Search Results

The prediction of agents' future trajectory is a crucial task in supporting advanced driver-assistance systems (ADAS) and plays a vital role in ensuring safe decisions for autonomous driving (AD). Currently, prevailing trajectory prediction methods heavily rely on high-definition maps (HD maps) as a source of prior knowledge. While HD maps enhance the accuracy of trajectory prediction by providing information about the surrounding environment, their widespread use is limited due to their high cost and legal restrictions. Furthermore, due to object occlusion, limited field of view, and other factors, the historical trajectory of the target agent is often incomplete This limitation significantly reduces the accuracy of trajectory prediction. Therefore, this paper proposes ETSA-Pred, a mapless trajectory prediction model that incorporates enhanced temporal modeling and spatial self-attention.

Ammonia has attracted the attention of a growing number of researchers in recent years. However, some properties of ammonia (e.g., low laminar burning velocity, high ignition energy, etc.) inhibit its direct application in engines. Several routes have been proposed to overcome these problems, such as oxygen enrichment, partial fuel cracking strategy and co-combustion with more reactive fuels. Improving the reactivity of ammonia from the oxidizer side is also practical. Ozone is a highly reactive oxidizer which can be easily and rapidly generated through electrical plasma and is an effective promoter applicable for a variety of fuels. The dissociation reaction of ozone increases the concentration of reactive radicals and promotes chain-propagating reactions. Thus, obtaining accurate rate constants of reactions related to ozone is necessary, especially at elevated to high pressure range which is closer to engine-relevant conditions.

In order to meet increasingly stringent emission regulations, it is significance to establish a control- oriented transient NOx and PM emission prediction model and improve the accuracy and real-time performance. In this study, the prediction model of transient PM emissions based on Transformer is established. In terms of model accuracy and real-time performance, Transformer emission prediction model is compared with Multilayer perceptron (MLP) neural network and Long-Short Term Memory (LSTM) emission prediction model. The results show that the performance of Transformer transient emission prediction model is superior to other model structures, it can be used for real-time prediction.

Connected autonomy brings with it the means of significantly increasing vehicle Energy Economy (EE) through optimal Eco-Driving control. Much research has been conducted in the area of autonomous Eco-Driving control via various methods. Generally, proposed algorithms fall into the broad categories of rules-based controls, optimal controls, and meta-heuristics. Proposed algorithms also vary in cost function type with the 2-norm of acceleration being common. In a previous study the authors classified and implemented commonly represented methods from the literature using real-world data. Results from the study showed a tradeoff between EE improvement and run-time and that the best overall performers were meta-heuristics. Results also showed that cost functions sensitive to the 1-norm of acceleration led to better performance than those which directly minimize the 2-norm.

Exhaust Gas Recirculation (EGR) coolers are widely used on diesel engines to reduce in-cylinder NOx formation. A common problem is the accumulation of a fouling layer inside the heat exchanger, mainly due to thermophoresis that leads to deposition of particulate matter (PM), and condensation of hydrocarbons (HC) from the diesel exhaust. From a recent investigation of deposits from field samples of EGR coolers, it was confirmed that the densities of their deposits were much higher than reported in previous studies. In this study, the experiments were conducted in order to verify hypotheses about deposit growth, especially densification. An experimental set up which included a custom-made shell and tube type heat exchanger with six surrogate tubes was designed to control flow rate independently, and was installed on a 1.9 L L-4 common rail turbo diesel engine.

Making manned and remotely-controlled wheeled and tracked vehicles easier to drive, especially off-road, is of great interest to the U.S. Army. If vehicles are easier to drive (especially closed hatch) or if they are driven autonomously, then drivers could perform additional tasks (e.g., operating weapons or communication systems), leading to reduced crew sizes. Further, poorly driven vehicles are more likely to get stuck, roll over, or encounter mines or improvised explosive devices, whereby the vehicle can no longer perform its mission and crew member safety is jeopardized. HMI technology and systems to support human drivers (e.g., autonomous driving systems, in-vehicle monitors or head-mounted displays, various control devices (including game controllers), navigation and route-planning systems) need to be evaluated, which traditionally occurs in mission-specific (and incomparable) evaluations.

The automotive industry widely accepted the launch of electric vehicles in the global market, resulting in the emergence of many new areas, including battery health, inverter design, and motor dynamics. Maintaining the desired thermal stress is required to achieve augmented performance along with the optimal design of these components. The HVAC system controls the coolant and refrigerant fluid pressures to maintain the temperatures of [Battery, Inverter, Motor] in a definite range. However, identifying the prominent factors affecting the thermal stress of electric vehicle components and their effect on temperature variation was not investigated in real-time. Therefore, this article defines the vector electric vehicle thermal operating point (EVTHOP) as the first step with three elements [instantaneous battery temperature, instantaneous inverter temperature, instantaneous stator temperature].

Visual Place Recognition (VPR) excels at providing a good location prior for autonomous vehicles to initialize the map-based visual SLAM system, especially when the environment changes after a long term. Condition change and viewpoint change, which influences features extracted from images, are two of the major challenges in recognizing a visited place. Existing VPR methods focus on developing the robustness of global feature to address them but ignore the benefits that local feature can auxiliarily offer. Therefore, we introduce a novel hierarchical place recognition method with both global and local features deriving from homologous VLAD to improve the VPR performance. Our model is weak supervised by GPS label and we design a fine-tuning strategy with a coupled triplet loss to make the model more suitable for extracting local features.

Increasingly strict emission regulations and unfavorable economic climate bring severe challenges to the energy conservation of marine low-speed engine. Besides traditional methods, the energy and exergy analysis could acknowledge the losses of fuel from a global perspective to further improve the engine efficiency. Therefore, the energy and exergy analysis is conducted for a marine low-speed engine based on the experimental data. Energy analysis shows the exhaust gas occupies the largest proportion of all fuel energy waste, and it rises with the increment of engine load. The heat transfer consumes the second largest proportion, while it is negatively correlated to engine load. The energy analysis indicates that the most effective way to improve the engine efficiency is to reduce the energy wasted by exhaust gas and heat transfer. However, the latter exergy analysis demonstrates that there are other effective approaches to improve the engine efficiency.

To improve the combustion and emission characteristics of the large-bore marine engines, the spray is usually designed as an inter-spray impingement to promote the fuel-air mixing process, which implies frequent droplet collisions. Properly describing the collision dynamics of liquid droplets has been of interest in the field of spray modeling for marine engine applications. In this context, this work attempts to develop an accurate and efficient methodology for modeling impinging sprays under engine-like conditions. Experimental validations in terms of spray penetration and morphology are initially carried out at different operating conditions considering the parametric variations of ambient temperature and pressure, where the measurements are performed on a large-scale constant volume chamber with two symmetrical injectors.

Fuel film deposited on fuel injector tips used in gasoline direct injection engines, otherwise known as nozzle tip wetting, has been identified as an essential source of particle emissions. Attempts have been made to reduce nozzle tip wetting by the optimization design of nozzle geometry parameters. However, relevant investigations are still limited to emission measurements and corresponding indirect analysis. Due to the lack of related visualization research, the mechanism of nozzle tip wetting formation and its link with nozzle internal flow are still unclear. To clarify the influence of spray hole length on nozzle tip wetting and the underlying mechanisms, the dynamic formation process and the fuel film area evolution of nozzle tip wetting were visualized directly using laser-induced fluorescence technique and photomicrography technique.

Semi-trucks, specifically class-8 trucks, have recently become a platform of interest for autonomy systems. Platooning involves multiple trucks following each other in close proximity, with only the lead truck being manually driven and the rest being controlled autonomously. This approach to semi-truck autonomy is easily integrated on existing platforms, reduces delivery times, and reduces greenhouse gas emissions via fuel economy benefits. Level 1 SAE fuel studies were performed on class-8 trucks operating with the Auburn Cooperative Adaptive Cruise Control (CACC) system, and fuel savings up to 10-12% were seen. Enabling platooning autonomy required the use of radar, global positioning systems (GPS), and wireless vehicle-to-vehicle (V2V) communication. Poor measurements and state estimates can lead to incorrect or missing positioning data, which can lead to unnecessary dynamics and finally wasted fuel.

Autonomy for multiple trucks to drive in a fixed-headway platoon formation is achieved by adding precision GPS and V2V communications to a conventional adaptive cruise control (ACC) system. The performance of the Cooperative ACC (CACC) system depends heavily on the reliability of the underlying V2V communications network. Using data recorded on precision-instrumented trucks at both ACM and NCAT test tracks, we provide an understanding of various effects on V2V network performance: Occlusions - non-line-of-sight (NLOS) between the Tx and Rx antenna may cause network signal loss. Rain - water droplets in the air may cause network signal degradation. Antenna position - antennas at higher elevation may have less ground clutter to deal with. RF interference - interference may cause network packet loss. GPS outage - outages caused by tree cover, tunnels, etc. may result in degraded performance. Road curvature - curves may affect antenna diversity.

This paper presents experimental results that validate eco-driving and eco-heating strategies developed for connected and automated vehicles (CAVs). By exploiting vehicle-to-infrastructure (V2I) communications, traffic signal timing, and queue length estimations, optimized and smoothed speed profiles for the ego-vehicle are generated to reduce energy consumption. Next, the planned eco-trajectories are incorporated into a real-time predictive optimization framework that coordinates the cabin thermal load (in cold weather) with the speed preview, i.e., eco-heating. To enable eco-heating, the engine coolant (as the only heat source for cabin heating) and the cabin air are leveraged as two thermal energy storages. Our eco-heating strategy stores thermal energy in the engine coolant and cabin air while the vehicle is driving at high speeds, and releases the stored energy slowly during the vehicle stops for cabin heating without forcing the engine to idle to provide the heating source.

Homogeneous lean combustion is expected to be a key technology to further improve the combustion and reduce emissions of spark-ignition direct-injection engines. The application of lean combustion is facing many challenges such as slow flame propagation and combustion fluctuations. Under severe operating conditions such as low-speed lean-burn conditions, the weak in-cylinder airflow worsens the fuel and air mixing yielding difficulties in stable flame kernel initiation and consequently deteriorating flame propagation. In this study, the effect of flash boiling spray on flame kernel generation, flame propagation, engine performance, and exhaust emissions of the spark ignition direct injection (SIDI) engine under homogenous lean-burn conditions are investigated. A single-cylinder four-stroke optical SIDI engine was used in this study. The in-cylinder flash boiling and subcooled sprays during engine operation were compared using the Mie scattering technique.

To alleviate the shortage of petroleum resources and the air pollution caused by the burning of fossil fuels, the development of renewable fuels has attracted widespread attention. Among the various renewable fuels, ethanol can be produced from biomass and does not require much modification when applied to practical engines, so it has been widely used. However, ethanol fuel has a higher heat of vaporization than gasoline, it is difficult to evaporate and atomize under cold start conditions. Besides, the catalyst has not reached the conversion temperature at this time, resulting in lower conversion efficiency. These factors all lead to higher pollutant emission levels in ethanol-gasoline blends. To solve the above problems, this research used visualization techniques to compare the effects of flash boiling and high-energy ignition technologies on the in-cylinder combustion process and pollutant emission of ethanol-gasoline blends fuel.

Multiple models to estimate motion sickness (MS) have been proposed in the literature; however, few capture the influence of visual cues, limiting the models’ ability to predict MS that closely matches experimental MS data. This is especially significant in the presence of conflicts between visual and vestibular sensory signals. This paper provides an analysis of the gaps within existing MS estimation models and addresses these gaps by proposing the visual-vestibular motion sickness (VVMS) model. In this paper, the structure of the VVMS model, associated model parameters, and mathematical and physiological justification for selecting these parameters are presented. The VVMS model integrates vestibular sensory dynamics, visual motion perception, and visual-vestibular cue conflict to determine the conflict between the sensed and true vertical orientation of the passenger.

With the development of economic globalization, the speed of development of container terminals is also very rapid. Under the pressure brought by the surge in throughput, the unmanned and intelligent terminals will become the future development direction of terminals. As the cornerstone of the unmanned terminal, the positioning technology provides the most basic position information for system scheduling, path planning, real-time correction, and loading and unloading. Therefore, this paper is aimed to design a low-cost, high-precision, and easy-to-maintain unmanned dock positioning system in order to better solve the problem of unmanned dock positioning. The main research content of this paper is to design a positioning algorithm for unmanned terminal Automated Guided Vehicle (AGV) based on single-line lidar, including point cloud data acquisition, background filtering, point cloud clustering, vehicle position extraction, and result optimization.

Precise trajectory prediction of surrounding vehicles is critical for decision-making of autonomous vehicles, and learning-based approaches are well recognized for the robustness. However, state-of-the-art learning-based methods ignore (1) the feasibility of the vehicle’s multimodal state information for prediction and (2) the mutually exclusive relationship between the global traffic scene receptive fields and the local position resolution when modeling vehicles’ interactions, which may influence prediction accuracy. Therefore, we propose a “vehicle descriptor”-based long short-term memory (LSTM) model with the dilated convolutional social pooling (VD+DCS-LSTM) to cope with the above issues.

In recent years, with the increase of traffic accidents and traffic jams, lane change, as one of the most important and commonly automatic driving operations for autonomous vehicles, is receiving attention in academia. It is considered to be one of the important solutions that play an important role in improving road traffic safety and efficiency. However, most existing lane-changing models are rule-based lane-changing models. These models only assume a one-direction impact of surrounding vehicles on the lane-changing vehicle. In fact, lane change is a process of mutual interaction between vehicles due to the complexity and uncertainty of the traffic environment. Moreover, the safety and efficiency of existing lane-changing decision algorithms need to be improved. In this paper, we proposed a multivehicle cooperative control approach with a distributed control structure to control the model.