提出一种由神经网络训练模糊控制规则的自适应模糊控制器，并应用于附加力外环的机器人力／位置控制。在不改变一般工业机器人原有位置控制器的前提下，实现力／位置自适应模糊控制。实验结果表明，该方法可使机器人控制系统对工作环境接触刚度的自适应能力得到显著改善。 An adaptive fuzzy controller trained by neural network to fuzzy control rules was proposed and applied to the robot force / position control with additional force outer ring. Under the premise of not changing the original position controller of the general industrial robot, the force / position adaptive fuzzy control is realized. The experimental results show that this method can significantly improve the adaptability of the robot control system to the contact stiffness of the working environment.