Application Of Multi Sensor Fusion In Autonomous Vehicle Localization Under Sensor Anomalies

Application of Multi-Sensor Fusion in Autonomous Vehicle Localization Under Sensor Anomalies

Multi-sensor Fusion-based Vehicle Localization

Multi-sensor fusion (MSF) is believed to be a promising tool for vehicular localization in urban environments. Due to the differences in principles and performance of various onboard vehicle sensors, MSF inevitably suffers from heterogeneous sources and vulnerability to cyber-attacks. Therefore, an essential requirement of MSF is the capability of providing a consumer-grade solution that operates in real-time, is accurate, and immune to abnormal conditions with guaranteed performance and quality of service for location-based applications. In other words, an MSF algorithm depends heavily on data synchronization, cost, an accurate process model, a prior knowledge of covariance matrices, integrity assessments, and security against cyber-attacks. Multi-sensor Fusion-based Vehicle Localization addresses trending technologies in MSF-based vehicle localization and outlines some insights into the unsettled issues and their potential solutions. The discussions and outlook are presented as a collection of key topics, including multi-sensor measurement data processing, sensory selection, filtering, integrity assessment, and cybersecurity. Click here to access the full SAE EDGETM Research Report portfolio. 9781468608854 9781468608861 https://doi.org/10.4271/EPR2024023

AI Embedded Assurance for Cyber Systems

The rapid growth and reliance on cyber systems have permeated our society, government, and military which is demonstrated in this book. The authors discuss how AI-powered cyber systems are designed to protect against cyber threats and ensure the security and reliability of digital systems using artificial intelligence (AI) technologies. As AI becomes more integrated into various aspects of our lives, the need for reliable and trustworthy AI systems becomes increasingly important. This book is an introduction to all of the above-mentioned areas in the context of AI Embedded Assurance for Cyber Systems. This book has three themes. First, the AI/ML for digital forensics theme focuses on developing AI and ML powered forensic tools, techniques, software, and hardware. Second, the AI/ML for cyber physical system theme describes that AI/ML plays an enabling role to boost the development of cyber physical systems (CPS), especially in strengthening the security and privacy ofCPS. Third, the AI/ML for cyber analysis theme focuses on using AI/ML to analyze tons of data in a timely manner and identify many complex threat patterns. This book is designed for undergraduates, graduate students in computer science and researchers in an interdisciplinary area of cyber forensics and AI embedded security applications. It is also useful for practitioners who would like to adopt AIs to solve cyber security problems.