Many studies have been performed on the position/force control of robot man
ipulators. Since the desired position and force required to realize certain
tasks are usually designated in the operational space, the controller shou
ld adapt itself to an environment and generate the control force vector in
the operational space. On the other hand, the friction of each joint of a r
obot manipulator is a serious problem since it impedes control accuracy. Th
erefore, the friction should be effectively compensated for in order to rea
lize precise control of robot manipulators. Recently, soft computing techni
ques (fuzzy reasoning, neural networks and genetic algorithms) have been pl
aying an important role in the control of robots. Applying the fuzzy-neuro
approach (a combination of fuzzy reasoning and neural networks), learning/a
daptation ability and human knowledge can be incorporated into a robot cont
roller. In this paper, we propose a two-stage adaptive robot manipulator po
sition/force control method in which the uncertain/unknown dynamic of the e
nvironment is compensated for in the task space and the joint friction is e
ffectively compensated for in the joint space using soft computing techniqu
es. The effectiveness of the proposed control method was evaluated by exper
iments.