Premium
Two/Infinity Norm Criteria Resolution of Manipulator Redundancy at Joint‐Acceleration Level Using Primal‐Dual Neural Network
Author(s) -
Zhang Yug,
Cai BingHuang,
Yin JiangPing,
Zhang Lei
Publication year - 2012
Publication title -
asian journal of control
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.769
H-Index - 53
eISSN - 1934-6093
pISSN - 1561-8625
DOI - 10.1002/asjc.485
Subject(s) - weighting , piecewise , mathematical optimization , artificial neural network , mathematics , variational inequality , norm (philosophy) , quadratic programming , control theory (sociology) , redundancy (engineering) , computer science , algorithm , artificial intelligence , medicine , mathematical analysis , control (management) , political science , law , radiology , operating system
A two/infinity norm criteria (termed bi‐criteria) weighting scheme is proposed in this paper for resolving manipulator redundancy at the joint‐acceleration level. This bi‐criteria scheme is aimed at remedying discontinuity‐points and torque‐instability problems which arise in pure infinity‐norm acceleration‐minimization schemes. By incorporating joint physical limits, the proposed bi‐criteria redundancy‐resolution scheme can finally be formulated as a quadratic program ( QP ) subject to equality constraint, inequality constraint and bound constraint simultaneously. As a real‐time QP solver with simple piecewise‐linear dynamics and higher computational efficiency, the primal‐dual neural network based on linear variational inequalities ( LVI ) is presented in this paper to solve online such a bi‐criteria weighting scheme. Computer‐simulation results based on a PMUA 560 robot arm illustrate the advantages and efficacy of such a neural weighting scheme.