摘要
利用反向传播神经网络(BP-NN)学习算法,对机器人臂的重力补偿进行了研究。给出了机器人臂各关节扭矩的重力项理论计算公式及其连杆参数识别方法,同时,对BP-NN算法进行了详细分析,利用BP-NN来处理机器人臂重力项并进行试验。试验结果表明,采用该学习算法得到的机器人臂重力项输出值和实测值基本一致,能有效减少机器人臂重力项计算量,达到实时控制的目的。
By adopting back propagation-neural network（BP-NN） learning algorithm,the gravity compensation for robot arm is researched.Both the theoretical computation formula of the gravity item of each joint torque in robot arm,and the parameter identification method of their linkages is given,in addition,the BP-NN algorithm is analyzed in detail,and the BP-NN is used to handle the gravity items of robot arm and the experiment is conducted.The test result shows that the output value of gravity items of robot arm learnt with the BP-NN is basically conforming with the measured value,and the work load of calculation for the gravity items of robot arm is effectively reduced and the real-time control can be carried out.
出处
《自动化仪表》
CAS
北大核心
2012年第2期22-24,共3页
Process Automation Instrumentation
基金
国家留学基金资助项目(编号:2007102654)
重庆市教委2011年度科学技术研究基金资助项目(编号:KJ112203)
关键词
BP-NN
最小二乘法
重力补偿
实时控制
参数识别
Back propagation-neural network（BP-NN） Least square method Gravity compensation Real-time control Parameter identification
