A variational approach to determination of maximum throw-able workspace of robotic manipulators in optimal ball pitching motion

This study is aimed at finding the entire points that a manipulator can launch an object onto by an optimal motion. These points are called throw-able workspace, which are located outside the reachable workspace of the robot. From an optimization point of view, some throwing parameters can be found to decrease motion cost. In this paper, by using this concept, the best combination of throwing and trajectory planning is attempted. The proposed method consists of two basic ideas: first, defining the optimal throwing problem as the optimal control problem (OCP) and solving it using the indirect solution approach based on the fundamental lemma of calculus of variations. To achieve the best release angle and speed, the throwing equation of motion is applied as a moving-end boundary condition (BC). Second, based on the obtained optimal throwing, an algorithm is presented to calculate the maximum throw-able workspace. The simulation results demonstrate the effectiveness of the proposed framework for both single link and spatial two-degree-of-freedom throwing robots.