Abstract
Typical solutions based on nonlinear constrained optimization-based strategies are hard to find and usually demand for higher level of computation. In this paper two techniques for transforming the initial nonlinear optimization into an approximate convex optimization are presented and tested for a rigid manipulator modeled with a feedforward neural network. The results have shown that the overall performance is enhanced when performing an approximate feedback hnearization.
Cite
CITATION STYLE
Botto, M. A., Te Braake, H. A. B., & Sá Da Costa, J. (1996). Solving nonlinear MBPC through convex optimization: A Comparative study using neural networks. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 1112 LNCS, pp. 593–598). Springer Verlag. https://doi.org/10.1007/3-540-61510-5_101
Register to see more suggestions
Mendeley helps you to discover research relevant for your work.
