Integration of Active Systems for a Global Chassis Control Design

Vehicle chassis control active systems (braking, suspension, steering and driveline), from the first ABS/ESC control unit to the current advanced driver assistance systems (ADAS), are progressively revolutionizing the way of thinking and designing the vehicle, improving its interaction with the surrounding world (V2V and V2X) and have led to excellent results in terms of safety and performances (dynamic behavior and drivability). They are usually referred as intelligent vehicles due to a software/hardware architecture able to assist the driver for achieving specific safety margin and/or optimal vehicle dynamic behavior. Moreover, industrial and academic communities agree that these technologies will progress till the diffusion of the so called autonomous cars which are able to drive robustly in a wide range of traffic scenarios. Different autonomous vehicles are already available in Europe, Japan and United States and several solutions have been proposed for smart cities and/or small public area like university campus. In this context, the present research activity aims at improving safety, comfort and performances through the integration of global active chassis control: the purposes are to study, design and implement control strategies to support the driver for achieving one or more final target among safety, comfort and performance. Specifically, the vehicle subsystems that are involved in the present research for active systems development are the steering system, the propulsion system, the transmission and the braking system. The thesis is divided into three sections related to different applications of active systems that, starting from a robust theoretical design procedure, are strongly supported by objective experimental results obtained fromHardware In the Loop (HIL) test rigs and/or proving ground testing sessions. The first chapter is dedicated to one of the most discussed topic about autonomous driving due to its impact from the social point of view and in terms of human error mitigation when the driver is not prompt enough. In particular, it is here analyzed the automated steering control which is already implemented for automatic parking and that could represent also a key element for conventional passenger car in emergency situation where a braking intervention is not enough for avoiding an imminent collision. The activity is focused on different steering controllers design and their implementation for an autonomous vehicle; an obstacle collision avoidance adaptation is introduced for future implementations. Three different controllers, Proportional Derivative (PD), PD+Feedforward (FF) e PD+Integral Sliding Mode (ISM), are designed for tracking a reference trajectory that can be modified in real-time for obstacle avoidance purposes. Furthermore, PD+FF and PD+ISM logic are able to improve the tracking performances of automated steering during cornering maneuvers, relevant fromthe collision avoidance point of view. Path tracking control and its obstacle avoidance enhancement is also shown during experimental tests executed in a proving ground through its implementation for an autonomous vehicle demonstrator. Even if the activity is presented for an autonomous vehicle, the active control can be developed also for a conventional vehicle equipped with an Electronic Power Steering (EPS) or Steer-by-wire architectures. The second chapter describes a Torque Vectoring (TV) control strategy, applied to a Fully Electric Vehicle (FEV) with four independent electric motor (one for each wheel), that aims to optimize the lateral vehicle behavior by a proper electric motor torque regulation. A yaw rate controller is presented and designed in order to achieve a desired steady-state lateral behaviour of the car (handling task). Furthermore, a sideslip angle controller is also integrated to preserve vehicle stability during emergency situations (safety task). LQR, LQR+FF and ISM strategies are formulated and explained for yaw rate and concurrent yaw rate/sideslip angle control techniques also comparing their advantages and weakness points. The TV strategy is implemented and calibrated on a FEV demonstrator by executing experimental maneuvers (step steer, skid pad, lane change and sequence of step steers) thus proving the efficacy of the proposed controller and the safety contribution guaranteed by the sideslip control. The TV could be also applied for internal combustion engine driven vehicles by installing specific torque vectoring differentials, able to distribute the torque generated by the engine to each wheel independently. The TV strategy evaluated in the second chapter can be influenced by the presence of a transmission between themotor (or the engine) and wheels (where the torque control is supposed to be designed): in addition to the mechanical delay introduced by transmission components, the presence of gears backlashes can provoke undesired noises and vibrations in presence of torque sign inversion. The last chapter is thus related to a new method for noises and vibration attenuation for a Dual Clutch Transmission (DCT). This is achieved in a new way by integrating the powertrain control with the braking system control, which are historically and conventionally analyzed and designed separately. It is showed that a torsional preload effect can be obtained on transmission components by increasing the wheel torque and concurrently applying a braking wheel torque. For this reason, a pressure following controller is presented and validated through a Hardware In the Loop (HIL) test rig in order to track a reference value of braking torque thus ensuring the desired preload effect and noises reduction. Experimental results demonstrates the efficacy of the controller, also opening new scenario for global chassis control design. Finally, some general conclusions are drawn and possible future activities and recommendations are proposed for further investigations or improvements with respect to the results shown in the present work.