Search Results

Convolutional neural networks are the de facto method of processing camera, radar, and lidar data for use in perception in ADAS and L4 vehicles, yet their operation is a black box to many engineers. Unlike traditional rules-based approaches to coding intelligent systems, networks are trained and the internal structure created during the training process is too complex to be understood by humans, yet in operation networks are able to classify objects of interest at error rates better than rates achieved by humans viewing the same input data.

In pursuit of safety validation of automated driving functions, efforts are being made to accompany real world test drives by test drives in virtual environments. To be able to transfer highly automated driving functions into a simulation, models of the vehicle’s perception sensors such as lidar, radar and camera are required. In addition to the classic pulsed time-of-flight (ToF) lidars, the growing availability of commercial frequency modulated continuous wave (FMCW) lidars sparks interest in the field of environment perception. This is due to advanced capabilities such as directly measuring the target’s relative radial velocity based on the Doppler effect. In this work, an FMCW lidar sensor simulation model is introduced, which is divided into the components of signal propagation and signal processing. The signal propagation is modeled by a ray tracing approach simulating the interaction of light waves with the environment.

Autonomous Driving is being utilized in various settings, including indoor areas such as industrial halls. Additionally, LIDAR sensors are currently popular due to their superior spatial resolution and accuracy compared to RADAR, as well as their robustness to varying lighting conditions compared to cameras. They enable precise and real-time perception of the surrounding environment. Several datasets for on-road scenarios such as KITTI or Waymo are publicly available. However, there is a notable lack of open-source datasets specifically designed for industrial hall scenarios, particularly for 3D LIDAR data. Furthermore, for industrial areas where vehicle platforms with omnidirectional drive are often used, 360° FOV LIDAR sensors are necessary to monitor all critical objects. Although high-resolution sensors would be optimal, mechanical LIDAR sensors with 360° FOV exhibit a significant price increase with increasing resolution.

In the evolving landscape of automated driving systems, the critical role of vehicle localization within the autonomous driving stack is increasingly evident. Traditional reliance on Global Navigation Satellite Systems (GNSS) proves to be inadequate, especially in urban areas where signal obstruction and multipath effects degrade accuracy. Addressing this challenge, this paper details the enhancement of a localization system for autonomous public transport vehicles, focusing on mitigating GNSS errors through the integration of a LiDAR sensor. The approach involves creating a 3D map using the factor graph-based LIO-SAM algorithm based on GNSS, vehicle odometry, IMU and LiDAR data. The algorithm is adapted to the use-case by adding a velocity factor and altitude data from a Digital Terrain model. Based on the map a state estimator is proposed, which combines high-frequency LiDAR odometry based on FAST-LIO with low-frequency absolute multiscale ICP-based LiDAR position estimation.

Aeroacoustics is important in the automotive industry, as it significantly influences driving comfort. Particularly in the case of battery electric vehicles (BEVs), the flow noise is already crucial at lower driving speeds, since these generate barely any drive noise and the masking effects produced by the engine are eliminated. Due to the increasing importance of drag minimization and elimination of the exhaust system, the underbody of BEVs is typically very streamlined and exhibits a low acoustic interference potential. However, even small geometric modifications to the vehicle can lead to changes in the flow around the vehicle and consequently to significant noise sources. Thus, significant flow resonances in the low frequency range below 30 Hz have been detected on certain vehicle configurations. Initial investigations have shown that the flow around the front wheel spoilers is relevant for the development of the flow phenomenon.

When travelling in an open-jet wind tunnel, the path of an acoustic wave is affected by the flow causing a shift of source positions in acoustical maps of phased arrays outside the flow. The well-known approach of Amiet attempts to correct for this effect by computing travel times between microphones and map points based on the assumption that the boundary layer of the flow, the so-called shear-layer, is infinitely thin and refracts the acoustical ray in a conceptually analogy to optics. However, in reality, the turbulent nature of both the not-so thin shear-layer and the acoustic emission process itself causes an additional smearing of sources in acoustic maps, which in turn causes deconvolution methods based on these maps - the most prominent example being CLEAN-SC - to produce certain ring effects, so-called halos, around sources.

Design verification and quality control of automotive components require the analysis of the source location of ultra-short sound events, for instance the engaging event of an electromechanical clutch or the clicking noise of the aluminium frame of a passenger car seat under vibration. State-of-the-art acoustic cameras allow for a frame rate of about 100 acoustic images per second. Considering that most of the sound events introduced above can be far less than 10ms, an acoustic image generated at this rate resembles an hard-to-interpret overlay of multiple sources on the structure under test along with reflections from the surrounding test environment. This contribution introduces a novel method for visualizing impulse-like sound emissions from automotive components at 10x the frame rate of traditional acoustic cameras.

Many technical projects, most vehicle and component testing, and all accident reconstructions, product failure analyses, and other forensic investigations, require photographic documentation. Roadway evidence disappears, tested or wrecked vehicles are repaired, disassembled, or scrapped, and components can be tested for failure. Photographs are frequently the only evidence that remains of a wreck, or the only records of subjects before or during tests. Making consistently good images during any inspection is a critical part of the evaluation process. 

Photographs and video recordings of vehicle crashes and accident sites are more prevalent than ever, with dash mounted cameras, surveillance footage, and personal cell phones now ubiquitous. The information contained in these pictures and videos provide critical information to understanding how crashes occurred, and  analyze physical evidence. This course teaches the theory and techniques for getting the most out of digital media, including correctly processing raw video and photographs, correcting for lens distortion, and using photogrammetric techniques to convert the information in digital media to usable scaled three-dimensional data.

SAE EDGE Research Reports provide examinations significant topics facing mobility industry today including Connected Automated Vehicle Technologies Electrification Advanced Manufacturing

SAE EDGE Research Reports provide examinations significant topics facing mobility industry today including Connected Automated Vehicle Technologies Electrification Advanced Manufacturing

SAE EDGE Research Reports provide examinations significant topics facing mobility industry today including Connected Automated Vehicle Technologies Electrification Advanced Manufacturing

SAE EDGE Research Reports provide examinations significant topics facing mobility industry today including Connected Automated Vehicle Technologies Electrification Advanced Manufacturing

Testing of ducted fuel injection (DFI) in a single-cylinder engine with production-like hardware previously showed that adding a duct structure increased soot emissions at the full load, rated speed operating point [1]. The authors hypothesized that the DFI flame, which travels faster than a conventional diesel combustion (CDC) flame, and has a shorter distance to travel, was being re-entrained into the on-going fuel injection around the lift-off length (LOL), thus reducing air entrainment into the on-going injection. The engine operating condition and the engine combustion chamber geometry were duplicated in a constant pressure vessel. The experimental setup used a 3D piston section combined with a glass fire deck allowing for a comparison between a CDC flame and a DFI flame via high-speed imaging. CH* imaging of the 3D piston profile view clearly confirmed the re-entrainment hypothesis presented in the previous engine work.

This paper addresses the issues of long-term signal loss in localization and cumulative drift in SLAM-based online mapping and localization in autonomous valet parking scenarios. A GPS, INS, and SLAM fusion localization framework is proposed, enabling centimeter-level localization with wide scene adaptability at multiple scales. The framework leverages the coupling of LiDAR and Inertial Measurement Unit (IMU) to create a point cloud map within the parking environment. The IMU pre-integration information is used to provide rough pose estimation for point cloud frames, and distortion correction, line and plane feature extraction are performed for pose estimation. The map is optimized and aligned with a global coordinate system during the mapping process, while a visual Bag-of-Words model is built to remove dynamic features.

One of the main challenges of autonomous driving is to integrate different modules, such as perception, planning, control, and communication, that work together to enable the vehicle to drive safely and efficiently. A key module of autonomous driving is the vehicle localization system, which estimates the vehicle's position in the environment, and provides guidance for the optimal route. The vehicle localization system is essential for ensuring the safety of autonomous driving. This paper proposes a vehicle localization method based on visual simultaneous localization and mapping (SLAM) using a monocular camera. The method captures images of the environment with a monocular camera and extracts ORB (Oriented FAST and rotated BRIEF) features from them. It then tracks the features across the images and constructs a sparse map of the scene. The map is used to estimate the vehicle's pose, which is the position and orientation of the vehicle, in local coordinates.

Unlike conventional heat shrink tubes or enclosure systems which only seals wires and splices on the outside, a novel Acrylate based sealing technology developed and introduced by Eurotech is a low viscosity fluid formulated to be applied to the splices either in liquid droplets or by dipping, utilizes fast capillary-wicking action and quick self-cure inside the wires to form a robust, cost effective, flexible, impenetrable seal to prevent moisture damage of wire harnesses and associated electrical components. This technology is an enabler of new wire harness architectures currently limited by the shortcomings of conventional sealing products such as heat shrink tubes which come up short when the splice configurations or geometries become too complex or difficult for sealing from the outside.

Ammonia shows promise as an alternative fuel for internal combustion engines (ICEs) in reducing CO2 emissions due to its carbon-free nature and well-established infrastructure. However, certain drawbacks, such as the high ignition energy, the narrow flammability range, and the extremely low laminar flame speed, limit its widespread application. The dual fuel (DF) mode is an appealing approach to enhance ammonia combustion. The combustion characteristics of ammonia-diesel dual fuel mode and ammonia-PODE3 dual fuel mode were experimentally studied using a full-view optical engine and the high-speed photography method. The ammonia energy ratio (ERa) was varied from 40% to 60%, and the main injection energy ratio (ERInj1) and the main injection time (SOI1) were also varied in ammonia-PODE3 mode.

Hydrogen–diesel dual-fuel combustion processes were visualized using an optically accessible rapid compression and expansion machine (RCEM). A hydrogen-air mixture was introduced into the combustion chamber, and a pilot injection of diesel fuel was used as the ignition source. A small amount of diesel fuel was injected as the pilot fuel at injection pressures of 40, 80, and 120 MPa using a common rail injection system. The injection amounts of diesel fuel were varied as 3, 6, and 13 mm3. The amount of hydrogen was manipulated by varying the total excess air ratio (λtotal) at 3 and 4. The RCEM was operated at a constant speed of 900 rpm, and the in-cylinder pressure and temperature at the top dead center (TDC) were set as 5 MPa and 700 K, respectively. The combustion processes were visualized via direct photography and hydroxyl (OH*) chemiluminescence photography using a high-speed camera and an image intensifier.

Water removal from Proton Exchange Membrane (PEM) Fuel Cell (FC) mainly involves two phenomena: some of the emerging droplets will roll on the Gas Diffusion Layer (GDL), others may impact channel walls and start sliding along the airflow direction. This different behaviour is linked to the hydrophobic/hydrophilic nature of the surface the water is moving on. In this paper, the walls of the channel of a FC were characterized by applying optical techniques. The deposition of droplets on the channel wall led to an evaluation of the proper range for Contact Angle Hysteresis (CAH = 55° - 45°), and due to the high wettability of the surface, droplets dimension was defined with a dimensionless parameter B/H. Under high crossflow condition (15 m/s) a sliding behaviour was observed. The channel features determined through image processing were used as boundary conditions for a 2D CFD two phase simulation employing the Volume of Fluid (VOF) model to keep track of the fluids interface.