Search Results

Abstract Real-time reconstruction of 3D environment attributed with semantic information is significant for a variety of applications, such as obstacle detection, traffic scene comprehension and autonomous navigation. The current approaches to achieve it are mainly using stereo vision, Structure from Motion (SfM) or mobile LiDAR sensors. Each of these approaches has its own limitation, stereo vision has high computational cost, SfM needs accurate calibration between a sequences of images, and the onboard LiDAR sensor can only provide sparse points without color information. This paper describes a novel method for traffic scene semantic segmentation by combining sparse LiDAR point cloud (e.g. from Velodyne scans), with monocular color image. The key novelty of the method is the semantic coupling of stereoscopic point cloud with color lattice from camera image labelled through a Convolutional Neural Network (CNN).

Abstract Additive manufacturing (AM) technologies can produce lighter parts; reduce manual assembly processes; reduce the number of production steps; shorten the production cycle; significantly reduce material consumption; enable the production of prostheses, implants, and artificial organs; and produce end-user products since it is used in many sectors for many reasons; it has also started to be used widely, especially in the field of aerospace. In this study, polylactic acid (PLA) was preferred for the antenna substrate because it is environmentally friendly, easy to recycle, provides convenience in production design with a three-dimensional (3D) printer, and is less expensive compared to other available materials. Copper (Cu) tape and graphene filament were employed for the antenna patch component due to their benefits.

Abstract Composite ceramic brake discs are made of ceramic material reinforced with carbon fibers and offer exceptional advantages that translate directly into higher vehicle performance. In the case of an electric vehicle, it could increase the range of the vehicle, and in the case of conventional internal combustion engine vehicles, it means lower fuel consumption (and consequently lower CO2 emissions). These discs are typically characterized by complex internal geometries, further complicated by the presence of drilling holes on both friction surfaces. To estimate the aerothermal performance of these discs, and for the thermal management of the vehicle, a reliable model for predicting the air flowing across the disc channels is needed. In this study, a real carbon-ceramic brake disc with drilling holes was investigated in a dedicated test rig simulating the wheel corner flow conditions experimentally using the particle image velocimetry technique and numerically.

Abstract Autonomous Racing is gaining increasing publicity as an attractive showcase of state-of-the-art technologies and the enhanced algorithms used for autonomous driving. The Indy Autonomous Challenge (IAC) tackled autonomous high-speed wheel-to-wheel racing at the famous Indianapolis Motor Speedway (IMS) in October 2021. To solve this problem, advanced autonomous driving algorithms were developed by each team. In this article, we display a multi-sensor localization approach developed for usage in the IAC. To decouple the lateral and longitudinal position of the ego vehicle, a trackbound coordinate system is used that can be transformed to conventional Cartesian coordinates. The longitudinal motion of the car is tracked via a modified version of the OpenVSLAM that outputs the progress of the already driven distance.

Abstract With the introduction of advanced lightweight materials with complex microstructures and behaviors, more focus is put on the accurate determination of their forming limits, and that can only be possible through experiments as the conventional theoretical models for the forming limit curve (FLC) prediction fail to perform. Despite that, CAE engineers, designers, and toolmakers still rely heavily on theoretical models due to the steep costs associated with formability testing, including mechanical setup, a large number of tests, and the cost of a stereo digital image correlation (DIC) system. The international standard ISO 12004-2:2021 recommends using a stereo DIC system for formability testing since two-dimensional (2D) DIC systems are considered incapable of producing reliable strains due to errors associated with out-of-plane motion and deformation.

Abstract A modular, stationary IC engine blow rig for differential and integral flow field measurements using particle image velocimetry (PIV) has been developed. Unlike conventional PIV blow rigs, the given design is capable of operating under near-reality charge air conditions, that is, highly pressurized, hot intake air supply at high flow rates. Its conceptual flexibility as well as peripheral infrastructure allow for comprehensive and wide-ranging flow field analysis. Because of a modular architecture, it is neither confined to a specific cylinder head design nor limited solely to the application of PIV for differential flow field analysis. It also already accounts for direct inlet flow determination through an additional PIV access point upstream of the cylinder head. The inlet and outlet ducts have been designed with regular shapes and smooth walls, such that a digital twin-type CFD model of the blow rig is conveniently feasible.

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 The following work presents a new concept for optical accessibility of a single-cylinder medium-speed large-bore marine engine from the concept development to the implementation and feasibility investigation in a test bench observing the flame chemiluminescence of dual-fuel (DF) combustion. The design’s feasibility is verified using conjugated heat transfer (CHT) and finite element method (FEM) simulation during the whole design process presented herein. Assumptions made for the simulation, e.g., of the mount between the optical component and the steel engine parts, are evaluated in pretesting setups presented and described as follows. The optical access is made to withstand steady-state full-load operating conditions and is proofed so. The optical access is designed for an engine with a bore of 350 mm and a stroke of 440 mm.

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 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 In this study, a parametric thoracic spine (T-spine) model was developed to account for morphological variations among the adult population. A total of 84 CT scans were collected, and the subjects were evenly distributed among age groups and both sexes. CT segmentation, landmarking, and mesh morphing were performed to map a template mesh onto the T-spine vertebrae for each sampled subject. Generalized procrustes analysis (GPA), principal component analysis (PCA), and linear regression analysis were then performed to investigate the morphological variations and develop prediction models. A total of 13 statistical models, including 12 T-spine vertebrae and a spinal curvature model, were combined to predict a full T-spine 3D geometry with any combination of age, sex, stature, and body mass index (BMI). A leave-one-out root mean square error (RMSE) analysis was conducted for each node of the mesh predicted by the statistical model for every T-spine vertebra.

Abstract The fuel injection pressures used in gasoline direct injection (GDI) engines have increased in recent years to improve fuel efficiency and reduce emissions. Current GDI engines use injection pressures of up to 350 bar, and there is evidence that even higher fuel injection pressures could yield further improvements in atomization. Higher injection pressures could also improve mixture formation by increasing the spray velocity; however, the research with higher injection pressures over 1000 bar is limited due to a limit of mechanical components. This manuscript summarizes experimental investigations into the effect of injection pressure, injection mass, and nozzle shape on spray-induced air motion with ultrahigh injection pressure over 1000 bar.

Abstract The dynamic characteristics of tires are critical in the overall vibrations of vehicles because the tire-road interface is the only medium of energy transfer between the vehicle and the road surface. Obtaining the natural frequencies and mode shapes of the tire helps in improving the comfort of the passengers. The vibrational characteristics of structures are usually obtained by performing conventional impact hammer modal testing, in which the structure is excited with an impact hammer and the response of the structure under excitation is captured using accelerometers. However, this approach only provides the response of the structure at a few discrete locations, and it is challenging to use this procedure for rotating structures. Digital Image Correlation (DIC) helps in overcoming these challenges by providing the full-field response of the structure.

Abstract With the increasing electrification of road vehicle powertrains, new NVH challenges are emerging. Due to the temporary deactivation of the internal combustion engine or even its complete elimination, the interior sound is primarily determined by noise shares of the electrical components and shares of tire and wind noise. In particular, the high-frequency tonal noise components induced by the electric powertrain are often dominant. These components usually have a negative impact on the perceived sound quality of the total vehicle. Major noise-causing components of the electric powertrain are the electric machine and the transmission. Since acoustic optimization is a complex and time-consuming process, this article develops and presents a novel method for the automated separation of the noise shares of the electric machine and the transmission gears in the vehicle interior.

Abstract The image from monocular camera is processed to detect depth information of the obstacles viewed by the rearview cameras of vehicle door side. The depth information recognized from a single, two-dimensional image data can be used for the purpose of blind spot area detection. Blind spot detection function is contributing to enhance the vehicle safety in scenarios such as lane-change and overtaking driving. In this article the depth cue information is inferred from the feature comparison between two image blocks selected within a single image. Convolutional neural network model trained by deep learning process with good enough accuracy is applied to distinguish if an obstacle is far or near for a specified threshold in the vehicle blind spot area. The application study results are demonstrated by the offline calculations with real traffic image data.

Abstract The article contains the results of operational investigations of deposit formation on external and internal surfaces of injector nozzles of the marine self-ignition engines during their operational use. The aim of this article is to introduce an image analysis method for global assessment of the quantity and quality of injector nozzle deposits in piston internal combustion engines. The components of medium-speed marine engines fueled with distillation and residual fuels were investigated. Digital images of new and used injector nozzles without deposits and with random deposits formed after natural operation on marine ships, respectively, were taken. Macro and microscopy images of external surfaces were taken in a shadowless tent and were illuminated with low-temperature lamps. The characteristic surfaces of the injector nozzles were virtually separated from the white background.

Abstract The greatest frictional contributor in an internal combustion engine is the contact between the piston ring pack and cylinder liner. Therefore, an improved lubrication regime has the potential to raise engine efficiency while lowering emissions, aiding to meet environmental regulations. Previous ultrasonic measurements of the oil film thickness (OFT) between piston rings and the cylinder liner in a marine engine have been subject to several unexpected trends. This article refines the measurement to identify and remove these factors, the trends were found to have arisen due to the detection of ultrasonic reflections from the piston ring outside of the expected alignment zone. The extent of these undesired reflections is thought to be due to the liner thickness providing a relatively large distance for spreading of the ultrasonic wavefront.

Abstract Single point incremental forming (SPIF) is a robust and new technique. In the recent research scenario, materials properties such as microstructure, micro-texture analysis, and crystal structure can be accessed through characterization non-destructive techniques, e.g., scanning electron microscope (SEM), electron backscattered diffraction (EBSD), and X-ray diffraction (XRD). XRD is a non-destructive method for analyzing the fine structure of materials. This study explores how process variables such as wall angle, step size, feed rate, and forming speed affect the parts of large-, medium-, and small-sized truncated cones of aluminum alloy AA3003-O sheet. Several cone parts of truncated cones are used in this investigation to implement Scherrer’s method. The two primary determining factors peak height and crystallite size are assessed for additional analysis in the present research.

Abstract The development of automated driving systems (ADS) necessitates procedures to validate system safety. The reliability of an ADS’s environment perception provided by lidar, radar, and camera sensors is of special interest in this context, because perception errors can be safety-critical. In this article, we formalize the reliability-based validation of environment perception for safe automated driving and discuss associated challenges. We describe a potential solution to a perception reliability validation by deriving performance requirements at the sensor level. We then summarize statistical methods to learn sensor perception reliabilities in field tests, on proving grounds, and through virtual simulations. With the developed safety validation framework, we show that, potentially, one can validate the safety of an ADS with feasible test effort.

Abstract The present study deals with the investigation on microstructural and mechanical properties of friction stir processed (FSPed) pure Aluminum (Al)-Graphene Nanoplatelets (GNPs) composites. Composite specimens such as castings were made by blending 0.5 wt.%, 1.0 wt.%, 1.5 wt.%, and 2.0 wt.% of GNPs in pure Al matrix using the ultrasonic-assisted stir casting technique (UASCT). Also for enhancement of mechanical properties via grain refinement the friction stir processing (FSP) has been employed, as well as mechanical properties like tensile strength and microhardness were evaluated. Moreover, the microstructural analysis were done using Scanning Electron Microscope (SEM), Field Emission Scanning Electron Microscope (FESEM), transmission electron microscopy (TEM), and X-Ray Diffraction (XRD) examination were also performed for inspecting the changes occurred during synthesis of the fabricated composites after FSP.