“…Fog nodes can operate independently, allowing critical operations to be completed locally during network outages. Cloud-based redundancy protects important data and computing processes, improving robotic infrastructure reliability [23].…”
Robotics has been transformed by machine learning (ML), enabling intelligent and adaptive autonomous systems. By delivering massive computational resources and real-time data, fog/cloud computing and the Internet of Things boost ML-based robotics. Intelligent and linked robotics have emerged from fog/cloud computing, IoT, and machine learning. Robots using distributed computing, real-time IoT data, and advanced machine learning algorithms could alter industries and improve automation. To maximize its potential, this revolutionary combination must overcome several obstacles. This paper discusses the benefits and drawbacks of integrating technologies. It offer rapid model training and deployment for robots ML algorithms like deep learning and reinforcement learning. Case studies demonstrate how this combination might enhance robotics across industries. This study discusses the benefits and drawbacks of fog/cloud computing, IoT, and machine learning in robots. We propose solutions for security and privacy, resource management, latency and bandwidth, interoperability, energy efficiency, data quality, and bias. By proactively addressing these difficulties, we can establish a secure, efficient, and privacy-conscious robotic ecosystem where robots seamlessly interact with the physical world, improving productivity, safety, and human-robot collaboration. As these technologies progress, appropriate integration and ethical principles are needed to maximize their benefits to society.
“…Fog nodes can operate independently, allowing critical operations to be completed locally during network outages. Cloud-based redundancy protects important data and computing processes, improving robotic infrastructure reliability [23].…”
Robotics has been transformed by machine learning (ML), enabling intelligent and adaptive autonomous systems. By delivering massive computational resources and real-time data, fog/cloud computing and the Internet of Things boost ML-based robotics. Intelligent and linked robotics have emerged from fog/cloud computing, IoT, and machine learning. Robots using distributed computing, real-time IoT data, and advanced machine learning algorithms could alter industries and improve automation. To maximize its potential, this revolutionary combination must overcome several obstacles. This paper discusses the benefits and drawbacks of integrating technologies. It offer rapid model training and deployment for robots ML algorithms like deep learning and reinforcement learning. Case studies demonstrate how this combination might enhance robotics across industries. This study discusses the benefits and drawbacks of fog/cloud computing, IoT, and machine learning in robots. We propose solutions for security and privacy, resource management, latency and bandwidth, interoperability, energy efficiency, data quality, and bias. By proactively addressing these difficulties, we can establish a secure, efficient, and privacy-conscious robotic ecosystem where robots seamlessly interact with the physical world, improving productivity, safety, and human-robot collaboration. As these technologies progress, appropriate integration and ethical principles are needed to maximize their benefits to society.
“…Additionally, to prevent inefficient massive data aggregation to centralized cloud computing in the conceived ultra-largescale e-Health system with the rapid increase in the number of users, fog computing has emerged, since it is located physically closer to users [4]. To integrate the aforementioned strengths of cloud and fog computing, the fog-cloud hierarchical structure, as well as efficient fog-cloud resource management, has provided a glimmer of hope to support high portability and automatic provisioning for future e-Health development [5]. There are plenty of efforts for resource management among fog and cloud nodes, e.g., resource mapping [6] and task scheduling [4], where task classification is essential to identifying the features of both computing nodes (e.g., computational capacity, potential latency, etc.)…”
Section: Introductionmentioning
confidence: 99%
“…Support vector machine (SVM)-based task classification which is efficient in handling the defined latency-sensitive critical tasks is proposed. It is necessary to note that although deep learning algorithms increasingly gain markets, shallow machine learning (e.g., SVM) with low computational costs still presents strengths for latency-sensitive e-Health applications [5].…”
Fog–cloud-based hierarchical task-scheduling methods are embracing significant challenges to support e-Health applications due to the large number of users, high task diversity, and harsher service-level requirements. Addressing the challenges of fog–cloud integration, this paper proposes a new service/network-aware fog–cloud hierarchical resource-mapping scheme, which achieves optimized resource utilization efficiency and minimized latency for service-level critical tasks in e-Health applications. Concretely, we develop a service/network-aware task classification algorithm. We adopt support vector machine as a backbone with fast computational speed to support real-time task scheduling, and we develop a new kernel, fusing convolution, cross-correlation, and auto-correlation, to gain enhanced specificity and sensitivity. Based on task classification, we propose task priority assignment and resource-mapping algorithms, which aim to achieve minimized overall latency for critical tasks and improve resource utilization efficiency. Simulation results showcase that the proposed algorithm is able to achieve average execution times for critical/non-critical tasks of 0.23/0.50 ms in diverse networking setups, which surpass the benchmark scheme by 73.88%/52.01%, respectively.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.