During the interaction process of a manipulator executing a grasping task, to ensure no damage to the object, accurate force and position control of the manipulator’s end-effector must be concurrently implemented. To address the computationally intensive nature of current hybrid force/position control methods, a variable-parameter impedance control method for manipulators, utilizing a gradient descent method and Radial Basis Function Neural Network (RBFNN), is proposed. This method employs a position-based impedance control structure that integrates iterative learning control principles with a gradient descent method to dynamically adjust impedance parameters. Firstly, a sliding mode controller is designed for position control to mitigate uncertainties, including friction and unknown perturbations within the manipulator system. Secondly, the RBFNN, known for its nonlinear fitting capabilities, is employed to identify the system throughout the iterative process. Lastly, a gradient descent method adjusts the impedance parameters iteratively. Through simulation and experimentation, the efficacy of the proposed method in achieving precise force and position control is confirmed. Compared to traditional impedance control, manual adjustment of impedance parameters is unnecessary, and the method can adapt to tasks involving objects of varying stiffness, highlighting its superiority.