Q.P. Chu

Fuzzy logic based full-envelope autonomous flight control for an atmospheric re-entry spacecraft

An intelligent and autonomous flight control system for an atmospheric re-entry vehicle is investigated, based on fuzzy logic control and aerodynamic inversion computation. A common PD-Mamdani fuzzy logic controller is designed for all the five re-entry flight regions characterized by different actuator configurations. A linear transformation to the controller inputs is applied to tune the controller performance for different flight regions while using the same fuzzy rule base and inference engine. An autonomous actuator allocation algorithm is developed, based on the aerodynamic inversion computation, to cover all the five actuator configurations with the same fuzzy logic controller. Simulation results of tracking both a bench mark trajectory and a given nominal re-entry trajectory are presented to evaluate the control performance.

Fuzzy logic based attitude control of the spacecraft X-38 along a nominal re-entry trajectory ☆

Abstract Fuzzy logic control techniques are investigated for applications in the intelligent re-entry flight control of the ESA–NASA crew return vehicle. Three PD-Mamdani fuzzy controllers are constructed to control the inner-loop attitude dynamics, simulated by a fully nonlinear 3 degree-of-freedom simulator of the CRV. Each controller uses an angle tracking error and its derivative to calculate a commanded control surface deflection of the simulator. The input-domains are partitioned with 5 membership functions, resulting in 25 fuzzy rules for each rule-base. The output-domains are partitioned with 9 membership functions. The Mamdani controllers use a standard max–min inference process and a fast center of area method to calculate the crisp control signals. Simulation results show the ability to track a reference trajectory with acceptable performance, though the real strength of a nonlinear fuzzy logic controller is yet to be proven with more demanding benchmark trajectories.

Atmospheric Reentry Modeling and Simulation

The analysis, being overpredictivebut simple and fast, was then applied to six other sabot models, as given in Figs. 9 and 10, to examine their relative lifting capability. These sabot models and dimensions were generated by the PRODAS code. These sabots are for both the 120and 105-mm calibers, as provided in Table 2. Note that sabot model 3 has four petals instead of the more common three-petal design. All calculations were made at Mach 4.5 and the same conditions.The resulting front bourreletnormal force on each petalmodel is shown in Fig. 11. Sabot models and their corresponding normal forces are provided in Table 2. These predicted forces are to be considered overestimated, possibly by a factor of about 1.6 (based on the CFD results for model 1). An actual test is recommended to be made to provide a validity for the CFD results as well. Meanwhile, the comparative results for the different sabot conx8e gurationsprovide some insight about the relative effectiveness of different front bourrelet designs in producing lift.

Inferring the driver's lane change intention using context-based dynamic Bayesian networks

This paper presents the results of the design and evaluation of a context based intent inference system for highway lane changes. Traditionally, intent inference for this task is based on information from the subject vehicle only. Because the context, e.g. traffic situation, plays an important role in the drivers situational awareness, a study was performed to investigate what results can be achieved by including this context information in the intent inference system. It was observed that traditional state based systems have a 90% accuracy and reach their maximum performance much earlier than a simple lateral threshold. The addition of various context information variables to the state based system decreased performance, caused by the models having problems capturing the complexity of the context data.

Online Adaptive Critic Flight Control using Approximated Plant Dynamics

A relatively new approach to adaptive flight control is the use of reinforcement learning methods such as the adaptive critic designs. Controllers that apply reinforcement learning methods learn by interaction with the environment and their ability to adapt themselves online makes them especially useful in adaptive and reconfigurable flight control systems. This paper is focused on two types of adaptive critic design, one is action dependent and the other uses an approximation of the plant dynamics. The goal of this paper is to gain insight into the theoretical and practical differences between these two controllers, when applied in an online environment with changing plant dynamics. To investigate the practical differences the controllers are implemented for a model of the general dynamics F-16 and the characteristics of the controllers are investigated and compared to each other by conducting several experiments in two phases. First the controllers are trained offline to control the baseline F-16 model, next the dynamics of the F-16 model are changed online and the controllers will have to adapt to the new plant dynamics. The result from the offline experiments show that the controller with the approximated plant dynamics has a higher success ratio for learning to control the baseline F-16 model. The online experiments further show that this controller outperforms the action dependent controller in adapting to changed plant dynamics