拟仿真实验教学项目(教高[2021]22157);河南省实验教学示范中心项目(教高[2011]562)。

摘　　要：末端触觉是智能机器人实现对外部环境做出直接反应的必需媒介。触觉传感器的弹性层易导致表面接触压力测量结果出现偏差,降低机械手的抓取力度的合理性。提出智能机器人末端触觉传感器压力误差补偿方法。建立触觉传感器三维力学模型,分析表面接触应力与弹性层之间的关系,得到传感器压力特征;计算弹性层的低通效应值,通过传感器测量值推理得到表面压力值,利用线性迭代反卷积算法对压力误差补偿。实验结果表明,研究方法应用下,压力误差补偿的均方误差接近于10-3,精度与理想效果更接近,上述方法获取的传感器对不同大小作用力的电学响应结果与实际数据具有较高一致性。Terminal tactile is a necessary medium for intelligent robot to make a direct response to the external environment.At present,the elastic layer of tactile sensor may lead to the deviation of surface contact pressure,reducing the grasping force of manipulator.Therefore,a method of pressure error compensation intelligent robot was proposed based on terminal tactile sensor.Firstly,a three-dimensional mechanical model of tactile sensor was constructed.Secondly,the relationship between surface contact stress and elastic layer was analyzed,and then the pressure characteristic of sensor was obtained.Thirdly,the low-pass value of the elastic layer was calculated,and then the surface pressure value was inferred through the measured value of sensor.Finally,the linear iterative deconvolution method was adopted to compensate the pressure error.The experimental results show that the mean square error of pressure error compensation is close to 10-3 after using the proposed method;This accuracy is closer to the ideal value.In addition,the electrical response results of the sensor to different forces obtained by the proposed method have high consistency with actual data.

关 键 词：智能机器人 触觉传感器 压力误差补偿 三维力学模型 低通效应 

分 类 号：TP642[自动化与计算机技术—控制理论与控制工程]