Hybrid vehicles are an important technology for reducing oil use and transportation-related emissions. It is well-known that hybrid and electric vehicles are often designed and tested using standard cycles such as the Highway Fuel Economy Test (HWY), Urban Dynamometer Driving Schedule (UDDS), and the US06 Supplementary Federal Test Procedure (US06). However, this begs the questions: How does real world driving compare to these cycles? Can a vehicle be designed using real world driving data which saves fuel in the real world compared to a vehicle designed using standard cycles? This thesis investigates this issue using a set of 5000km of real world driving data by light-duty pickup trucks, with the goal to optimize the fuel savings of a mild hybrid truck. The challenge with using a model-based design approach on thousands of kilometers of real driving data is the long model run-time required to iterate through plant and control parameters. Thus, this work develops a novel script which reduces optimization time by 78%. The key is to run the full model of the non-hybrid truck one time on the full driving data set, and then use the resulting vehicle speed, engine efficiency, engine torque, and engine speed, as inputs to the faster script. The script is then used to quickly iterate through the driving data set many times to find optimal control and plant parameters. In this work, exhaustive search is used; however, evolutionary optimization algorithms could also be used and would benefit from the fast script iteration on real world driving cycles. Overall, the use of the real world driving set for design of the mild hybrid truck resulted in a 7.10% decrease in fuel consumption compared to the non-hybrid truck, while the use of standard driving cycles for design resulted in a 5.45% fuel consumption decrease compared to the non-hybrid truck. / Thesis / Master of Applied Science (MASc)

This thesis studies the optimal control of vehicular systems, focusing on the solution of minimum-lap-time problems for a Formula 1 car. The basic optimal control theory is summarised as an infinite-dimensional extension of optimisation theory. The relevant numerical techniques for optimisation and integral approximation are compared in view of the application to vehicle systems. The classical brachistochrone problem is revisited from an optimal control perspective, with two vehicle-relevant generalisations. Closed-form solutions are derived for both the optimal trajectory and transit time. Problems involving a steerable disc rolling on the interior surface of a hemisphere are studied. For three-dimensional problems of this type, which involve rolling bodies and nonholonomic constraints, numerical solutions are used. The identification of 3D race track models from measured GPS data is treated as a problem in the differential geometry of curves and surfaces. Curvilinear coordinates are adopted to facilitate optimal control solutions. The track is specified in terms of three displacement-dependent curvatures and two edge variables. The differential model is smoothed using numerical optimal control techniques. The Barcelona track is considered as an illustrative example. The minimum-lap-time problem for a Formula 1 car on a flat track is solved using direct transcription. The driven line and multiple car setup parameters are optimised simultaneously. It is shown that significant lap-time reductions can be obtained from track-specific setup parameter optimisation. Reduced computing times are achieved using a combination of analytical derivatives, model non-dimensionalisation and problem scaling. The optimal control of the car on a 3D track is studied; the results are compared with flat-track solutions. Contemporary kinetic energy-recovery systems are studied and compared with future hybrid kinetic-thermal energy-recovery systems. It is demonstrated that these systems can produce contemporary lap time using approximately two-thirds of the fuel required by present-day vehicles.

Due to the conditions peculiar to underwater, distinctive approaches are required to solve the navigation and path planning problem of an unmanned underwater vehicle (UUV). In this study, first of all, a detailed 6 degrees-of-freedom (DOF) mathematical model is formed, including the coupled non-linear forces and moments acting on an underwater vehicle. The hydrodynamic coefficients which correspond to the geometry of the vehicle which the model is based on are calculated using the strip theory. After the mathematical model is obtained, by applying appropriate linearization on the model, &ldquo / Linear Quadratic Regulator (LQR)&rdquo / control method is implemented to govern the surge, heave, pitch and yaw motions of the underwater vehicle. Path planning algorithm of the vehicle is based on tracking the waypoints. Permutation of the waypoints is obtained by solving the &ldquo / Travelling Salesman Problem (TSP)&rdquo / via genetic algorithm. Linked with that, &ldquo / Rapidly-Exploring Random Trees (RRT)&rdquo / algorithm is introduced into the path planning algorithm to avoid collisions in environments with obstacles. Underwater navigation solution is based on the &ldquo / Inertial Navigation System (INS)&rdquo / outputs, located on the vehicle. To correct the long-term drift of the inertial solution, &ldquo / Kalman Filter&rdquo / based integration algorithm is used and external aids such as &ldquo / Global Navigation Satellite System (GNSS)&rdquo / , &ldquo / Ultra-Short Baseline (USBL)&rdquo / acoustic navigation system and attitude sensors have been utilized. The control method, path planning and navigation algorithms used in this study are verified by simulation results.

This thesis is focused on individual wheel actuators in road vehicles intended for vehicle motion control. Particular attention is paid to electro-mechanical actuators and how they can contribute to improving vehicle dynamics and safety. The employment of individual wheel actuators at the vehicle's four corner results in a large degree of over-actuation. Over-actuation has a potential of exploiting the vehicle's force constraints at a high level and of controlling the vehicle more freely. One important reason for using over-actuated vehicles is their capability to assist the driver to experience the vehicle as desired. This thesis demonstrates that critical situations close to the limits can be handled more efficiently by over-actuation. To maximise the vehicle performance, all the available actuators are systematically exploited within their force constraints.  Therefore, force constraints for the individually controlled wheel are formulated, along with important restrictions that follow as soon as a reduction in the degrees of freedom of the wheel occurs. Particular focus is directed at non-convex force constraints arising from combined tyre slip characteristics. To evaluate the differently actuated vehicles, constrained control allocation is employed to control the vehicle. The allocation problem is formulated as an optimisation problem, which is solved by non-linear programming. To emulate realistic safety critical scenarios, highly over-actuated vehicles are controlled and evaluated by the use of a driver model and a validated complex strongly non-linear vehicle model. it is shown that, owing to the actuator redundancy, over-actuated vehicles possess an inherent capacity to handle actuator faults, with less need for extra hardware or case-specific fault-handling strategies. / QC 20100722

Multi-body system dynamics is one of the most important theoretical achievementsin mechanics. With the development of the theory, corresponding commercialsoftware packages have been developed and are used for modellingand simulation of complicated large systems, such as air planes and vehicles.This kind of virtual prototypes can be used for studies and assessments ofreal systems even before the real systems are built. As a result, the high costprototype building and prototype testing can be saved, so as the time can bereduced. This is just the demand of modern industry. This theory can beapplied on the vehicle-virtual road interaction study which has been used inthis thesis.This thesis suggests a target velocity prole for a heavy vehicle which driveson tough road. Having uneven and hilly road, actual driving conditions arechanged as the driver runs the vehicle. Drivers can perceive the road conditionwith their visual organ and sense of balance and then they control theirvehicles more safely by re ecting various conditions of this target velocityprole. Without this process, the driving-stabilization on slope and twistingroads would fall considerably, and the problem could be directly connected tooverturning. This thesis, moreover, will show how to acquire the road data,extract the velocity prole, and verify the performance of the suggested velocityprole through virtual road test.In vehicle-virtual road interaction simulation, multi body system (MBS) dynamicswith software Adams has been employed to model an articulatedhauler. The simulation has been validated by velocity prole test data andcompared to the former velocity prole. This method can be used for estimatingthe eects of dynamic forces on the frame so that the load design canbe assessed in vehicle design process.This project is in collaboration with Volvo Construction Equipment AB,Braas, Sweden.

The thesis presents an intermediate modelling approach to study transient behaviour of vehicle systems, with emphasis put on simplified yet accurate representation of important system elements. A representative non-linear vehicle model is developed in MA TLAB/Simulink environment, where non-linear characteristics of tyre, suspension and braking system are included to capture the dynamic behaviour of a vehicle under transient conditions. The novel aspect of this work is the application of a representative full vehicle-tyre-ABS integrated set-up to study the complicated interaction between tyre and anti-lock braking, under a range of demanding operating conditions, including combined cornering and braking. The modelling methodology involves development of low end vehicle models, based on the Newton-Euler formulation. Subsequently, an intermediate vehicle model is devised, where more details are incorporated such as additional DOF to capture the sprung mass motion in space, along with its non-linear interactions with the un-sprung masses, large angle effects, kinematics of steering/wheels and an appropriate tyre model suitable for transient manoeuvres. Particular attention is paid to the suspension system modelling, through inclusion of non-linear effects in springs, dampers, bump-stops, and anti-roll bars, along with the jacking and anti-dive effects using the virtual work method. The model also incorporates a hydraulic brake model, based on the reduced order brake system dynamics for realistic simulation of the braking manoeuvres. A complex multi-body ADAMS/Chassis model, with much greater level of detail, has also been established to extensively compare and enhance the realistic behaviour of the intermediate vehicle model. During the simulation exercise, the intermediate vehicle model has shown good agreement with the complex ADAMS model, thus justifying the accurate representation of vehicle.non-linear characteristics, particularly the suspension system. The realistic behaviour of the vehicle model is further ascertained with a reliable GPS enabled test vehicle, by performing number of manoeuvres on test tracks, including combined cornering and braking. A representative 4-channel conventional ABS system is modelled and integrated in the intermediate vehicle model. The ABS adopts generic peak seeking approach, employing wheel deceleration and brake slip as control variables. External braking inputs, in form of stepped pressure pulses, are also separately used to represent the transient braking system dynamics. In the current work, different transient tyre models based on the single point contact approach and using Magic Formula steady-state characteristics are applied, while studying the influence of their dynamic behaviour on the ABS system. By employing a representative ABS system in a multi-body vehicle model and considering the particularly demanding situation of combined braking I cornering, it is shown that the models which are adequate for pure braking might struggle when the complicated full vehicle dynamics are excited. It is shown that the first order relaxation length approach may not be sufficient to fully satisfy the requirements of an ABS braking, especially if the relaxation length is not modelled as a variable dependent on tyre slip. In comparison, the modelling approach, where the carcass compliances and contact patch properties are explicitly represented, can handle the oscillatory tyre behaviour associated with ABS braking, in a far more accurate manner. In comparison to the earlier studies, which were mostly conducted for straight-line braking, this thesis stresses the fact that the tyre behaviour can be influenced by the complex interaction of handling and braking, and hence the effect should be captured while investigating or evaluating the performance of a tyre model in relation with ABS simulation.

Mine haul road maintenance is traditionally done at scheduled intervals or after regular inspection. Both these methods can lead to unwarranted expenditure, either through over-maintaining the road, or failure to recognise significant deterioration, resulting in an increase in vehicle operating costs. Predictive maintenance management models for unpaved roads have been developed in recent years. These methods work well in a trivial environment where variables such as traffic volume can be predicted. However, many mining systems are too complex for such models to be effective. This work investigates the possibility of using haul truck response to aid haul road maintenance management. The approach adopted for the study was twofold: Firstly, can truck response data be used to recognise specific road defects, in terms of location, type and size? This is important since different defect types require different road maintenance strategies. Secondly, can road roughness be measured on a qualitative basis? With the emphasis on road defect reconstruction, a mathematical modelling approach was adopted. The truck was characterised in terms of its suspension and tyre properties. Dynamic truck response data was acquired during field measurements in which the vehicle was driven over defects of known dimensions. With these data sets available, mathematical modelling and simulation was possible. Quarter vehicle and seven degree of freedom vehicle models played a vital role in this work by laying a foundation in the use of haul truck response for the purpose of road defect reconstruction. A modelling methodology that is based on dynamic equilibrium of an independent front unsprung mass of the truck is proposed in which the vertical dynamic tyre force and eventually the road geometry is calculated. It is shown that defects can be reconstructed from measured truck response data with an accuracy sufficient to fulfil the requirements of defect recognition for road maintenance management purposes. Secondly, a preliminary investigation into the qualitative assessment of road condition via truck response measurements was conducted. The inherent response properties of the truck pertaining to road roughness measurement were studied and some correlation between measured suspension motion and road roughness measured with a high speed profilometer was found. / Dissertation (MEng (Mechanical))--University of Pretoria, 2005. / Mechanical and Aeronautical Engineering / MEng / Unrestricted

No description available.

Cette thèse traite de la synthèse et de la réalisation d’un intégrateur d’ordre non entier borné en fréquence. La réalisation est faite par un réseau constitué d’un faible nombre de cellules capacitives et dissipatives. La première partie de ce mémoire s’attache à développer des méthodes permettant de déterminer les paramètres physiques des éléments du réseau à partir des quatre paramètres de haut niveau qui caractérisent l’intégrateur d’ordre non entier à réaliser. Les spécificités liées à une réalisation en technologie hydropneumatique sont détaillées. Il est montré, dans un contexte d’isolation vibratoire, qu’elles conduisent à des performances remarquables de robustesse du degré de stabilité et de robustesse de la rapidité vis-à-vis des variations de la masse suspendue, et ce, malgré l’existence de non-linéarités. Les non-linéarités sont étudiées à l’aide des séries de Volterra. La seconde partie est consacrée à l’application au secteur de l’automobile des résultats de la première partie. La synthèse et la réalisation d’une suspension CRONE hydractive, suspension multi-états dont le mode souple assure la robustesse du degré de stabilité de la caisse vis-à-vis des variations de la masse suspendue, sont proposées et validées en simulation sur un modèle de véhicule à 14 degrés de liberté. / The thesis deals with the synthesis and the realisation of a band limited fractional differentiator. The realisation is made thanks to a small number of resistive and capacitive cells (RC cells). The first part of this thesis is about some new methods to compute the physical parameters of the RC cells from the 4 high-level parameters of the band limited fractional differentiator. The specificities of a realisation using hydropneumatic technology are detailed. It is shown that, in vibration isolation, they lead to remarkable performances. The stability degree robustness and the rapidity robustness towards the variation of the sprung mass value are obtained in spite of non- linearities. Volterra serie expansion is used to study the non-linearities. The second part is about the application of the previous results to the automotive field. The design and the realisation of an hydractive CRONE suspension is proposed. An hydractive CRONE suspension is a suspension with several operating modes and which allows to obtain the stability degree robustness. The hydractive CRONE suspension is then test with a 14 degrees of freedom model of a car.