This paper presents an approach, based on the qualitative theory of nonlinear dynamical systems, for the analysis and design of control systems for autonomous articulated vehicles. In particular, backing-up maneuvers are considered. These maneuvers are dangerous because the vehicles tend to `jackknife' when the angle between the tractor and the trailer is greater than a certain value. This danger is increased when the actuators are saturated because they can cause a lack of controllability in the state space and control becomes complicated. The paper proposes a two-level control system with an orientation controller in the lower level and a look-ahead controller in the higher. In the paper, a general orientation feedback controller is presented. This control strategy represents a wide family of different nonlinear controllers and can be applied with different particular control laws. The paper applies the qualitative theory of nonlinear dynamical systems to study the stability of the control system. Thus, the stability of the system under some constraints is demonstrated. The paper includes both simulation and experimental results of the implementation on the Romeo 4R autonomous vehicle, which is a full-size autonomous electrical golf cart with a trailer.
This paper presents an advanced model and control structure of a non-linear hybrid system: a gasoline direct injection (GDI) engine. The engine management system proposed in the paper is based on a hierarchical structure that uses Takagi-Sugeno fuzzy models of the engine to define piecewise linear controllers. Thanks to the hierarchical controller structure, engine idle speed regulation is tackled as a high-level requirement and therefore no specific low-level controller is needed. Moreover, the two combustion modes are correctly handled to optimize the engine efficiency. A global stability analysis of the GDI engine working either in homogeneous mode or in stratified mode is also presented, based on bifurcation analysis. Bifurcation analysis can be applied to a wide range of different controllers (even close-toproduction), and therefore it can be of significant practical value.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.