Modeling and longitudinal control of commercial heavy vehicle equipped with variable compression braking

Over the last ten years, there has been a significant improvement of the reliability and efficiency of the commercial heavy vehicle powertrain. This transformation is primarily achieved by using lightweight material, by reducing aerodynamic drag and frictional losses, and by employing efficient diesel engines. This contributes to the increased fuel economy associated with diesel engines, but on the other hand, it does not provide means for decelerating the moving engine parts. An increase in operational vehicle speed combined with decrease of the natural retarding capabilities in modern powertrains creates challenging braking requirements. Engine and truck manufacturers respond to the need by developing additional retarding mechanisms.; In this dissertation, we develop models and longitudinal speed controllers for heavy commercial vehicles equipped with a variable compression braking mechanisms. In particular, we develop a detailed crankangle based simulation model for a six cylinder, 350 hp diesel engine equipped with a compression brake using a zero-dimensional modeling approach in addition to physical equations and static engine maps provided by the manufacturers. The model is capable of describing the intrinsic interactions between individual cylinder intake and exhaust processes, turbocharger dynamics during combustion and braking modes, and the transition between those modes.; Based on averaging and identification of the instantaneous torque response for changes in brake valve timing and fuel flow, we derive a low-order control oriented engine model. This work bridges the gap between the detailed crankangle based models developed in the engine design community, and the first order lag representation of diesel engine torque response used in the vehicle dynamics community.; Moreover, we integrate the compression brake actuator with the service brakes and design a PI-controller that emulates the driver's actions on long grades. The controller uses the engine speed measurement to activate the service brakes only when retarding power of the compression brake is insufficient. We also design a controller that is robust to variations in vehicle mass and road grade using robust control methodologies. Finally, we compare the robust controller performance with the performance of an adaptive control scheme.