Inequality-based Manipulator-Obstacle Avoidance Using the LVI-based Primal-dual Neural Network | IEEE Conference Publication | IEEE Xplore