Xianjia Yu

LiDAR-based SLAM systems are highly sensitive to adverse conditions such as occlusion, noise, and field-of-view (FoV) degradation, yet existing robustness evaluation methods either lack physical grounding or do not capture sensor-specific behavior. This paper presents a sensor-aware, phenomenological framework for simulating interpretable LiDAR degradations directly on real point clouds, enabling controlled and reproducible SLAM stress testing. Unlike image-derived corruption benchmarks (e.g., SemanticKITTI-C) or simulation-only approaches (e.g., lidarsim), the proposed system preserves per-point geometry, intensity, and temporal structure while applying structured dropout, FoV reduction, Gaussian noise, occlusion masking, sparsification, and motion distortion. The framework features autonomous topic and sensor detection, a modular configuration with four predefined severity tiers (light–extreme), and real-time performance ($< 5$ ms per frame for solid-state LiDAR and $< 20$ ms for dense, wide-FoV spinning LiDAR). The implementation is Docker-containerized and compatible with ROS workflows. Experimental validation across three LiDAR models and five state-of-the-art SLAM systems reveals distinct patterns of robustness shaped by sensor design and environmental context. The open-source implementation provides a practical foundation for benchmarking LiDAR-based SLAM under physically meaningful degradation scenarios.

Lidar technology has been widely employed across various applications, such as robot localization inGNSS-denied environments and 3D reconstruction. Recent advancements have introduced different lidar types, including cost-effective solid-state lidars such as the Livox Avia and Mid-360. The Mid-360, with its dome-like design, is increasingly used in portable mapping and uncrewed aerial vehicle(UAV) applications due to its low cost, compact size, and reliable performance. However, the lack of datasets that include dome-shaped lidars, such as the Mid-360, alongside other solid-state and spinning lidars significantly hinders the comparative evaluation of novel approaches across platforms. Additionally, performance differences between low-cost solid-state and high-end spinning lidars (e.g., Ouster OS series) remain insufficiently examined, particularly without an Inertial Measurement Unit (IMU) in odometry. To address this gap, we introduce a novel dataset comprising data from multiple lidar types, including the low-cost Livox Avia and the dome-shaped Mid-360, as well as high-end spinning lidars such as the Ouster series. Notably, to the best of our knowledge, no existing dataset comprehensively includes dome-shaped lidars such as Mid-360 alongside both other solid-state and spinning lidars. In addition to the dataset, we provide a benchmark evaluation of state-of-the-art SLAM algorithms applied to this diverse sensor data. Furthermore, we present a quantitative analysis of point cloud registration techniques, specifically point-to-point, point-to-plane, and hybrid methods, using indoor and outdoor data collected from the included lidar systems. The outcomes of this study establish a foundational reference for future research in SLAM and 3D reconstruction across heterogeneous lidar platforms.

Federated learning (FL) is increasingly recognised for addressing security and privacy concerns in traditional cloud-centric machine learning (ML), particularly within personalised health monitoring such as wearable devices. By enabling global model training through localised policies, FL allows resource-constrained wearables to operate independently. However, conventional first-order FL approaches face several challenges in personalised model training due to the heterogeneous non-independent and identically distributed (non-iid) data by each individual’s unique physiology and usage patterns. Recently, second-order FL approaches maintain the stability and consistency of non-iid datasets while improving personalised model training. This study proposes and develops a verifiable and auditable optimised second-order FL framework BFEL (blockchain enhanced federated edge learning) based on optimised FedCurv for personalised healthcare systems. FedCurv incorporates information about the importance of each parameter to each client’s task (through fisher information matrix) which helps to preserve client-specific knowledge and reduce model drift during aggregation. Moreover, it minimizes communication rounds required to achieve a target precision convergence for each client device while effectively managing personalised training on non-iid and heterogeneous data. The incorporation of ethereum-based model aggregation ensures trust, verifiability, and auditability while public key encryption enhances privacy and security. Experimental results of federated CNNs and MLPs utilizing mnist, cifar-10, and PathMnist demonstrate framework’s high efficiency, scalability, suitability for edge deployment on wearables, and significant reduction in communication cost.

ROS (Robot Operating System) has become the de facto standard in robotics research and development, with ROS 2, in particular, offering enhanced support for real-time communication, distributed systems, and scalable multi-robot applications. These capabilities have driven its widespread adoption across academia, industry, and the open-source community. However, deploying ROS 2 applications across heterogeneous hardware platforms remains a complex task—especially in scenarios that require tightly coordinated, multi-agent systems. In such cases, the failure of a single agent can propagate disruptions throughout the system. A representative example is Ultra-wideband (UWB)-based multi-robot relative localization, where inter-robot dependencies are essential for maintaining accurate relative positioning. While Kubernetes offers powerful features for automated deployment and orchestration, its integration with ROS 2 has not yet been thoroughly evaluated within the context of specific robotic applications. This paper addresses this gap by integrating Kubernetes with ROS 2 in a UWB-based multi-robot localization system, using UWB ranging error mitigation as a representative application. An edge cluster comprising five NVIDIA Jetson Nano devices and one laptop is orchestrated using Kubernetes, with a Jetson Nano node mounted on each robot. We deploy Long Short-Term Memory (LSTM)-based error mitigation modules on the edge nodes and systematically induce failures in various combinations of these modules. The system’s resilience and robustness are then assessed by analyzing position errors under different failure scenarios.

Lidar technology has been widely employed across various applications, such as robot localization in GNSS-denied environments and 3D reconstruction. Recent advancements have introduced different lidar types, including cost-effective solid-state lidars such as the Livox Avia and Mid-360. The Mid-360, with its dome-like design, is increasingly used in portable mapping and unmanned aerial vehicle (UAV) applications due to its low cost, compact size, and reliable performance. However, the lack of datasets that include dome-shaped lidars, such as the Mid-360, alongside other solid-state and spinning lidars significantly hinders the comparative evaluation of novel approaches across platforms. Additionally, performance differences between low-cost solid-state and high-end spinning lidars (e.g., Ouster OS series) remain insufficiently examined, particularly without an Inertial Measurement Unit (IMU) in odometry. To address this gap, we introduce a novel dataset comprising data from multiple lidar types, including the low-cost Livox Avia and the dome-shaped Mid-360, as well as high-end spinning lidars such as the Ouster series. Notably, to the best of our knowledge, no existing dataset comprehensively includes dome-shaped lidars such as Mid-360 alongside both other solid-state and spinning lidars. In addition to the dataset, we provide a benchmark evaluation of state-of-the-art SLAM algorithms applied to this diverse sensor data. Furthermore, we present a quantitative analysis of point cloud registration techniques, specifically point-to-point, point-to-plane, and hybrid methods, using indoor and outdoor data collected from the included lidar systems. The outcomes of this study establish a foundational reference for future research in SLAM and 3D reconstruction across heterogeneous lidar platforms.

Edge Intelligence (EI) serves as a critical enabler for privacy-preserving systems by providing AI-empowered computation and distributed caching services at the edge, thereby minimizing latency and enhancing data privacy. The integration of blockchain technology further augments EI frameworks by ensuring transactional transparency, auditability, and system-wide reliability through a decentralized network model. However, the operational architecture of such systems introduces inherent vulnerabilities, particularly due to the extensive data interactions between edge gateways (EGs) and the distributed nature of information storage during service provisioning. To address these challenges, we propose an autonomous computing model along with its interaction topologies tailored for privacy-critical and time-sensitive health applications. The system supports continuous monitoring, real-time alert notifications, disease detection, and robust data processing and aggregation. It also includes a data transaction handler and mechanisms for ensuring privacy at the EGs. Moreover, a resource-efficient one-dimensional convolutional neural network (1D-CNN) is proposed for the multiclass classification of arrhythmia, enabling accurate and real-time analysis of constrained EGs. Furthermore, a secure access scheme is defined to manage both off-chain and on-chain data sharing and storage. To validate the proposed model, comprehensive security, performance, and cost analyses are conducted, demonstrating the efficiency and reliability of the fine-grained access control scheme.

Recent advancements in lidar technology have led to improved point cloud resolution as well as the generation of 360 degrees, low-resolution images by encoding depth, reflectivity, or near-infrared light within each pixel. These images enable the application of deep learning (DL) approaches, originally developed for RGB images from cameras to lidar-only systems, eliminating other efforts, such as lidar-camera calibration. Compared with conventional RGB images, lidar imagery demonstrates greater robustness in adverse environmental conditions, such as low light and foggy weather. Moreover, the imaging capability addresses the challenges in environments where the geometric information in point clouds may be degraded, such as long corridors, and dense point clouds may be misleading, potentially leading to drift errors. Therefore, this paper proposes a novel framework that leverages DL-based colorization and super-resolution techniques on lidar imagery to extract reliable samples from lidar point clouds for odometry estimation. The enhanced lidar images, enriched with additional information, facilitate improved keypoint detection, which is subsequently employed for more effective point cloud downsampling. The proposed method enhances point cloud registration accuracy and mitigates mismatches arising from insufficient geometric information or misleading extra points. Experimental results indicate that our approach surpasses previous methods, achieving lower translation and rotation errors while using fewer points.

Edge Intelligence (EI) serves as a critical enabler for privacy-preserving systems, providing artificial intelligence(AI) powered computation and distributed caching services at the edge, thereby minimizing latency and enhancing data privacy. The integration of blockchain technology further strengthens these frameworks by ensuring transactional transparency, auditability, and system-wide reliability through a decentralised model. However, this operational architecture introduces inherent vulnerabilities, primarily due to the extensive data interactions between edge gateways (EGs) and the distributed nature of information storage during service provisioning. To address these challenges, we propose an autonomous computing pipeline along with its interaction topologies tailored for privacy-critical and time-sensitive health applications. The proposed system supports continuous monitoring, real-time heart rate rythm analysis, alert notifications, and robust data processing and aggregation at the edge. It incorporates a dedicated data transaction handler and privacy assurance mechanisms within the EGs. Furthermore, a resource-efficient one-dimensional convolutional neural network (1D-CNN) is proposed for the multiclass classification of arrhythmia, enabling accurate and real-time analysis utilising EGs. A secure access scheme is also defined to manage both off-chain and on-chain data sharing and storage. The proposed model is validated through comprehensive security, performance, and cost analyses, which demonstrate the efficiency and reliability of its fine-grained access control system.

Federated learning (FL) has become one of the key methods for privacy-preserving collaborative learning, as it enables the transfer of models without requiring local data exchange. Within the FL framework, an aggregation algorithm is recognized as one of the most crucial components for ensuring the efficacy and security of the system. Existing average aggregation algorithms typically assume that all client-trained data holds equal value or that weights are based solely on the quantity of data contributed by each client. In contrast, alternative approaches involve training the model locally after aggregation to enhance adaptability. However, these approaches fundamentally ignore the inherent heterogeneity between different clients' data and the complexity of variations in data at the aggregation stage, which may lead to a suboptimal global model. To address these issues, this study proposes a novel dual-criterion weighted aggregation algorithm involving the quantity and quality of data from the client node. Specifically, we quantify the data used for training and perform multiple rounds of local model inference accuracy evaluation on a specialized dataset to assess the data quality of each client. These two factors are utilized as weights within the aggregation process, applied through a dynamically weighted summation of these two factors. This approach allows the algorithm to adaptively adjust the weights, ensuring that every client can contribute to the global model, regardless of their data's size or initial quality. Our experiments show that the proposed algorithm outperforms several existing state-of-the-art aggregation approaches on both a general-purpose open-source dataset, CIFAR-10, and a dataset specific to visual obstacle avoidance.

The increased data transmission and number of devices involved in communications among distributed systems make it challenging yet significantly necessary to have an efficient and reliable networking middleware. In robotics and autonomous systems, the wide application of ROS 2 brings the possibility of utilizing various networking middlewares together with DDS in ROS 2 for better communication among edge devices or between edge devices and the cloud. However, there is a lack of comprehensive communication performance comparison of integrating these networking middlewares with ROS 2. In this study, we provide a quantitative analysis for the communication performance of utilized networking middlewares including MQTT and Zenoh alongside DDS in ROS 2 among a multiple host system. For a complete and reliable comparison, we calculate the latency and throughput of these middlewares by sending distinct amounts and types of data through different network setups including Ethernet, Wi-Fi, and 4G. To further extend the evaluation to real-world application scenarios, we assess the drift error (the position changes) over time caused by these networking middlewares with the robot moving in an identical square-shaped path. Our results show that CycloneDDS performs better under Ethernet while Zenoh performs better under Wi-Fi and 4G. In the actual robot test, the robot moving trajectory drift error over time (96 s) via Zenoh is the smallest. It is worth noting we have a discussion of the CPU utilization of these networking middlewares and the perfosrmance impact caused by enabling the security feature in ROS 2 at the end of the paper.