Camera-In-The-Loop Test Bench Development for Advanced Driver Assistance Systems Applications
MetadataShow full item record
Publisher:The Ohio State University
Series/Report no.:The Ohio State University. Department of Mechanical and Aerospace Engineering Honors Theses; 2019
As the prevalence of ADAS and autonomous vehicles rises, the demands placed on the sensors and controllers they rely on also increases. To accommodate for this increase in demand, more robust validation methods will be required to enable reliable, fast, and comprehensive sensor and control process testing. For the camera sensor, one of the sensors used in many ADAS systems, a hardware-in-the-loop validation method, camera-in-the-loop (CIL), is useful as it allows for the camera sensor to be tested in real-world scenarios while also generating data to be used in the verification of the controller. However, taking live video from inside vehicles can be a costly and slow process that hinders the development of autonomous systems. Instead, if the environment that the camera sensor observes is simulated virtually using a high-fidelity modeler, similar results can be achieved in a less costly and more timely fashion. Using software would also allow for the rapid generation of new, specified test cases allowing for greater control over algorithm testing. In this project, a Mobileye 6, a camera sensor equipped with computer vision is mounted on a test bench and faces a digital display. The display visualizes an environment simulated in PreScan that shows a realistic roadway from the view of the camera. The multiple components in this test bench are then connected through a CAN bus that sends vehicle speed characteristics. In doing so, the Mobileye’s computer vision interprets the displayed images and outputs valuable information such as object classification, object orientation, and various ADAS functions like lane departure warnings with a high degree of accuracy. Additionally, modifications to the test cases can be made in very little time while keeping the accuracy of the results consistent.
1st Place in OSU 2019 Undergraduate Research Festival 3 Minute Thesis Competition
Academic Major: Mechanical Engineering
Ohio State EcoCAR Student Project Team
Items in Knowledge Bank are protected by copyright, with all rights reserved, unless otherwise indicated.