Automation of logistics enhances efficiency, reduces costs, and minimizes human error. Image processing—particularly vision-based AI—enables real-time tracking, object recognition, and intelligent decision-making, thereby improving supply chain resilience. This study addresses the challenge of deploying deep learning-based object detection on resource-constrained embedded platforms, such as NVIDIA Jetson devices on UAVs and ground robots, for real-time logistics applications. Specifically, we provide a comprehensive comparative analysis of YOLOv5 and YOLOv8, evaluating their performance in terms of inference speed, accuracy, and dataset-specific metrics using both the Common Objects in Context (COCO) dataset and a novel, custom logistics dataset tailored for aerial and ground-based logistics scenarios. A key contribution is the development of a user-friendly graphical user interface (GUI) for selective object visualization, enabling dynamic interaction and real-time filtering of detection results—significantly enhancing practical usability. Furthermore, we investigate and compare deployment strategies in both Python 3.9 and C# (ML. NET v3 and .NET Framework 7) environments, highlighting their respective impacts on performance and scalability. This research offers valuable insights and practical guidelines for optimizing real-time object detection deployment on embedded platforms in UAV- and ground robot-based logistics, with a focus on efficient resource utilization and enhanced operational effectiveness.

Although it is a fundamental technique for autonomous navigation in GNSS-denied environments, tightly-coupled multi-sensor fusion has advanced faster than conventional narrative assessments. Using a PRISMA-guided review, bibliometrics, and Latent Dirichlet Allocation (LDA) topic modeling ($K=8, C_v=0.62$), we analyze a corpus of \Nincluded{} included publications (2015–2026) to give a data-driven synthesis of the domain. Resilience appears as an increasingly explicit topic across the corpus: more recent publications highlight degeneracy, outliers, and event-driven sensing inside state-estimation pipelines, in contrast to previous work that concentrated on nominal robustness. The literature from 2024 to 2025 particularly shows four trends (2026 is a half year in our export).:(i) measurement-domain integration of 5G, WiFi Round-Trip Time (WiFi-RTT), and ultra-wideband (UWB) signals; (ii) event-based and neuromorphic vision for high-dynamic-range scenarios; (iii) decentralized swarm cooperative localization; and (iv) semantic-cognitive map integration for managing geometric degeneracy. We also find a reporting gap: a lot of studies don't include computational performance data, which makes real-world validation more difficult. The analysis scripts and machine-readable corpus metadata are available upon request. Lastly, using three synthesis tools—a conceptual degradation--response framework for navigation resilience, an empirically supported taxonomy of degeneracy-mitigation mechanisms, and an engineering decision matrix for choosing sensor--estimator topologies under computational and environmental constraints—we convert the review findings into useful recommendations.