This paper addresses the problem of controller design for constrained robots. A control algorithm, described as a combination of the direct adaptive and the variable structure method, is presented for the trajectory tracking of an end-effector on a constrained surface with specified constraint forces. This scheme has certain advantages, both with respect to computational efficiency and with respect to design. With respect to computational efficiency, the calculation of the regressor is avoided. With respect to design, since the number of parameter update laws required in the adaptation is independent of the number of links of the robot, the difficulty of tuning numerous adaptation gains is avoided.