FUZZY ADAPTIVE CONTROLLER-DESIGN FOR THE JOINT SPACE CONTROL OF AN AGRICULTURAL ROBOT

The proper execution of agricultural robotic tasks needs the use of ad aptive control techniques. This fact is mainly due to the nature of th e systems to control, which are difficult-to-model and time-varying sy stems. After a review of previous works concerning adaptive control, a solution using a fuzzy adaptive controller is studied for the joint c ontrol of such robots. An analytic representation of a particular fuzz y system is first developed to deduce useful conclusions for the contr oller design. Then, a specialized learning architecture is used to all ow the reconstruction of an error signal required for a gradient metho d for on-line modification of the consequent part of the inference rul es of a Sugeno's fuzzy controller. At the same time, a second level co nstituted by static rules (meta-rules) is introduced to cope with some limits of the learning architecture. Clustering of some rules is prop osed to be able to learn those that are not fired most of the time but essential for unusual robot motions. Thanks to this new structure, th e controller is dedicated to each meta-rule, and the number of rules w ith respect to a solution without meta-rule is considerably reduced. S imulation results during large on-line variations in system parameters derived from a typical example of an agricultural robot show the effe ctiveness of the proposed approach. The controller stability is verifi ed by using the so-called cell-to-cell mapping algorithm. Finally, the feasibility of the implementation of this algorithm in low-end hardwa re is shown. (C) 1998 Elsevier Science B.V. All rights reserved.