Search Results

Abstract In the United States (USA), transportation is the largest single source of greenhouse gas (GHG) emissions, representing 27% of total GHGs emitted in 2020. Eighty-three percent of these came from road transport, and 57% from light-duty vehicles (LDVs). Internal combustion engine (ICE) vehicles, which still form the bulk of the United States (US) fleet, struggle to meet climate change targets. Despite increasingly stringent regulatory mechanisms and technology improvements, only three US states have been able to reduce their transport emissions to the target of below 1990 levels. Fifteen states have made some headway to within 10% of their 1990 baseline. Largely, however, it appears that current strategies are not generating effective results. Current climate-change mitigation measures in road transport tend to be predominantly technological.

Abstract This article presents an investigation of energy transfer, flame propagation, and emissions formation mechanisms in a four-cylinder, downsized and boosted, spark ignition engine fuelled by either directly injected compressed natural gas (DI CNG) or gasoline (GDI). Three different charge preparation strategies are examined for both fuels: stoichiometric engine operation without external dilution, stoichiometric operation with external exhaust gas recirculation (EGR), and lean burn. In this work, experiments and engine modelling are first used to analyze the energy transfer throughout the engine system. This analysis shows that an early start of fuel injection (SOI) improves fuel efficiency through lower unburned fuel energy at low loads with stoichiometric DI CNG operation.

Abstract The equivalent consumption minimization strategy (ECMS) is an instantaneous optimization method implemented online for hybrid electric vehicles (HEVs) to improve fuel economy. To fulfill the near-optimal performance of ECMS, equivalent factors (EFs) must be well tuned for different powertrains and driving cycles. This study proposes a hierarchical offline optimization framework which tunes the penalty value of state of charge (SOC) balance in the outer layer and optimizes EFs based on heuristic algorithms in the inner layer. A comprehensive analysis is conducted to evaluate three heuristic algorithms, including the genetic algorithm (GA), the nonlinear-inertia-decreasing particle swarm optimization algorithm (NLPSO), and the novel firefly algorithm (FA). The traversal optimization method (TOM) is chosen as the benchmark. Besides, a sensitivity analysis is carried out to reveal the impact of the penalty value on the battery SOC balance.

Abstract Vehicle-to-infrastructure (V2I) connectivity technology presents the opportunity for vehicles to perform autonomous longitudinal control to navigate safely and efficiently through sequences of V2I-enabled intersections, known as connected corridors. Existing research has proposed several control systems to navigate these corridors while minimizing energy consumption and travel time. This article analyzes and compares the simulated performance of three different autonomous navigation systems in connected corridors: a V2I-informed constant acceleration kinematic controller (V2I-K), a V2I-informed model predictive controller (V2I-MPC), and a V2I-informed reinforcement learning (V2I-RL) agent. A rules-based controller that does not use V2I information is implemented to simulate a human driver and is used as a baseline. The performance metrics analyzed are net energy consumption, travel time, and root-mean-square (RMS) acceleration.

Abstract Electric power steering (EPS) systems have become the most advantageous steering system used in vehicles. They provide better fuel efficiency and a more compact design over traditional hydraulic power steering (HPS) systems. However, EPS systems are afflicted with unwanted noise and vibration that can undermine the safety of drivers. This article presents a mathematical framework for vibration analysis in a column-type EPS system. The steering column is modeled as a continuous clamped column. The equations of motion are derived using Hamilton’s principle, and explicit expressions are presented for the frequency and transmissibility equations. A three-degrees-of-freedom (3-DOF) dynamic model is also presented by an approximation of the stiffness, damping, and mass of the steering column. The results of the proposed analytical models are validated using ANSYS simulation.

Abstract Among the myriad of potential hybrid powertrain architectures, selecting the optimal for an application is a daunting task. Whenever available, computer models greatly assist in it. However, some aspects, such as pollutant emissions, are difficult to model, leaving no other option than to test. Validating plausible options before building the powertrain prototype has the potential of accelerating the vehicle development even more, doing so without shipping components around the world. This work concerns the design of a system to virtually couple—that is, avoiding physical contact—geographically distant test rigs in order to evaluate the components of a powertrain. In the past, methods have been attempted, either with or without assistance of mathematical models of the coupled components (observers). Existing methods are accurate only when the dynamics of the systems to couple are slow in relation to the communication delay.

Abstract An energy management strategy (EMS) has an essential role in ameliorating the efficiency and lifetime of the powertrain components in a hybrid fuel cell vehicle (HFCV). The EMS of intelligent HFCVs is equipped with advanced data-driven techniques to efficiently distribute the power flow among the power sources, which have heterogeneous energetic characteristics. Decentralized EMSs provide higher modularity (plug and play) and reliability compared to the centralized data-driven strategies. Modularity is the specification that promotes the discovery of new components in a powertrain system without the need for reconfiguration. Hence, this article puts forward a decentralized reinforcement learning (Dec-RL) framework for designing an EMS in a heavy-duty HFCV. The studied powertrain is composed of two parallel fuel cell systems (FCSs) and a battery pack.

Abstract Deep learning (DL)-based approaches enable unprecedented control paradigms for propulsion systems, utilizing recent advances in high-performance computing infrastructure connected to modern vehicles. These approaches can be employed to optimize diesel aftertreatment control systems targeting the reduction of emissions. The optimization of the Trapped Soot Load (TSL) reduction in the Diesel Particulate Filter (DPF) is such an example. As part of the diesel aftertreatment system, the DPF stores the soot particles resulting from the combustion process in the engine. Periodically, the stored soot is oxidized during a DPF regeneration event. The efficiency of such a regeneration influences the fuel economy, and potentially the service interval of the vehicle. The quality of a regeneration depends on the operating conditions of the DPF, the engine, and the ability to complete the regeneration event.

Abstract This article proposes a novel approach of a hybrid system of physics and data-driven modeling for accurately estimating the state of charge (SOC) of a battery. State of Charge (SOC) is a measure of the remaining battery capacity and plays a significant role in various vehicle applications like charger control and driving range predictions. Hence the accuracy of the SOC is a major area of interest in the automotive sector. The method proposed in this work takes the state-of-the-art practice of Kalman filter (KF) and merges it with intelligent capabilities of machine learning using neural networks (NNs). The proposed hybrid system comprises a physics-based battery model and a plurality of NNs eliminating the need for the conventional KF while retaining its features of the predictor-corrector mechanism of the variables to reduce the errors in estimation.

Abstract Turbocharging can provide a low cost means for increasing the power output and fuel economy of an internal combustion engine. Currently, turbocharging is common in multi-cylinder engines, but due to the inconsistent nature of intake air flow, it is not commonly used in single-cylinder engines. In this article, we propose a novel method for turbocharging single-cylinder, four-stroke engines. Our method adds an air capacitor-an additional volume in series with the intake manifold, between the turbocharger compressor and the engine intake-to buffer the output from the turbocharger compressor and deliver pressurized air during the intake stroke. We analyzed the theoretical feasibility of air capacitor-based turbocharging for a single-cylinder engine, focusing on fill time, optimal volume, density gain, and thermal effects due to adiabatic compression of the intake air.

Abstract Nowadays, the need for a more sustainable mobility is fostering powertrain electrification as a way of reducing the carbon footprint of conventional vehicles. On the other side, the presence of multiple energy sources significantly increases the powertrain complexity and requires the development of a suitable Energy Management System (EMS) whose performance can strongly affect the fuel economy potential of the vehicle. In such a framework, this article proposes a novel methodology to reverse engineer the control strategy of a test case P2 Plug-in Hybrid Electric Vehicle (PHEV) through the analysis of experimental data acquired in a wide range of driving conditions. In particular, a combination of data obtained from On-Board Diagnostic system (OBD), Controller Area Network (CAN)-bus protocol, and additional sensors installed on the High Voltage (HV) electric net of the vehicle is used to point out any dependency of the EMS decisions on the powertrain main operating variables.

Abstract Effective circumstance perception technology is the prerequisite for the successful application of autonomous driving, especially the detection technology of traffic objects that affects other tasks such as driving decisions and motion execution in autonomous vehicles. However, recent studies show that a single sensor cannot perceive the surrounding environment stably and effectively in complex circumstances. In the article, we propose a multi-scale feature fusion framework that exploits a dual backbone network to extract camera and radar feature maps and performs feature fusion on three different feature scales using a new fusion module. In addition, we introduce a new generation mechanism of radar projection images and relabel the nuScenes dataset since there is no other suitable autonomous driving dataset for model training and testing.

Abstract Machine learning (ML) for predicting wireless channels of vehicular communications networks has attracted interest in recent years. Beamforming is a technique used to selectively transmit and receive data in a desired direction. The receiver should be capable of choosing the right beam at the right time. The usage of adaptive antenna scanning, i.e., scanning all the beams and choosing the best beam will result only in 30% accuracy, which means 70% of the data will be lost. This article studied a multiagency long short-term memory (LSTM) beamforming prediction model based on signal strength to forecast optimum beams within each beacon interval (BI) for cellular vehicle-to-everything (C-V2X) systems. The model combines the outputs of several parallel prediction models resulting in an enhanced accuracy of prediction. Simulation data validated the effectiveness of the proposed prediction model on the university campus, resulting in a 24% improvement in prediction accuracy.

Abstract The two-wheel system design is widely used in various mobile tools, such as remote-control vehicles and robots, due to its simplicity and stability. However, the specific wheel and body models in the real world can be complex, and the control accuracy of existing algorithms may not meet practical requirements. To address this issue, we propose a double inverted pendulum on mobile device (DIPM) model to improve control performances and reduce calculations. The model is based on the kinetic and potential energy of the DIPM system, known as the Euler-Lagrange equation, and is composed of three second-order nonlinear differential equations derived by specifying Lagrange. We also propose a stable feedback control method for mobile device drive systems. Our experiments compare several mainstream reinforcement learning (RL) methods, including linear quadratic regulator (LQR) and iterative linear quadratic regulator (ILQR), as well as Q-learning, SARSA, DQN (Deep Q Network), and AC.

Abstract This article presents a novel approach to optimize the placement of light detection and ranging (LiDAR) sensors in autonomous driving vehicles using machine learning. As autonomous driving technology advances, LiDAR sensors play a crucial role in providing accurate collision data for environmental perception. The proposed method employs the deep deterministic policy gradient (DDPG) algorithm, which takes the vehicle’s surface geometry as input and generates optimized 3D sensor positions with predicted high visibility. Through extensive experiments on various vehicle shapes and a rectangular cuboid, the effectiveness and adaptability of the proposed method are demonstrated. Importantly, the trained network can efficiently evaluate new vehicle shapes without the need for re-optimization, representing a significant improvement over classical methods such as genetic algorithms.

Abstract Heavy-duty (HD) diesel engines are the primary propulsion systems in use within the transportation sector and are subjected to stringent oxides of nitrogen (NOx) and particulate matter (PM) emission regulations. The objective of this study is to develop a robust calibration technique to optimize HD diesel engine for performance and emissions to meet current and future emissions standards during certification and real-world operations. In recent years, California - Air Resources Board (C-ARB) has initiated many studies to assess the technology road maps to achieve Ultra-Low NOx emissions for HD diesel applications [1]. Subsequently, there is also a major push for the complex real-world driving emissions as the confirmatory and certification testing procedure in Europe and Asia through the UN-ECE and ISO standards.

Abstract The demand for contactless, rapid manufacturing has increased over the years, especially during the COVID-19 pandemic. Additive manufacturing (AM), a type of rapid manufacturing, is a computer-based system that precisely manufactures products. It proves to be a faster, cheaper, and more efficient production system when integrated with cloud-based manufacturing (CBM). Similarly, the need for semiconductors has grown exponentially over the last five years. Several companies could not keep up with the increasing demand for many reasons. One of the main reasons is the lack of a workforce due to the COVID-19 protocols. This article proposes a novel technique to manufacture semiconductor chips in a fast-paced manner. An algorithm is integrated with cloud, machine vision, sensors, and email access to monitor with live feedback and correct the manufacturing in case of an anomaly.

Abstract Lane changes are stressful maneuvers for drivers, particularly during high-speed traffic flows. However, modeling driver’s lane-changing decision and implementation process is challenging due to the complexity and uncertainty of driving behaviors. To address this issue, this article presents a personalized Lane-Changing Model (LCM) for Advanced Driver Assistance System (ADAS) based on deep learning method. The LCM contains three major computational components. Firstly, with abundant inputs of Root Residual Network (Root-ResNet), LCM is able to exploit more local information from the front view video data. Secondly, the LCM has an ability of learning the global spatial-temporal information via Temporal Modeling Blocks (TMBs). Finally, a two-layer Long Short-Term Memory (LSTM) network is used to learn video contextual features combined with lane boundary based distance features in lane change events.

Abstract Mankind’s quest for clean alternative energy sources pushes the boundaries of science and technology every passing day. Increasing environmental and human health concerns caused by conventional fuels underscores the importance of identifying an energy source which is sustainable in terms of production, emissions, and wide application. Hythane (hydrogen-methane gas) is a contender for applications ranging from transportation and combined heating-power generation to cooking. In land transportation, the use of gaseous hythane for internal combustion engines shows better performance, enhanced combustion, and lower emission than conventional liquid hydrocarbon fuels. The use of liquefied Natural Gas (NG) and hydrogen in aircraft and ships is a steppingstone to more the wide-scale use of hythane in air, sea, and rail transportation sectors which could lower emissions and operating costs.

Abstract Many consider hydrogen to be the automobile fuel of the future. Indeed, it has numerous characteristics that makes it very attractive. Hydrogen has a much higher energy density than gasoline, can be produced from water, and its only emission is water. However, there are numerous challenges associated with hydrogen. In particular, the production of hydrogen is a key issue. Currently, most hydrogen is developed from methane, resulting in hydrogen having a carbon footprint. New investments into electrolysis from renewable energy sources is showing promise as an alternative for generating hydrogen. Further, the distribution of hydrogen poses many problems, requiring substantial infrastructure to support a hydrogen economy. Additionally, hydrogen storage is a key issue since most conventional storage mechanisms are overly bulky. If these three issues can be addressed, hydrogen is posed for being a key fuel as the world tries to move away from fossil fuels.