Impact of solar radiation and ambient temperature on automotive sensors (Virtual Vehicle)
Virtual testing and early lifetime predictions of electronic components are two huge topics with respect to automated vehicles. Since thousands of kilometres driven are needed in order to test and validate automated driving functions, digital twins of all related components and realistic driving profiles, as well as profiles for environmental influences, are needed to relieve the testing effort and to be able to increase the reliability.
Within the iRel40 project, Virtual Vehicle developed annual usage profiles for shuttle pods that enable sensor manufacturers to predict sensor usage in automotive applications. Additionally, the influence of environmental factors can be studied and used for fault-predictions or lifetime analysis.
Virtual Vehicle developed a data-driven model of the system temperature of an Ibeo LUX Lidar. Based on measurements recorded under different weather conditions, the developed model is able to predict the temperature inside the sensor, based on sensor runtime, ambient temperature and solar radiance. Two models are available, one for shaded setup of the LIDAR, and one for LIDAR positioning in direct sunlight. An exemplary prediction of the LIDAR temperature in combination with one day of the annual usage profile can be seen in the picture below.
Figure 1 Prediction of LIDAR system temperature based on novel weather and driving profiles for shuttle pod application. The LIDAR temperature shows the comparison between a shaded LIDAR positioning (blue) unshaded positioning (red).
Such temperature predictions can for example be used to map lab-applications, such as Burn-In for semiconductors, to in-field applications. In iRel40 this was done for the LIDAR example described above. Further details can be found in [1] and [2].
Reference
[1] Stephanie Grubmueller, Pamela Innerwinkler, 2022A Framework for the Determination of realistic Usage Profiles for Automated Shuttle Pods, International Conference on Connected Vehicle and Expo (ICCVE), 2022, pp.1–6.
[2] Stephanie Grubmueller, Pamela Innerwinkler, Marlies Mischinger, Horst Lewitschnig, Konstantin Posch, Selim Solmaz., 2023, Impact of solar radiation and ambient temperature on the early lifetime estimation of an automotive LIDAR, IEEE International Automated Vehicle Validation Conference 2023, Manuscript submitted for publication
Keywords
Virtual testing, system temperature model, burn-in time estimation, annual usage profiles