Receding horizon control of a 3 DOF helicopter using online estimation of aerodynamic parameters

Abstract

This study presents a numerical implementation of fast nonlinear model predictive control (NMPC) and nonlinear moving horizon estimation (NMHE) for the trajectory tracking problem of a 3 degree of freedom (DOF) helicopter. The motivation behind using the NMPC instead of its linear counterpart is that the helicopter is operated over nonlinear regions. Moreover, this system has cross-couplings that make the control of the system even more complicated. What is more, according to our simulation scenario, the system has a time-varying dynamical model because it has time-varying parameters which are estimated online using NMHE and the extended Kalman filter (EKF) throughout the control. Although NMHE is computationally more demanding, its capability of incorporating the constraints encourages us to utilize NMHE rather than EKF. Two reference trajectories, namely, sinusoidal and square-like, are tracked, and owing to the better learning capability of NMHE over EKF, the NMPC-NMHE closed-loop control framework is able to track both reference signals with more accuracy than the NMPC-EKF control framework, even under parameter uncertainties. Thanks to the ACADO toolkit, the combined average execution time is 4 milliseconds, demonstrating the potential of the proposed framework for real-time aerospace applications using relatively cheaper processors.

Practical control implementation of tri-tiltRotor flying wing unmanned aerial vehicles based upon active disturbance rejection control

“…In the lateral cascade channel, the lateral offset reference, y c , is the reference input of the y lateral offset channel, and the intermediate variable, c , is regarded as the reference input of the roll angle, , channel. From the lateral state equations (17) and (18), the relative equations can be rewritten as…”

“…Second, the required calculation time and the computing resources available on autopilot inhibit these optimization solutions. In Papachristos et al., 12,13 Alexis et al., 14 Kamel et al., 15 Liu et al., 16 and Mehndiratta and Kayacan, 17 a powerful model predictive control method was designed to deal with the issues of path tracking based on commercial autopilot for thrust-vectoring of a TTR–UAV, but the method’s calculation cost can prove prohibitive. The UAV proposed in Fang et al.…”

Adaptive model predictive control design using multiple model second level adaptation for parameter estimation of two‐degree freedom of helicopter model

“…Many difficulties have been arrived to design suitable control algorithms for unmanned helicopter due to its high nonlinearity, inter‐axis cross‐coupling, unstable system dynamics, and parametric uncertainty during the last two decades 3‐5 . Active research is continuing to develop controllers for UAVs, which can work successfully in spite of all difficulties 5‐7 …”

“…Recently, a learning‐based model predictive control (LBMPC) technique has been developed in Reference 20, where a modified extended Kalman filter (EKF) is used to perform state estimation and learn updated model parameters. Whereas, in References 7,21‐23, a nonlinear moving horizon estimation has been used for nonlinear system to improve the performance of LBMPC.…”

Neural network-based adaptive fault tolerant tracking control for unmanned autonomous helicopters with prescribed performance

“…During the past decades, considerable research and increasing attention have been done on the design, analysis, and operation of the unmanned autonomous helicopters (UAHs) due to their widely application prospects in the field of military and civilian. [1][2][3] However, unlike other mechanical systems, the UAHs are special controlled objects with the characters of underactuation, multivariate, strong coupling and nonlinearity. 4 Especially for the medium-scale UAHs, who possess the merits of fast speed, high altitude, long cruise and large payload compared with the small-scale UAHs, it is still a challenge to design effective and robust controller for them.…”