【作者】 强艳辉；

【导师】 温淑焕；

【作者基本信息】 燕山大学 ， 控制理论与控制工程， 2009， 硕士

【摘要】 利用X-Y平台进行抛光,打磨等作业时,平台与环境接触,在精确位置跟踪的同时需对接触表面施加一定的期望力。针对此种受限X-Y平台,本文在参考国内外相关文献的基础上,对其进行了力控制研究,主要研究内容如下:首先,在X-Y平台的约束动力学方程基础上,提出了一种在不改变X-Y平台位置控制器的前提下实现力控制的方案,即系统的外环为力控制回路,根据力误差修正参考位置,使平台与环境的实际接触力跟踪期望力。在力控制回路,利用改进Elman网络在线辨识未知环境,不需要环境位置和刚度的先验知识,这种控制方法有误差补偿作用,对力控制中的干扰和环境不确定因素具有鲁棒性,仿真结果表明了控制方案的有效性。其次,运用力/位置混合控制方法对X-Y平台进行力控制。在提出的力/位置混合控制策略中,位置控制回路采用PD控制,同时,将RBF神经网络用于X-Y定位平台的力控制回路中,利用RBF神经网络学习力控制中的不确定上界,并与反馈控制器结合,进一步确保了控制系统的稳定性,有效地提高了系统的精度和自适应能力。最后,本文将阻抗控制应用于X-Y平台的力控制中,阻抗控制的关键在于阻抗参数的选择,根据接触力和平台位置、速度的变化对阻抗模型参数进行实时模糊调节,减少了受限运动中力干扰的影响,提高了全局力控制效果。更多还原

【Abstract】 The end-effector of a X-Y positioning table is required to keep precise postion tracking and keep a contact force along the outward normal of the constraint surface in tasks such as deburring and grinding. Base on the relevant literature at home and abroad, we did some researches on force control for X-Y table. The main research of this paper is concluded as follows.First, a kind of force control strategy for X-Y table is proposed on the premise that the position controller is not changed, namely, the outer loop is force control loop. The reference position is modified according to the force error so that the expected contact force between the X-Y table and the environment can be achieved. In force control loop, unknown environments is identificated by a modified Elman neural network online. The prior knowledge of the environment is not required. This method can compensate the errors of environment and guarantees the robustness of force control. Simulation results show that this control strategy is effective.Second, a kind of hypbrid force/position controller based on RBF neural network is presented for constrained X-Y positioning table. PD controler is used in position control. In force control, the RBF neural network is applied to learning the upper bound of system’s uncertainties and proportion controller strengthens the completeness of this control strategy. The results of numerical simulation demonstrate the stability and robustness of the system. The precision and adaptability is improved effectively under the proposed control strategy.Finally, an impedance control algorithm is presented in force control for X-Y positioning table. The key of this algorithm is to select the proper impedance parameters. So the parameters of impedance model are adjusted fuzzily real-time according to the change of force, positon and volocity of X-Y table. This method can reduce disturbance in constrained motion and can improve the global force control performance.更多还原