This study proposes an adaptive impedance force control using fuzzy neural networks (FNNs) for robot manipulator including actuator dynamics with uncertainties and operated in unknown environment. The system uncertainties are estimated by FNNs, and the corresponding adaptive impedance control is derived based on Lyapunov stability approach. Besides, the stiffness coefficient of contact environment is estimated by gradient method with convergent theory. The proposed approach does not calculate the regressor matrix which is a significant simplification in implementation. Simulation results are introduced to illustrate the effectiveness and performance of our approach.