Testing Scenario Library Generation for Connected and Automated Vehicles: An Adaptive Framework
نویسندگان
چکیده
How to generate testing scenario libraries for connected and automated vehicles (CAVs) is a major challenge faced by the industry. In previous studies, evaluate maneuver of scenario, surrogate models (SMs) are often used without explicit knowledge CAV under test. However, performance dissimilarities between SM test usually exist, it can lead generation suboptimal libraries. this article, an adaptive library (ATSLG) method proposed solve problem. A customized specific model generated through process. To compensate leverage each CAV, Bayesian optimization techniques applied with classification-based Gaussian Process Regression newly designed acquisition function. Comparing pre-determined library, be tested evaluated in more efficient manner library. validate method, cut-in case study investigated results demonstrate that further accelerate evaluation process few orders magnitude.
منابع مشابه
An adaptive cruise control for connected energy-saving electric vehicles
We present an energy-saving cooperative adaptive cruise control (eco-CACC), which minimizes the energy consumption of autonomous electric vehicles. The approach leverages a trajectory preview from the preceding vehicle, and conciliates inter-vehicular distance reduction and speed profile smoothing. The problem is tackled with a nonlinear MPC approach. Rather than tracking a reference trajectory...
متن کاملAn Automated Testing Framework for Statistical Testing of GUI Applications
It is known to be inherently more difficult and laborintensive to functionally test software applications that employ a graphical user interface front-end, due to the vast GUI input space. We propose an automated testing framework for functional and statistical testing of GUI-driven applications, using a combination of two rigorous software specification and testing methods and integrating them...
متن کاملAn Adaptive Course Generation Framework
Existing adaptive e-learning methods are supported by student (user) profiling for capturing student characteristics, and course structuring for organizing learning materials according to topics and levels of difficulties. Adaptive courses are then generated by extracting materials from the course structure to match the criteria specified in the student profiles. In addition, to handle advanced...
متن کاملImplementing Fog Concept in the Connected Vehicles Scenario
.............................................................................................................. IV TABLE OF CONTENTS ........................................................................................... V Chapter
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: IEEE Transactions on Intelligent Transportation Systems
سال: 2022
ISSN: ['1558-0016', '1524-9050']
DOI: https://doi.org/10.1109/tits.2020.3023668