Hierarchical Control Policy Research Articles

An autonomous radiation source detection policy based on deep reinforcement learning with generalized ability in unknown environments

Autonomous radiation source detection has long been studied for radiation emergencies. Compared to conventional data-driven or path planning methods, deep reinforcement learning shows a strong capacity in source detection while still lacking the generalized ability to the geometry in unknown environments. In this work, the detection task is decomposed into two subtasks: exploration and localization. A hierarchical control policy (HC) is proposed to perform the subtasks at different stages. The low-level controller learns how to execute the individual subtasks by deep reinforcement learning, and the high-level controller determines which subtasks should be executed at the current stage. In experimental tests under different geometrical conditions, HC achieves the best performance among the autonomous decision policies. The robustness and generalized ability of the hierarchy have been demonstrated.

Identification of Waste Handling Compliance of Sharp Objects at Universitas Gadjah Mada Academic Hospital Yogyakarta

Background: Health care processes that occur in the hospital cannot be separated from factor control and prevention of infection. One of the biggest challenges of the current hospital is to be able to reduce the number of infections that occur during the delivery of health services in progress.Objective: This research aimed to identify compliance with the infection control policy concerning health workers and sharp objects in order to prevent spread of infection and to decide upon the chain of infection transmission in Universitas Gadjah Mada Academic Hospital, Yogyakarta, Indonesia.Design: This research was a descriptive analytic study. Data were collected by observation methods using a survey instrument in the form of a check list. The processing of data involved factor analysis and narratives in descriptive form with percentages and tables.The study population involved total sampling of 30 percent of the total measures undertaken by syringe healthcare workers who work in the ER, Intensive Care Installation, Operation Room and Midwifery Care Lounge. The sample set included 40 activities. The Observation Sheet was based on the Guide PPI from the 2011 Nurses Association of Infection Prevention and Control.Results: The incidence of needle stick during January - April 2015 increased significantly to 4 events. Rate of compliance officers in managing sharps waste was: 67.5% with 25 % of corporate compliance officers not doing recapping or closure of needles (minimum scale). A sharp object was not given to others as much as 82.5% (Medium scale). When intending to give a sharp object to another person then the container was used as much as 57.5% (minimum scale). Sharps waste were put into a special puncture resistant and waterproof box 97.5% (scale of Good). Safety box 3/4 full of sharps waste was sealed or sealed and thrown into the garbage storage area as much as 75% (minimum scale).Conclusions: The availability of adequate infrastructure, e.g. Safe Injecting SPO, SPO sharps waste management, availability of containers for storing syringe when it will be handed over to another officer, and the availability of adequate safety box needs to be improved. Workers should strictly follow the NSI (Needle Stick Injury) prevention and control program and execute compliance based on SPO guided by the universal precautions and hierarchical control policy.

Hierarchical control policy for dynamic resource management in grid virtual organization

This paper proposes a hierarchical control system in grid virtual organization. The hierarchical system can be decomposed into multiple application groups, which can be further decomposed into multiple applications. At the top of the hierarchy, the global controller controls the gross allocation of resources to the groups. At the next level down, the group controller coordinates the local deployments of all applications that consume the local allocation of resources. At the lowest level, the local controllers adjust the local resource usages to optimize the utility of single application. The hierarchical control system considers all applications and coordinates all layers of grid architecture upon any changes. According to different time granularity, we adopt a different control scheme. The global control considers all applications and coordinates three layers of grid architecture in response to large system changes at coarse time granularity, while local control adapts a single application to small changes at fine granularity. This paper adopts utility-driven cross layer optimization for grid applications to find a system wide optimization and solves the cross-layer optimization by using pricing based decomposition. A set of hierarchical utility functions is used to measure the performance of the grid system that follows the system, group and application hierarchy. This paper uses total utility to measure the overall quality of grid system. The experiments are conducted to test the performance of the hierarchical control algorithms.