Dynamic control allocation for a stratospheric airship with redundant control systems

This paper proposes a dynamic control allocation method based on constrained optimal quadratic programming for the stratospheric airship with redundant systems. To ensure a given control command is well distributed among an available set of actuators, the control system is divided into two levels. A high-level controller is based on dynamic-inverse method to derive the pseudo-control law for the low-level controller. The control input of the actuators can be obtained by two principles, one is keeping the control input close to the desired steady-state value, and the other is minimizing the change between the current control input and the preceding one as well as the one before. This approach considers both physical constraints and dynamic responses of actuators. In addition, when a fault occurs, it can change the weight coefficient of the actuator to ensure that the fault is well distributed among the fault-free actuators without reconfiguring the high-level controller. Several simulations illustrate the feasibility of the control law and show advantages for practical application.