Integrity of Visual Navigation—Developments, Challenges, and Prospects

Abstract

Global navigation satellite systems (GNSSs) are widely used in applications nowadays. From maritime transportation to civil aviation, from route guidance to location-based services on smartphones, GNSS is a core technique for positioning. However, significant challenges with new applications in specific environments persist. Autonomous driving, unmanned aerial vehicles (UAV), and urban air mobility (UAM) require reliable navigation in deep urban areas. In these environments, the performance of GNSS is degenerated due to shadowing by tall buildings and strong multipath signals. It is difficult to solve these problems using GNSS technologies only. Multiple sensor fusion is necessary to deliver sufficient navigation performance required by such applications.Visual navigation using cameras as part of a multi-sensor solution offers a number of advantages in such applications. Firstly, the number of visual cues for camera-based localization algorithms is large in dense urban environments where GNSS faces its greatest challenges. The associated detailed visual information leads to precise estimates of the camera's pose (position and orientation). Secondly, the

Search for or Navigate to? Dual Adaptive Thinking for Object Navigation

“…Embodied AI (Duan et al 2022) is a challenging research topic that requires agents to use well-developed intuitive tasks (e.g., classification (Wang et al 2019) and detection (Liu et al 2020)) to complete complex logical tasks (e.g., navigation (Zhu, Meurer, and Günther 2022) and interaction (Shridhar et al 2020)) in real-world environments. Humans often use more than one way of thinking when completing these complex logical tasks.…”

Multisensor Localization Architecture for High-Accuracy and High-Integrity Land-based Applications

“…4 that a fault detection and elimination (FDE) function block is added after each step of processing (marked as red blocks). The design of the FDE functions follows the multi-domain error monitoring concept proposed for visual navigation integrity by Zhu et al (2022). The neural network outputs are first independently denoised and validated according to projective geometry.…”

Failure Mode Analysis (FMA) for Visual-Based Navigation for UAVs in Urban Environment

“…Commonly integrity monitoring in visual navigation faces several challenges and requires new solutions. Figure 1 illustrates familiar error sources in visual positioning that need to be considered to ensure the system's integrity [3]. Integrating multiple sensors can enhance performance and reduce the impact of failure in the system.…”

“…Additionally, the existing systems mentioned above have not discussed sensor failures in reference systems. Therefore, adopting the FMEA approach allows the systems to identify faults in reference systems with sensors and consider correlation and propagation of error sources [3]. Along with two local filters and one master filter, GNSS and IMU are the adding sensors to bound VINS error.…”