Cognitive Optimal-Setting Control of AIoT Industrial Applications With Deep Reinforcement Learning

Lai, Ying-Hsun; Wu, Tung-Cheng; Lai, Chin‐Feng; Yang, Laurence T.; Zhou, Xiaokang

doi:10.1109/tii.2020.2986501

Cited by 27 publications

(7 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The popular AIoT technology combines the IoT with AI technology to create numerous smart applications, such as smart homes, smart enterprises, and even smart cities (Gubbi et al, 2013 ; Lee and Lee, 2015 ; Lai et al, 2021 ). The diverse sensing technologies of AIoT and AI programming learning are also compatible with various educational strategies in engineering education, maker learning, project-based learning, and problem-oriented learning (Navghane et al, 2016 ; Lensing and Friedhoff, 2018 ), thus enabling students to integrate sensor applications with AI algorithms in order to create different smart applications and solve practical problems.…”

Section: Introductionmentioning

confidence: 99%

Exploring Computational Thinking Skills Training Through Augmented Reality and AIoT Learning

et al. 2021

Self Cite

View full text Add to dashboard Cite

Given the widespread acceptance of computational thinking (CT) in educational systems around the world, primary and higher education has begun thinking about how to cultivate students' CT competences. The artificial intelligence of things (AIoT) combines artificial intelligence (AI) and the Internet of things (IoT) and involves integrating sensing technologies at the lowest level with relevant algorithms in order to solve real-world problems. Thus, it has now become a popular technological application for CT training. In this study, a novel AIoT learning with Augmented Reality (AR) technology was proposed and explored the effect of CT skills. The students used AR applications to understand AIoT applications in practice, attempted the placement of different AR sensors in actual scenarios, and further generalized and designed algorithms. Based on the results of the experimental course, we explored the influence of prior knowledge and usage intention on students' CT competence training. The results show that proposed AIoT learning can increase students' learning intention and that they had a positive impact on problem solving and comprehension with AR technology, as well as application planning and design.

show abstract

Section: Introductionmentioning

confidence: 99%

Exploring Computational Thinking Skills Training Through Augmented Reality and AIoT Learning

et al. 2021

Self Cite

View full text Add to dashboard Cite

show abstract

“…where, cost (j) i = [s ji , s i ] represents the row vector of the jth row of the matrix and s ji ∈ S c , s i ∈ S s . Let the identity matrix be ) , I (2) , … , I (j) , … ,…”

Section: Pruning With Linear Weight Strategymentioning

confidence: 99%

An effective resource scheduling model for edge cloud oriented AIoT

Liang

Cui

Cao

et al. 2021

Concurrency and Computation

View full text Add to dashboard Cite

Artificial Intelligence of Things (AIoT) is a new research area in AI and IoT. For the massive data and connections at the edge of AIoT, how to schedule the resource load is a key problem to be solved. For the multi-objective resource scheduling problem in a cross-domain environment, this article proposes a resource scheduling method based on edge computing and multi-objective algorithms. First, a cloud-edge hybrid AIoT hierarchical network resource management architecture is constructed. Then, the resource scheduling problem is modeled and a linear weighting strategy is designed for search space "pruning", which is then optimized by three multi-objective algorithms. Finally, we conducted a series of experiments based on the iFogSim platform. The simulation results show that the proposed resource scheduling strategy can effectively reduce the processing delay of AIoT and effectively improve the energy utilization efficiency of devices in the network.

show abstract

“…In the last decade the AI field has experienced explosive growth, partially due to Deep Learning (DL) and Deep Neural Networks (DNN) with ground-breaking applications in multiple domains. Combining AI and IoT networks and devices, is a subject that has attracted the attention of both academic and private research interests (Zhang and Tao, 2020), while the application of AI algorithms to the IoT domain has been described as Artificial Intelligence-of Things (AIoT) (Lai et al, 2021). A few examples of the successful application of Deep Learning based AI methods in the IoT domain include either use-case or device specific applications such as wearable devices (Ravì et al, 2017), patient rehabilitation systems (Fan et al, 2014) and smart energy meters (Alahakoon and Yu, 2016).…”

Section: Introductionmentioning

confidence: 99%

IoT threat mitigation engine empowered by artificial intelligence multi-objective optimization

Mpatziakas

Drosou

Papadopoulos

et al. 2022

Journal of Network and Computer Applications

View full text Add to dashboard Cite

Cognitive Optimal-Setting Control of AIoT Industrial Applications With Deep Reinforcement Learning

Cited by 27 publications

References 11 publications

Exploring Computational Thinking Skills Training Through Augmented Reality and AIoT Learning

Exploring Computational Thinking Skills Training Through Augmented Reality and AIoT Learning

An effective resource scheduling model for edge cloud oriented AIoT

IoT threat mitigation engine empowered by artificial intelligence multi-objective optimization

Contact Info

Product

Resources

About