This review explores recent advancements in hydrogen gas (H₂) safety through the lens of artificial intelligence (AI) techniques. As hydrogen gains prominence as a clean energy source, ensuring its safe handling becomes paramount. The paper critically evaluates the implementation of AI methodologies, including artificial neural networks (ANN), machine learning algorithms, computer vision (CV), and data fusion techniques, in enhancing hydrogen safety measures. By examining the integration of wireless sensor networks and AI for real-time monitoring and leveraging CV for interpreting visual indicators related to hydrogen leakage issues, this review highlights the transformative potential of AI in revolutionizing safety frameworks. Moreover, it addresses key challenges such as the scarcity of standardized datasets, the optimization of AI models for diverse environmental conditions, etc., while also identifying opportunities for further research and development. This review foresees faster response times, reduced false alarms, and overall improved safety for hydrogen-related applications. This paper serves as a valuable resource for researchers, engineers, and practitioners seeking to leverage state-of-the-art AI technologies for enhanced hydrogen safety systems.

In artificial intelligence (AI), computer vision consists of intelligent models to interpret and recognize the visual world, similar to human vision. This technology relies on a synergy of extensive data and human expertise, meticulously structured to yield accurate results. Tackling the intricate task of locating and resolving blockages within sewer systems is a significant challenge due to their diverse nature and lack of robust technique. This research utilizes the previously introduced “S-BIRD” dataset, a collection of frames depicting sewer blockages, as the foundational training data for a deep neural network model. To enhance the model’s performance and attain optimal results, transfer learning and fine-tuning techniques are strategically implemented on the YOLOv5 architecture, using the corresponding dataset. The outcomes of the trained model exhibit a remarkable accuracy rate in sewer blockage detection, thereby boosting the reliability and efficacy of the associated robotic framework for proficient removal of various blockages. Particularly noteworthy is the achieved mean average precision (mAP) score of 96.30% at a confidence threshold of 0.5, maintaining a consistently high-performance level of 79.20% across Intersection over Union (IoU) thresholds ranging from 0.5 to 0.95. It is expected that this work contributes to advancing the applications of AI-driven solutions for modern urban sanitation systems.

There is an increasing trend of using automated and robotic systems for the tasks that are hazardous or inconvenient and dirty for humans. Sewers maintenance and cleaning is such a task where robots are already being used for inspection of underground pipes for blockages and damage. This paper reviews the existing robotic systems and various platforms and algorithms along with their capabilities and limitations being discussed. A typical mid-size city in a developing country, Pune, India is selected in order to understand the concerns and identify the requirements for developing robotic systems for the same. It is found that major concern of sewers are blockages but there is not enough information on both real-time detection and removal of it with robotic systems. On-board processing with computer vision algorithms has not been efficiently utilized in terms of performance and determinations for real-world implementations of sewer robotic systems. The review highlights the available methodologies that can be utilized in developing sewer inspection and cleaning robotic systems.

Water pollution has become a global issue and creating a severe problems due to contamination of various water sources such as agricultural waste, industrial waste, and sewage waste etc. So, the available sensory systems, nodes, and techniques plays a crucial role in identification of harmful pollutants in distinct water resources. The water pollutants monitoring techniques with prospered tools and sensors have been unfolded to find gap for future research maneuver. The need of real-time newer sensory system has been put forwarded for identification of mostly occurring harmful pollutants such as BOD, COD, TSS, and Hydrogen sulfide in examination of water quality. In this modern advancement, robotic systems have taken an intrinsic stand in diverse fields of humankind. An underground infrastructure is a base of modern society. In this paper, the existing methodologies and developed sewer robotic systems have been discussed and concluded on their applications, limitations and impact on realistic scenarios. Also, the cardinal point is revealed that previous art work focused only on sewer defect identification but not standardized work on sewer blockages detection and removal. So, the sewer robotic system with features of cost effectiveness and standardized accuracy matrix should be developed to resolve sewer blockage issues and followed by human scavenging. The survey outcomes extend a province of sewer robotics to resolve blockages issues of buried sewers of distinct diameters in real-time with substantial methodology.

Computer vision in consideration of automated and robotic systems has come up as a steady and robust platform in sewer maintenance and cleaning tasks. The AI revolution has enhanced the ability of computer vision and is being used to detect problems with underground sewer pipes, such as blockages and damages. A large amount of appropriate, validated, and labeled imagery data is always a key requirement for learning AI-based detection models to generate the desired outcomes. In this paper, a new imagery dataset S-BIRD (Sewer-Blockages Imagery Recognition Dataset) is presented to draw attention to the predominant sewers’ blockages issue caused by grease, plastic and tree roots. The need for the S-BIRD dataset and various parameters such as its strength, performance, consistency and feasibility have been considered and analyzed for real-time detection tasks. The YOLOX object detection model has been trained to prove the consistency and viability of the S-BIRD dataset. It also specified how the presented dataset will be used in an embedded vision-based robotic system to detect and remove sewer blockages in real-time. The outcomes of an individual survey conducted at a typical mid-size city in a developing country, Pune, India, give ground for the necessity of the presented work.