摘要
本文利用优化方法及递归神经网络实时求解器消除冗余机器手臂在运动过程中出现的角偏差问题。鉴于机器手臂都存在着关节物理约束,我们的优化方案也因此考虑关节极限和关节速度极限的躲避。更重要的是,本文详细分析了该成功解决关节角偏差问题的二次型性能指标的设计原理。仿真结果证实了该方法的可行性与有效性。
The joint angle drift problem of redundant robot manipulators could be remedied by using a quadratic-programming （QP） based problem formulation and its online neural QP solver. Joint limits and joint velocity limits are taken into consideration as well. The design principle of the quadratic performance index is investigated and proved via two different approaches. Simulation results substantiate the feasibility and effectiveness of such an optimization-based redundancy-resolution scheme.
出处
《微计算机信息》
2009年第1期265-267,共3页
Control & Automation
基金
机器手臂的基于二次规划的冗余度解析方案(60775050)基金颁发部门:国家自然科学基金委基金申请人:张雨浓
关键词
冗余度机器手臂
角偏差现象
二次型性能指标
梯度下降法
redundant robot manipulators
joint angle drift phenomenon
quadratic program
performance indices
gradient descent method.
