Provisional Access of Workflow Scheduling With Mobile Agents in Agricultural Application
N. Priyadharshini1, V. Narayani2
1N. Priyadharshini, Assistant Professor, Department of Computer Science at Sree Saraswathi Thyagaraja College, Pollachi, Tamil Nadu.
2Dr. V. Narayani, Assistant Professor, Department of Computer Science at St. Xavier‟s College, Palayamkotai, Tamil Nadu.
Manuscript received on March 12, 2020. | Revised Manuscript received on March 25, 2020. | Manuscript published on March 30, 2020. | PP: 3198-3207 | Volume-8 Issue-6, March 2020. | Retrieval Number: F8048038620/2020©BEIESP | DOI: 10.35940/ijrte.F8048.038620
Open Access | Ethics and Policies | Cite | Mendeley
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: In mobile computing, if agent experiences are assessed and available among end communication for environmental modelling, it helps in improving exploration load for unknown or unvisited circumstances. Therefore, it may speed learning procedure. As, building an accurate and effectual model with constraint time is also an essential factor, specifically for difficult conditions, this work initiates reinforcement model base learning approach based on work flow scheduling to acquire lesser memory consumption and effectual modelling. Here, two methods have been compared for attaining a real time experience and to produce virtual experiences as elapse time in learning process is reduced. However, this two modelling is appropriate for knowledge sharing. This analysis is inspired with knowledge sharing concept in multi agent based systems where agents has the competency to generate global modelling from scattering these models provided by individual agents. Subsequently, it may increase accuracy modelling; therefore it may offer valid experience for learning at earlier learning stage. To reduce make span process, anticipated model uses cost, reward and action techniques to grafting workflow scheduling need and resourceful experience from experienced system indeed of merging entire model. Simulation outcomes depict that anticipated scheduling model can acquire sample learning and efficiency model based acceleration in Multi-agent application objectives. Here, MATLAB environment is used for simulation. Metrics like cost, Make span is evaluated for agricultural dataset. Comparison is done with anticipated Dense mobile Network and Deep Q Network. Here, DMN shows better trade off than DQN model and more appropriate for agricultural dataset.
Keywords: Multi-agent, Mobile computing, agriculture, Dense Mobile network, Deep Q network,
Scope of the Article: Algorithm Engineering.