Viability and feasibility of constrained kinematic control of manipulators
Planning and control algorithms make an increasing use of optimization-based techniques, such as model predictive control. In case of manipulators with bounded joint positions, velocities, and accelerations, feasibility cannot be guaranteed if only box constraints (i.e. maximum and minimum values of the variables) are considered.
Additional constraints need to be added to shrink the set of admissible velocities and positions to a viable set.
We propose a method to construct a viable convex polyhedron that ensures feasibility of the optimal control problem by means of a given number of linear constraints.
Reference: Faroni, M., Beschi, M., Pedrocchi, N., and Visioli, A. (2018). Viability and feasibility of constrained kinematic control of manipulators. Robotics, 7(3):1–19.