DESIGN OF A FUZZY-LOGIC BASED ROBOTIC ADMITTANCE CONTROLLER

A fuzzy logic based admittance control approach is proposed for the co ntrol of robotic end-effector forces occurring during a typical automa ted robotic manufacturing task. The proposed admittance control approa ch provides the necessary nonlinear control actions required in a typi cal automated robotic manufacturing task, and at the same rime enables the incorporation of existing knowledge obtained from the process ope rator, in the design of the controller. This can significantly reduce the controller development time. The robotic deburring task is used as an example of a typical manufacturing task, although the technique ca n be easily extended to other manufacturing tasks. Automated robotic d eburring offers an attractive alternative to manual deburring in terms of reduced costs and improved quality of the finished parts. The appr oaches proposed for control of deburring using conventional control te chniques require an accurate process model and hence are not ideally s uited for control under conditions of uncertainty in the available inf ormation about the process. FUZZY logic control techniques offer an al ternative to conventional control methods used for the deburring task. A fuzzy logic rule base is designed, using the knowledge obtained fro m the deburring operator, to control the deburring robot in the presen ce of uncertainties in the burr size and location information. The fuz zy logic controller issues corrections to the nominal robot trajectory based on the burrs encountered. The corrected robot trajectory is the n input to a robot positional controller. Simulation results are prese nted to demonstrate the effectiveness of the proposed fuzzy logic base d admittance control scheme in controlling the automated robotic debur ring operation.