Repeatable Visibility Degradation using Water Spray for AD and ADAS TestingShow others and affiliations
2024 (English)In: 2024 IEEE International Automated Vehicle Validation Conference (IAVVC), Institute of Electrical and Electronics Engineers (IEEE), 2024Conference paper, Published paper (Refereed)
Abstract [en]
Automated vehicles and active safety functions based on sensor technology have been identified by the automotive industry as catalysts for improved safety, sustainability, accessibility, and efficiency. As technology advances, the applications of these systems are constantly expanding. Alongside these advancements, methods must be developed to evaluate and test AD and ADAS system performance and reliability in relevant and repeatable ways. This work outlines the main challenges in developing and evaluating a test method for generating road spray, a turbulent mix of fine water particles that reduce visibility caused by vehicles driving on wet surfaces. A hardware prototype and an appurtenant evaluation process were designed and produced to realize the test method. The evaluation process includes an automated software tool to quantify the prototype’s ability to degrade visibility and a method for automating sensor calibration for data collection at different locations and times. One of the key findings is the challenge of eliminating external disturbances in the test environment. Factors such as light and wind conditions significantly affect visibility through spray. The work concludes that controlling these factors is essential for achieving test repeatability. We successfully recreated road spray in a controlled environment, attenuating a sensor’s perceptive ability in steps of up to 80%, repeatedly within ±5-15%.
Place, publisher, year, edition, pages
Institute of Electrical and Electronics Engineers (IEEE), 2024.
Keywords [en]
spray, water particulates, ADAS, AD, automotive, visibility degradation, sensors, contrast, perception, adverse weather
National Category
Applied Mechanics
Identifiers
URN: urn:nbn:se:ri:diva-74971DOI: 10.1109/iavvc63304.2024.10786451Scopus ID: 2-s2.0-85216399888OAI: oai:DiVA.org:ri-74971DiVA, id: diva2:1893791
Conference
IEEE International Automated Vehicle Validation Conference (IAVVC)
Funder
Vinnova, 2021-025802024-08-302024-08-302026-04-16Bibliographically approved