Bi-criteria joint torque minimization of redundant robot arms using LVI-based primal-dual neural network

To diminish the discontinuity and divergence of infinity norm torque minimization scheme, a bi-criteria weighting scheme is proposed for online redundancy resolution of redundant robot arms. Such a scheme can easily be reformulated into a quadratic program (QP) subject to equality, inequality and bound constraints. To solve this QP problem online, a primal-dual dynamical-system solver is further presented based on linear variational inequalities (LVI). Compared to previous research work, the adopted QP-solver has simple piecewise-linear dynamics, and does not entail real-time matrix inversion. Computer simulations are performed based on a PUMA560 manipulator to verify the performance and effectiveness of the proposed torque-optimization method.