中文摘要：

提出了一种新的面向未知环境的智能预测算法，并将此算法应用于机器人力跟踪控制中，该方法利用机械手束端与未知受限环境产生的接触轨迹，通过模糊推理智能地预测阻抗控制模型中的参考轨迹，并根据力误差变化用参考比例因子对其进行调节，以适应未知环境刚度的变化．通过对阻抗模型参数进行模糊调节减少受限运动中的力误差，提高了全局的力控制效果．仿真结果证明了此算法的有效性。

英文摘要：

A novel intelligent predictive algorithm is proposed for unknown enviroments, and is applied to robotic force tracking control. According to the contact trajectory between the end-effector of the robot manipulator and the unknown constraint environment, the reference trajectory of the impedance controller is predicted intelligently by fuzzy reasoning. Meanwhile, it is tuned by reference scale factor depending upon the change of force error to accommodate change of the unknown stiffness. Moreover, impedance model parameters are adjusted fuzzily to reduce the force errors in the constrained motion and improve the global force control performance. Simulation results show the effectiveness of the scheme.