Holder, Martin ; Elster, Lukas ; Winner, Hermann (2022)
Digitalize the Twin: A Method for Calibration of Reference Data for Transfer Real-World Test Drives into Simulation.
In: Energies, 2022, 15 (3)
doi: 10.26083/tuprints-00020520
Artikel, Zweitveröffentlichung, Verlagsversion
Kurzbeschreibung (Abstract)
In the course of the development of automated driving, there has been increasing interest in obtaining ground truth information from sensor recordings and transferring road traffic scenarios to simulations. The quality of the "ground truth" annotation is dictated by its accuracy. This paper presents a method for calibrating the accuracy of ground truth in practical applications in the automotive context. With an exemplary measurement device, we show that the proclaimed accuracy of the device is not always reached. However, test repetitions show deviations, resulting in non-uniform reliability and limited trustworthiness of the reference measurement. A similar result can be observed when reproducing the trajectory in the simulation environment: the exact reproduction of the driven trajectory does not always succeed in the simulation environment shown as an example because deviations occur. This is particularly relevant for making sensor-specific features such as material reflectivities for lidar and radar quantifiable in dynamic cases.
| Typ des Eintrags: | Artikel |
|---|---|
| Erschienen: | 2022 |
| Autor(en): | Holder, Martin ; Elster, Lukas ; Winner, Hermann |
| Art des Eintrags: | Zweitveröffentlichung |
| Titel: | Digitalize the Twin: A Method for Calibration of Reference Data for Transfer Real-World Test Drives into Simulation |
| Sprache: | Englisch |
| Publikationsjahr: | 2022 |
| Publikationsdatum der Erstveröffentlichung: | 2022 |
| Verlag: | MDPI |
| Titel der Zeitschrift, Zeitung oder Schriftenreihe: | Energies |
| Jahrgang/Volume einer Zeitschrift: | 15 |
| (Heft-)Nummer: | 3 |
| Kollation: | 16 Seiten |
| DOI: | 10.26083/tuprints-00020520 |
| URL / URN: | https://tuprints.ulb.tu-darmstadt.de/20520 |
| Zugehörige Links: | |
| Herkunft: | Zweitveröffentlichung DeepGreen |
| Kurzbeschreibung (Abstract): | In the course of the development of automated driving, there has been increasing interest in obtaining ground truth information from sensor recordings and transferring road traffic scenarios to simulations. The quality of the "ground truth" annotation is dictated by its accuracy. This paper presents a method for calibrating the accuracy of ground truth in practical applications in the automotive context. With an exemplary measurement device, we show that the proclaimed accuracy of the device is not always reached. However, test repetitions show deviations, resulting in non-uniform reliability and limited trustworthiness of the reference measurement. A similar result can be observed when reproducing the trajectory in the simulation environment: the exact reproduction of the driven trajectory does not always succeed in the simulation environment shown as an example because deviations occur. This is particularly relevant for making sensor-specific features such as material reflectivities for lidar and radar quantifiable in dynamic cases. |
| Freie Schlagworte: | virtual validation, automated driving, ground truth, reference measurement, calibration method, simulation |
| Status: | Verlagsversion |
| URN: | urn:nbn:de:tuda-tuprints-205209 |
| Sachgruppe der Dewey Dezimalklassifikatin (DDC): | 600 Technik, Medizin, angewandte Wissenschaften > 620 Ingenieurwissenschaften und Maschinenbau |
| Fachbereich(e)/-gebiet(e): | 16 Fachbereich Maschinenbau 16 Fachbereich Maschinenbau > Fachgebiet Fahrzeugtechnik (FZD) |
| Hinterlegungsdatum: | 13 Apr 2022 11:13 |
| Letzte Änderung: | 14 Apr 2022 05:14 |
| PPN: | |
| Zugehörige Links: | |
| Export: | |
| Suche nach Titel in: | TUfind oder in Google |
 |
Frage zum Eintrag |
Optionen (nur für Redakteure)
 |
Redaktionelle Details anzeigen |
Drucken
Impressum