Artificial neural networks (ANNs) have recently become the focus of considerable attention in many disciplines, including robot control, where they can be used to solve highly nonlinear control problems. One of these ANN applications is that of the inverse kinematic problem, which is important in robot path planning. This paper outlines an iterative approach, using the offset error compensation method, to improve the accuracy of the inverse kinematic solution by using an ANN and the forward kinematic model of the robot. The method can be used as a feedforward position controller to generate accurate inverse kinematic information for general-purpose robotic manipulators.
PsaltisD.SiderisA.YamamuraA.A multi-layer neural networks controller. IEEE Control Systems Mag., April 1988, 17–21.
2.
KawatoM.UnoY.IsobeM.SuzukiR.A hierarchical model for voluntary movement and its application to robotics. In Proceedings of the First IEEE International Conference onNeural Networks, 1987, Vol. IV, pp. 573–582 (IEEE, New York).
3.
NguyenD. H.WidrowB.Neural networks for self-learning control system. IEEE Control Systems Mag., April 1990, 18–23.
4.
DenavitJ.HartenbergR. S.A kinematic notation for lower-pair mechanisms based on matrices. J. Appl. Mechanics, 1955, 77, 215–221.
5.
Arteaga-BravoF. J.Multilayer back-propagation network for learning the forward and inverse kinematics equations. In Proceedings of the International Joint Conference onNeural Networks, Washington D. C., 1990, Vol.2, pp. 319–322.
6.
GuezA.AhmadZ.Accelerated convergence in the inverse kinematics via multi-layer feedforward networks. In Proceedings of the International Joint Conference on Neural Networks, Washington D. C., 1989, Vol.2, pp. 341–347.
7.
GuoJ.CherkasskyV.A solution to the inverse kinematic problem in robotics using neural network processing. In Proceedings of the International Joint Conference onNeural Networks, Washington D. C., 1989, Vol.2, pp. 299–304.
8.
RumelhartD. E.HintonG. E.WilliamsR. J.Learning internal representations by error propagation. In Parallel Distributed Processing, 1986, Vol. 1, Ch. 8 (MIT Press).
9.
LouY. F.BrunnP.Using artificial neural networks for solving engineering problems. In Proceedings of the 31st International MATADOR Conference, 1995, pp. 635–640 (Macmillan).
10.
FunahashK.On approximate realization of continuous mappings by neural networks. Neural Networks, 1989, 2, 183–192.
FuK. S.GonzalesR. C.LeeC. S. G.Robotics: Control, Sensing, Vision, and Intelligence (McGraw-Hill, New York).
16.
VAL-II User's Manual, 1983 (Unimation Inc.).
17.
LeeC. S.G.ZieglerM.A geometric approach in solving the inverse kinematics of PUMA robots. IEEE Trans. Aerospace Electronic Systems, 1984, 20, 695–706.
18.
LouY. F.Non-conventional computing methods for manufacturing robot control. PhD thesis, Department of Mechanical Engineering, University of Manchester Institute of Science and Technology, 1996.
