An self-tuning Pl control with anti-windup scheme for the force regulation in robotic machining process is proposed in this paper. The experimentally identified robot dynamic model and nonlinear process model are used in the control algorithm design and simulation. Both the PI controller gains and the exact linearization procedure for the nonlinear process model are continuously tuned based on the on-line identified process parameters. Simulation results show that the proposed adaptive control scheme maintains the control system stability and performance in a wide range of operating conditions. The proposed algorithm was also verified via robotic machining experiments. © 2007 IEEE.