In response to the current and future energy and environment challenges, the automotive industry is strongly focusing on improving the fuel efficiency of vehicles. Although the electrification of automotive powertrains is clearly the principal path towards sustainable transportation, many opportunities still exist to improve the fuel economy of conventional vehicles. However, some of the technical solutions representing the state of the art in research and advanced development are difficult to benchmark in terms of their potential benefits for fuel consumption improvement. A greater understanding of the fuel energy utilization on the vehicle (here intended as a ‘system’ ) is therefore necessary in order to identify the readily available opportunities for efficiency improvements and, ultimately, to develop automobiles which are more fuel efficient. To this extent, this paper presents a review of the state of the art and technology trends in the field of energy management and recovery for automotive systems, with the primary focus on conventional powertrains. An understanding of the fuel energy utilization and dissipation associated with the vehicle subsystems (the engine, transmission, and chassis) is provided, as well as an overview of the opportunities and potential challenges in improving the fuel economy through system-level energy management, recovery, and harvesting. Finally, an overview of the most important solutions for managing energy dissipation, energy recovery, and harvesting is presented, discussing their potential for fuel economy improvement, technical readiness, and challenges. Wherever possible, projections on fuel economy improvements, based on either experimental data or simulations, are reported to provide opportunity for the assessment and comparison of current and future technologies.

The starter/alternator technology is considered an easily realizable hybrid electric vehicle (HEV) configuration to achieve significant fuel economy without compromising consumer acceptability. Several examples can be found in production or near-production vehicles, with implementation based on a spark ignition (SI) engine coupled with either a belted starter/alternator (BSA) or an integrated starter/alternator (ISA). One of the many challenges in successfully developing a starter/alternator HEV is to achieve engine start and stop operations with minimum passenger discomfort. This requires control of the electric motor to start and stop the engine quickly and smoothly, without compromising the vehicle noise, vibration, and harshness signature. The issue becomes more critical in the case of diesel hybrids, as the peak compression torque is much larger than in SI engines. This paper documents the results of a research activity focused on the control of the start and stop dynamics of a HEV with a belted starter/alternator. The work was conducted on a production 1.9 l common-rail diesel engine coupled to a 10.6 kW permanent magnet motor. The system is part of a series/parallel HEV powertrain, designed to fit a midsize prototype sport utility vehicle. A preliminary experimental investigation was done to assess the feasibility of the concept and to partially characterize the system. This facilitated the design of a control-oriented nonlinear model of the system dynamics and its validation on the complete HEV hardware. Model-based control techniques were then applied to design a controller for the belted starter/alternator, ensuring quick and smooth engine start operations. The final control design has been implemented on the vehicle. The research outcomes demonstrated that the BSA is effective in starting the diesel engine quickly and with very limited vibration and noise.

Deep Reinforcement Learning (DRL) has recently been applied to eco-driving to intelligently reduce fuel consumption and travel time. While previous studies synthesize simulators and model-free DRL (MFDRL), this work proposes a Safe Off-policy Model-Based Reinforcement Learning (SMORL) algorithm for eco-driving. SMORL integrates three key components, namely a computationally efficient model-based trajectory optimizer, a value function learned off-policy and a learned safe set. The advantages over the existing literature are three-fold. First, the combination of off-policy learning and the use of a physics-based model improves the sample efficiency. Second, the training does not require any extrinsic rewarding mechanism for constraint satisfaction. Third, the feasibility of trajectory is guaranteed by using a safe set approximated by deep generative models. The performance of SMORL is benchmarked over 100 trips against a baseline controller representing human drivers, a non-learning-based optimal controller, a previously designed MFDRL strategy, and the wait-and-see optimal solution. In simulation, SMORL reduces the fuel consumption by more than 21% while keeping the average speed comparable while compared to the baseline controller and demonstrates a better fuel economy while driving faster compared to the MFDRL agent and the non-learning-based optimal controller.

Connected and automated vehicles (CAVs), particularly those with a hybrid electric powertrain, have the potential to significantly improve vehicle energy savings in real-world driving conditions. In particular, the ecodriving problem seeks to design optimal speed and power usage profiles based on available information from connectivity and advanced mapping features to minimize the fuel consumption over an itinerary. This paper presents a hierarchical multilayer model predictive control (MPC) approach for improving the fuel economy of a 48 V mild-hybrid powertrain in a connected vehicle environment. Approximate dynamic programing (DP) is used to solve the receding horizon optimal control problem, whose terminal cost is approximated with the base policy obtained from the long-term optimization. The controller was tested virtually (with deterministic and Monte Carlo simulation) across multiple real-world routes, demonstrating energy savings of more than 20%. The controller was then deployed on a test vehicle equipped with a rapid prototyping embedded controller. In-vehicle testing confirm the energy savings obtained in simulation and demonstrate the real-time ability of the controller.

Connected and Automated Vehicles (CAVs), in particular those with multiple power sources, have the potential to significantly reduce fuel consumption and travel time in real world driving conditions. In particular, the eco-driving problem seeks to design optimal speed and power usage profiles based upon look-ahead information from connectivity and advanced mapping features, to minimize the fuel consumption over a given itinerary. In this work, the eco-driving problem is formulated as a Partially Observable Markov Decision Process (POMDP), which is then solved with a state-of-art Deep Reinforcement Learning (DRL) Actor Critic algorithm, Proximal Policy Optimization. An eco-driving simulation environment is developed for training and evaluation purposes. To benchmark the performance of the DRL controller, a baseline controller representing the human driver, a trajectory optimization algorithm and the wait-and-see deterministic optimal solution are presented. With a minimal onboard computational requirement and a comparable travel time, the DRL controller reduces the fuel consumption by more than 17% compared against the baseline controller by modulating the vehicle velocity over the route and performing energy efficient approach and departure at signalized intersections, over-performing the more computationally demanding trajectory optimization method.

Homogeneous charge compression ignition (HCCI) is a promising concept for internal combustion engines that can considerably decrease $NOx$ and soot emissions in part-load operations without penalizing fuel consumption. The HCCI combustion can be implemented in direct injection diesel engines without major modifications by introducing a specialized fuel injector in the intake port. This decouples the homogeneous mixture formation from the traditional in-cylinder injection, thus providing two fueling systems that can be used to optimize exhaust emissions and fuel consumption over the engine operating range. However, understanding and controlling the complex mechanisms and interactions driving the HCCI combustion process is still a difficult task. For this reason, it is essential to identify the most important control parameters and understand their influence on the auto-ignition process. The current work analyzes HCCI combustion with external mixture formation through experimental investigation and the definition of a control-oriented model. An extensive testing activity was performed on a passenger car diesel engine equipped with an external fuel atomizer to operate in HCCI mode. This provided an understanding of the process as well as experimental data to identify a mean value model of the system and its parameters. The model includes a thermodynamic combustion calculation that estimates the heat release, cylinder pressure, and the relevant variables for combustion control. The tool developed was then validated and used for analyzing the system behavior in steady state conditions. Finally, a description of the HCCI system behavior in transient operations is presented.