In this study, recursive Newton-Euler sensitivity equations are derived for robot manipulator motion planning problems. The dynamics and sensitivity equations depend on the 3 × 3 rotation matrices based on the moving coordinates. Compared to recursive Lagrangian formulation, which depends on 4 × 4 Denavit-Hartenberg (DH) transformation matrices, the moving coordinate formulation increases computational efficiency significantly as the number of matrix multiplications required for each optimization iteration is greatly reduced. A two-link manipulator time-optimal trajectory planning problem is solved using the proposed recursive Newton-Euler dynamics formulation. Only revolute joint is considered in the formulation. The predicted joint torque and trajectory are verified with the data in the literature. In addition, the optimal joint forces are retrieved from the optimization using recursive Newton-Euler dynamics.