A5100651125- Autonomous Vehicle Technology and Safety
- Adversarial Robustness in Machine Learning
- Advanced Neural Network Applications
- Traffic and Road Safety
- Anomaly Detection Techniques and Applications
- Machine Learning and Data Classification
- Ocean Waves and Remote Sensing
- Industrial Vision Systems and Defect Detection
- Human-Automation Interaction and Safety
- Target Tracking and Data Fusion in Sensor Networks
- Fluid Dynamics and Turbulent Flows
- Quality and Supply Management
- Safety Warnings and Signage
- Robotic Path Planning Algorithms
- Algal biology and biofuel production
- Medical Image Segmentation Techniques
- Lattice Boltzmann Simulation Studies
- Vehicle emissions and performance
- Spacecraft and Cryogenic Technologies
- Solar-Powered Water Purification Methods
- Nanofluid Flow and Heat Transfer
- Earthquake and Tsunami Effects
- Lung Cancer Diagnosis and Treatment
- Advanced Optical Sensing Technologies
- Coastal and Marine Dynamics
Technical University of Darmstadt
2022-2024
China Automotive Engineering Research Institute
2023
Hesse (Germany)
2023
Karlsruhe Institute of Technology
2023
Mercedes-Benz (Germany)
2023
Nagoya University
2012
Kyushu University
2005-2010
This paper describes a framework for establishing reference airway tree segmentation, which was used to quantitatively evaluate fifteen different extraction algorithms in standardized manner. Because of the sheer difficulty involved manually constructing complete standard from scratch, we propose construct using results all that are be evaluated. We start by subdividing each segmented into its individual branch segments. Each segment is then visually scored trained observers determine...
To evaluate perception components of an automated driving system, it is necessary to define the relevant objects. While urban domain popular among datasets, relevance insufficiently specified for this domain. Therefore, work adopts existing method in highway and expands different conceptualizations definitions are present literature, there a lack methods validate these definitions. presents novel validation leveraging motion prediction component. The leverages idea that removing irrelevant...
Many recent approaches for automated driving (AD) functions currently include components relying on deep neural networks (DNNs). One approach in order to test AD is the scenario-based approach. This work formalizes and evaluates parameter discretization process required yield concrete scenarios which an function can be tested. Using a common perception algorithm camera images, simulation case study conducted simple static scenario containing one other vehicle. The results are analyzed with...
Abstract In automated vehicles, environment perception is performed by various sensor types, such as cameras, radars, lidars, and ultrasonics. Simulation models of these sensors, required in virtual validation methods, are available degrees detail. However, proving the validity a subject research. New metrics methods for credibility assessment simulation needed to standardize process future. The so-called double metric (DVM) has shown advantages allows an intuitive interpretability results....
Hydraulic experiments were examined to investigate the characteristics of thermal convection based on inhomogeneous water surface cooling in a closed body where aquatic plants grew thickly.We demonstrated process generation and development from results temperature measurement experiments.This explained was divided into three stages.Aquatic disturb transportation at surface, aspect ratio cell changes by magnitude density difference near without plants.Moreover, when shield existed simulated...
Having efficient testing strategies is a core challenge that needs to be overcome for the release of automated driving. This necessitates clear requirements as well suitable methods testing. In this work, perception modules are considered with respect relevance. The concept relevance currently remains insufficiently defined and specified.In paper, we propose novel methodology by exemplary application collision safety in highway domain. Using general system use case specification,...
This study was conducted in a shallow reservoir for agriculture to examine the response characteristics of internal seiche wind action closed water body.A observational results revealed that frequency wave density interface change with time corresponding unsteady action, and delay which corresponds stress is about almost 25 minutes Shikinawa Lake.When Wedderburn number between 10 0 2 , cycle same as seiche.Moreovr, value cause vibration Lake.
<title>Abstract</title> In automated vehicles, environment perception is performed by various sensor types, such as cameras, radars, lidars, and ultrasonics. Simulation models of these, sensors required in virtual validation methods, are available degrees detail. However, proving the validity a subject research. New metrics methods for credibility assessment simulation needed to standardize process future. The so-called Double Validation Metric (DVM) has shown advantages allows an intuitive...
Having efficient testing strategies is a core challenge that needs to be overcome for the release of automated driving. This necessitates clear requirements as well suitable methods testing. In this work, perception modules are considered with respect relevance. The concept relevance currently remains insufficiently defined and specified. paper, we propose novel methodology by exemplary application collision safety in highway domain. Using general system use case specification, corresponding...
Despite the presence of classification task in many different benchmark datasets for perception automotive domain, few efforts have been undertaken to define consistent requirements. This work addresses topic by proposing a structured method generate structure. First, legal categories are identified based on behavioral requirements vehicle. structure is further substantiated considering two aspects collision safety objects as well perceptual categories. A hierarchy obtained applying an...
Despite the presence of classification task in many different benchmark datasets for perception automotive domain, few efforts have been undertaken to define consistent requirements. This work addresses topic by proposing a structured method generate structure. First, legal categories are identified based on behavioral requirements vehicle. structure is further substantiated considering two aspects collision safety objects as well perceptual categories. A hierarchy obtained applying an...
<p> Despite the presence of many different benchmark datasets and metrics, there is a lack consistent perception requirements. This work focuses on task 3D object detection which aims at perceiving location attributes dynamic objects in space. Most notably, definitions pass fail criteria for any given metric. addresses this topic by systematically considering human performance across variety visual tasks. approach yields interpretable metrics as well thresholds pass/fail criteria. A...
<p> Despite the presence of many different benchmark datasets and metrics, there is a lack consistent perception requirements. This work focuses on task 3D object detection which aims at perceiving location attributes dynamic objects in space. Most notably, definitions pass fail criteria for any given metric. addresses this topic by systematically considering human performance across variety visual tasks. approach yields interpretable metrics as well thresholds pass/fail criteria. A...
Automated driving requires reliable perception of the environment to ensure safety task. One common task is 3D object detection, which aims at perceiving location and attributes dynamic objects. This typically evaluated on different benchmark datasets, each propose metrics. However, these metrics generally lack consistency bear no relation safety. Most notably, there a consistent definitions pass/fail criteria for any given detection metric. In this work, issue addressed by systematically...
In a closed water body with compound cross section, we examined the effects of horizontal convection based on density difference characteristics thermal generated by radiational cooling at surface. An experiment for measuring temperature revealed that fluid exchange was done between deep and shallow parts. Moreover, flow visualization to investigate thermal, which are developmental factors in vertical convection, found generation frequency interval thermals higher case section than even...
For an application of the remote sensing technique on agriculture, image taken from anartificial satellite is used widely. The would not be useful in a field-scale precision agriculturedue to limited power resolution. purpose that adopted field-scaleprecision resolution should increased or by another techniqueshould used.<br><br>In this study, plants was using digital camera with band-pass filter. Theimage analyzed related soil moisture and nutrient condition. Broccoli planted ninepots...