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.

The continuous improvement of spark-ignition direct-injection (SIDI) engines is largely attributed to the enhanced understanding of air-fuel mixing and combustion processes. The intricate interaction between transient spray behavior and the ambient flow field is important to unveil the airflow dynamics during the spray injection process. This study investigates the fuel-spray boundary interactions under different superheated conditions by analyzing the ambient flow field pattern with constraint-based modeling (CBM). In the experimental setup, superheated conditions are facilitated by adjusting different fuel temperatures and ambient pressures. By adding the tracer particles containing Rhodamine 6G to the ambient air, the combined diagnostic of fluorescent particle image velocimetry (FPIV) and Mie-scattering is implemented to measure the velocity distribution and flow trajectory of the air surrounding the spray formation and propagation.

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.

The heat pump with low global warming potential (GWP) refrigerants is imperative for the electric vehicle (EV) to slow down global warming and extend the driving range while meeting passengers' thermal comfort in low ambient temperatures. However, there are no appropriate refrigerants. To provide long-term and environmental-friendly refrigerants in the heat pump for EVs, herein, we reported newly developed low-GWP refrigerant mixtures, i.e., DL3B, whose GWP is lower than 140, the flammability (lower flammability limit and burning velocity), saturation pressure, lubricant miscibility, material compatibility were experimentally tested. A test bench that can investigate the performance of an R410A prototype was built. The drop-in tests of the DL refrigerant were carried out to evaluate the capacities and COPs for both cooling and heating modes in the EV heat pump system.

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.

Gluing is an essential fastening step in the field of aircraft assembly except for riveting and bolting. Generally, the robotic programs of gluing are generated in CAM environment. Due to the positioning errors and deformation of the workpiece to be glued in the fixture, the nominal pose and the actual pose of the workpiece are no longer consistent with each other. The Robot trajectory of dispensing glue is adjusted manually according to the actual pose of the workpiece by robot teaching. In this paper, an on-line gluing path correction method is developed by 2D laser profile measurement. A pose calibration method for 2D laser profiler integrated into a gluing robot by measuring a fixed center point of a standard ball is proposed to identify the position and orientation of the laser sensor, which enables the accurate transforming coordinates between the robot frame and the sensor frame.

The automotive industry uses supercharging in combination with various EGR strategies to meet the increasing demand for Diesel engines with high efficiency and low engine emissions. The charge air is heated by the EGR and the compression in the turbocharger to such an extent that high NOx emissions and a reduction in engine performance occurs. For this reason, the charge air cooler cools down the charge air before it enters the air intake manifold. In case of low pressure EGR, the charge air possesses a high moisture content and under certain operating conditions an accumulation of condensate takes place within the charge air cooler. During demanding engine loads, the condensate is entrained from the charge air cooler into the combustion chamber, resulting in misfiring or severe engine damage.

Computational fluid dynamics (CFD) modeling has many potentials for the design and calibration of modern and future engine concepts, including facilitating the exploration of operation conditions and casting light on the involved physical and chemical phenomena. As more attention is paid to the matching of different fuel types and combustion strategies, the use of detailed chemistry in characterizing auto-ignition, flame stabilization processes and the formation of pollutant emissions is becoming critical, yet computationally intensive. Therefore, there is much interest in using tabulated approaches to account for detailed chemistry with an affordable computational cost. In the present work, the tabulated flamelet progress variable approach (TFPV), based on flamelet assumptions, was investigated and validated by simulating constant-volume Diesel combustion with primary reference fuels - binary mixtures of n-heptane and iso-octane.

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.

Developing advanced combustion mode has been the active area for high efficiency and ultra-low emissions of the next-generation internal combustion engines. In this paper, a series of experiments were conducted in a modified single-cylinder compression ignition engine for operating a brand-new combustion mode denoted as intelligent charge compression ignition (ICCI) mode. By using two common-rail systems, commercial gasoline and diesel were alternately directly injected into the cylinder through multi-injection strategies in the injection timing range of 50~320 °CA BTDC. Thus, the in-cylinder stratified condition can be flexibly and accurately adjusted in this unique combustion mode. The key injection parameters, such as gasoline injection timing and diesel split ratio, were investigated to explore their effects on engine combustion, emissions, and fuel consumption.

The energy management strategies (EMS) for hybrid electric vehicles (HEV) have a great impact on the fuel economy (FE). The Pontryagin's minimum principle (PMP) has been proved to be a viable control strategy for HEV. The optimal costate of the PMP control can be determined by the given information of the driving conditions. Since the full knowledge of future driving conditions is not available, this paper proposed a dynamic optimization method for PMP costate without the prediction of the driving cycle. It is known that the lower fuel consumption the method yields, the more efficiently the engine works. The selection of costate is designed to make the engine work in the high efficiency range. Compared with the rule-based control, the proposed method by the principle of Hamiltonian, can make engine working points have more opportunities locating in the middle of high efficiency range, instead of on the boundary of high efficiency range.

The reliability of electronic components is of increasing importance for further progress towards automated driving. Thermal aging processes such as electromigration is one factor that can negatively affect the reliability of electronics. The resulting failures depend on the thermal load of the components within the vehicle lifetime - called temperature collective - which is described by the temperature frequency distribution of the components. At present, endurance testing data are used to examine the temperature collective for electronic components in the late development stage. The use of numerical simulation tools within Vehicle Thermal Management (VTM) enables lifetime thermal prediction in the early development stage, but also represents challenges for the current VTM processes [1, 2]. Due to the changing focus from the underhood to numerous electronic compartments in vehicles, the number of simulation models has steadily increased.

A powerful and efficient turbocharger turbine benefits the engine in many aspects, such as better transient response, lower NOx emissions and better fuel economy. The turbine performance can be further improved by employing secondary flow injection through an injector over the shroud section. A secondary flow injection system can be integrated with a conventional turbine without affecting its original design parameters, including the rotor, volute, and back disk. In this study, a secondary flow injection system has been developed to fit for an asymmetric twin-scroll turbocharger turbine, which was designed for a 6-cylinder heavy-duty diesel engine, aiming at improving the vehicle’s performance at 1100 rpm under full-loading conditions. The shape of the flow injector is similar to a single-entry volute but can produce the flow angle in both circumferential and meridional directions when the flow leaves the injector and enters the shroud cavity.

Using highly efficient powertrain is one of the most important and effective approaches to increase the driving distance of electric vehicles (EVs). In this paper, a novel two-speed dual-clutch transmission (DCT) is proposed. The transmission is comprised of two traditional friction clutches and two-stage planetary gear sets. One clutch connects the input sun gear and the other connects the input carrier. The Simulink models including an electric motor and two-speed DCT are established. Gearshift schedule based on fuzzy logic which reflects the driver’s intensions is adopted to improve the dynamic and economic performance of the novel transmission. The simulation model is built using MATLAB/Simulink® to validate the effectiveness of the proposed gearshift schedule compared with the conventional two-parameter gearshift schedule. Simulation results show that both the dynamic and economic performance of the novel DCT for EVs are improved with the proposed fuzzy logic gearshift schedule.

With the rapid development of series-parallel hybrid electric vehicles (SPHEVs), mode transition from pure electrical drive to hybrid drive has attracted considerable attention. The presence of time delay due to response capacity of actuators and signal transmission of communication may cause decrease of speed tracking accuracy, even instable dynamics. Consequently, drivability of the SPHEV is unacceptable, and durability of the components is reduced. So far, plenty of control strategies have been proposed for mode transition, however, no previous research has been reported to deal with the time delay during mode transition. In this paper, a dynamic model with time delay of hybrid electric system is established. Next, a mode transition time-delay controller is proposed based on a two degree of freedom internal model controller (2-DOF-IMC).

Future CO2 emission legislation requires the internal combustion engine to become more efficient than ever. Of great importance is the boosting system enabling down-sizing and down-speeding. However, the thermodynamic coupling of a reciprocating internal combustion engine and a turbocharger poses a great challenge to the turbine as pulsating admission conditions are imposed onto the turbocharger turbine. This paper presents a novel approach to a turbocharger turbine development process and outlines this process using the example of an asymmetric twin scroll turbocharger applied to a heavy duty truck engine application. In a first step, relevant operating points are defined taking into account fuel consumption on reference routes for the target application. These operation points are transferred into transient boundary conditions imposed on the turbine.

Multiple injection strategies are commonly used in conventional Diesel engines due to the flexibility for optimizing heat-release timing with a consequent improvement in fuel economy and engine-out emissions. This is also desirable in low-temperature combustion (LTC) engines since it offers the potential to reduce unburned hydrocarbon and CO emissions. To better utilize these benefits and find optimal calibrations of split injection strategies, it is imperative that the fundamental processes of multiple injection combustion are understood and computational fluid dynamics models accurately describe the flow dynamics and combustion characteristics between different injection events. To this end, this work is dedicated to the identification of suitable methodologies to predict the multiple injection combustion process.