A proposal of manipulability based model predictive control for the parallelogram linkage

This paper proposes a control method for a parallelogram linkage robot manipulator. Parallelogram linkage can generate motion control with high precision. However, this technique is geometrically constrained in the task space as a result of its mechanical properties. The proposed method uses a model predictive controller to account for the geometric constraints of the system. The key concept behind the controller developed here is to use manipulability for the cost function and geometric constraints in the model predictive controller. Thus, we can design a trajectory-tracking controller using a manipulator and accounting for the geometric constraints. In the simulation results, the effectiveness of the proposed method was verified using the singularity avoidance problem as an example of geometric constraints.