An aero-engine optimal steady state controller based on reinforcement learning neural networks (NNs) was proposed in this paper. The presented reinforcement learning NNs can achieve the optimal control objective by constructing two interconnected modules (i.e. action module and critic module). For the state variable models of small perturbation on steady operating points, the double-variable control of an aero-engine is accomplished by two similar backing propagation (BP) NNs. The simulation results show that the presented controller has the perfect performance with the smooth transition process. It not only has strong anti-interference ability and adaptability, but also has excellent robustness to the change of aero-engine model parameters.

The problem of uncertain homing missile guidance is considered. The guidance control is designed with the state feedback of guaranteed cost control. On the basis of LMI, the robust stabilizing controller is presented, and a guidance algorithm is given which can successfully guide the missile to the target. Meanwhile, the upper bound of system performance is given. For a portable homing missile, the ideal trajectories are protracted. Simulations show the algorithm can successfully guide the missile to head-on or tail-on the maneuver target. Furthermore, for the same target, comparisons are carried out in the case of the different missile launching site. It is illustrated that the designed guidance algorithm is not only robust to disturbance but also small normal overload especially on the encountering time.

In order to strengthen the hovering precision of a small unmanned helicopter with a manipulator for completing a hovering operation, in this study, a double-loop control strategy of position and attitude is designed. In the position loop, a sliding mode controller is proposed. It attains high precision and strong robustness under continuous disturbance. With the manipulator continuously operating during helicopter hovering, the overall center of gravity and the moment of inertia of the system will also change. Therefore, a backstepping sliding mode controller is proposed in the attitude loop for controlling the hovering attitude of the helicopter. The high precision and strong robustness of the control system are proven. The digital simulation presents a position steady-state error of less than 2% under constant disturbance, and the hovering attitude angle fluctuation amplitude of the helicopter is less than 0.05 rad.

In this research, the focus is on the adaptive fuzzy neural networks (NNs) control of the single-link flexible robot manipulator, which achieves accurate trajectory tracking and improves the speed of position tracking. Firstly, in order to establish a single-link flexible model, the lumped spring-mass method and Lagrangian equation are used. Secondly, a full-state feedback control is proposed. Aiming at achieving the control objective, the Lyapunov stability method proves that the closed-loop system is uniformly ultimately bounded. Finally, Numerical simulations are carried out to verify the performance of the fuzzy NNs control. Furthermore, the control performance is compared with the proportion derivative control.