Quasi Time-Optimal Trajectory Generation for Pneumatic Drives Considering their Actuator Dynamics and Constraints

Pneumatic drives are commonly used in automation technology, where their motion must follow certain reference trajectories. To maximize productivity, these trajectories should be as fast as possible and at the same time still feasible to track. The limiting factors therein are that the pressure dynamics are not negligibly fast, the air mass flow through the control valves is subject to pressure-dependent constraints, and the dynamics of the mechanical and pneumatic subsystems are coupled to each other. The goal of this work is to generate quasi time-optimal trajectories for pneumatic drives considering all the aforementioned dynamics and constraints in a model-based way. As a foundation, it is first analyzed how the actuator dynamics and nonlinear state-dependent constraints affect the motion of the drive. Then, the quasi time-optimal control problem for trajectory generation is formulated and solved numerically offline. The resulting trajectories are validated through experiments on the drive. The experimental outcomes show that the trajectories are both dynamically feasible and utilize the available control input efficiently.