With the widespread adoption of the Robot Operating System (ROS), technologies for autonomous navigation in mobile robots have advanced considerably. ROS provides a modular navigation stack that integrates essential components, such as SLAM, localisation, global path planning, and obstacle avoidance, forming the foundation for applications including service robotics and autonomous driving. Nonetheless, achieving safe and reliable navigation in complex and dynamic environments remains a formidable challenge, due to the need for real-time perception of moving obstacles, sensor fusion requirements, and the demand for robust and efficient algorithms. This study presents a systematic examination of the ROS-based navigation stack and obstacle-avoidance mechanisms. The architecture and implementation principles of the core modules are analysed, along with a comparison of the features and application suitability of common local planners such as the Dynamic Window Approach (DWA) and Timed Elastic Band (TEB). The key technical challenges in autonomous navigation are summarised, and recent advancements are reviewed to outline emerging trends in ROS-based systems, including integration with deep learning, multi-robot coordination, and real-time optimisation. The findings contribute to a deeper theoretical understanding of robotic navigation and offer practical guidance for the design and development of autonomous systems.

Combustible gas leakage remains a critical safety concern in industrial and indoor environments, necessitating the development of detection systems that are both accurate and practically deployable. This study presents a wireless gas detection system that integrates a gas sensor array, a low-power microcontroller with Zigbee-based communication, and a Back Propagation (BP) neural network optimized via a sequential hybrid strategy. Specifically, Particle Swarm Optimization (PSO) is employed for global parameter initialization, followed by Dung Beetle Optimization (DBO) for local refinement, jointly enhancing the network’s convergence speed and predictive precision. Experimental results confirm that the proposed PSO-DBO-BP model achieves high correlation coefficients (above 0.997) and low mean relative errors (below 0.25%) for all monitored gases, including hydrogen, carbon monoxide, alkanes, and smog. The model exhibits strong robustness in handling nonlinear responses and cross-sensitivity effects across multiple sensors, demonstrating its effectiveness in complex detection scenarios under laboratory conditions within embedded wireless sensor networks.

Small object detection in traffic scenes presents unique challenges for mobile robots operating under constrained computational resources and highly dynamic environments. Unlike general object detection, small targets often suffer from low resolution, weak semantic cues, and frequent occlusion, especially in complex outdoor scenarios. This study systematically analyses the challenges, technical advances, and deployment strategies for small object detection tailored to mobile robotic platforms. We categorise existing approaches into three main strategies: feature enhancement (e.g., multi-scale fusion, attention mechanisms), network architecture optimisation (e.g., lightweight backbones, anchor-free heads), and data-driven techniques (e.g., augmentation, simulation, transfer learning). Furthermore, we examine deployment techniques on embedded devices such as Jetson Nano and Raspberry Pi, and we highlight multi-modal sensor fusion using Light Detection and Ranging (LiDAR), cameras, and Inertial Measurement Units (IMUs) for enhanced environmental perception. A comparative study of public datasets and evaluation metrics is provided to identify current limitations in real-world benchmarking. Finally, we discuss future directions, including robust detection under extreme conditions and human-in-the-loop incremental learning frameworks. This research aims to offer a comprehensive technical reference for researchers and practitioners developing small object detection systems for real-world robotic applications.