Adaptive Stability Augmentation and Model Predictive Trajectory Tracking Control for Flight Systems

• Applicants: Professor Dr.-Ing. Florian Holzapfel; Professor Dr.-Ing. Boris Lohmann
• Subject Area: Traffic and Transport Systems, Intelligent and Automated Traffic ; Automation, Mechatronics, Control Systems, Intelligent Technical Systems, Robotics
• Term: from 2014 to 2018
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 255465855

Flight control systems are mostly of cascaded structure with an inner stability augmentation system (control) and an outer trajectory tracking control (guidance). Herein, trajectory planning and control algorithms are designed in a way that they can fulfill the control task if the properties of the stabilized plant are as they were assumed in the design process. In the case of failure and damage, the plant degrades, so that: - the plants dynamics do not correspond anymore to the plant assumed in control design and the inner control loop can thus only stabilize the system if the amount of damage lies within its bounds of robustness - the safe flight envelope of the degraded system transforms to an unknown subset of the nominal plants safe flight envelope - the closed loop properties of the degraded plant with baseline control are different (or usually unknown) with respect to the underlying model of the trajectory control in a way that the desired trajectory is not achievable for the system at all or the trajectory controller cannot fulfill its taskIt is therefore proposed to:- identify the residual dynamics of the degraded system and its safe flight envelope - adapt the desired dynamics of the inner control loop to the remaining plant capabilities - actually reach the desired dynamics by means of adaptive control- perform the path planning in real-time and with guaranteed convergence based on the adjusted reference model - track the planned trajectoryThe main innovative feature of the planned approach is the problem separation with the help of physically motivated, nonlinear reference models into two subtasks:- Reference model adaption based on the identified dynamics and reference tracking by an adaptive control system (Institute of Flight System Dynamics) - Nonlinear model predictive control (NMPC) for online trajectory planning with guaranteed feasibility and cycle time (Institute of Automatic Control)In doing so, the adaptive control system handles mainly the rotation and attitude dynamics while the NMPC approach is allocated to translation and position dynamics. In the process of the research project it shall be accounted for aviation relevant constraints. Most notably, the formal prove of guaranteed performance is of high importance. The innovative approach's practical viability will be demonstrated with a six-rotor flight system (hexacopter).

• DFG Programme: Research Grants