Original papers
Context-aware control and monitoring system with IoT and cloud support

https://doi.org/10.1016/j.compag.2019.03.005Get rights and content

Highlights

  • IoT platforms needs to recognize the context in which applications are operating.

  • Competitive multi-layered hierarchical architecture of a context aware system.

  • Real-time control and monitoring of a process through an IoT platform.

  • A novel semantic interoperability architecture at the level of Sensor-Cloud interface.

  • Focusing on ensuring compatibility with requirements of IEC 61499 standard.

Abstract

The main goal of the paper is to integrate three emergent technologies (Internet of Things, Cloud Computing and Context awareness) in a multi-layered architecture for developing real-time process control agriculture application. For this, the paper presents original solutions for a Control and Monitoring unit (CMU) which performs real-time control as an entity running on an IoT platform, a Context-aware Control Platform (CaCP) with a three-tier architecture, serving as middleware mechanism for interfacing environmental sensors with IoT and Cloud and a four-level architecture to perform agriculture process control, that includes the CMU and CaCP modules. These solutions are validated by a case study application implemented on an IBM Bluemix IoT platform which performs automatic control of an irrigation system using context aware adaptation of controller parameters in response to environmental changes.

Introduction

“Internet of Things” (IoT) is a concept that has become increasingly disputed in the past decade and half. Although this concept does not have a definition which has been widely accepted in the engineering field until the present moment, it refers to a network of objects (devices) interconnected in a wireless network that have the ability to communicate and exchange information between them. Such networks of interconnected equipments began to be increasingly used nowadays in industrial applications.

This trend began to spread extensively in automation when it was found the need to introduce automation equipment in hospitals, office buildings, housings, manufacturing enterprises and so on. In particular, IoT is being used more and more often in association with precision agriculture (PA) technologies, for example sensing technologies for automated data collection and yield monitoring. One of these is nutrient management, often linked to the variable rate application (VRA) fertilizer spreading.

IoT platforms used in PA must be designed considering both the increase in the number of interconnected devices and the particular demands of users in different applications contexts. Therefore, an IoT platform will be successful only if it recognizes the context in which applications are operating.

This becomes possible by using context-aware computing, which acquires and analyzes relevant context information, and produces appropriate decisions to respond at contextual changes.

Context-aware systems (CAS) are systems that are able to adapt their operations to context changes without explicit user intervention. A CAS platform must explicitly provide by its components’ functionalities, context information and management action and also offers services to users using context information where relevance depends on the user’s task.

So, a context-aware platform can be considered a middleware support which allows the transfer of environmental information from the lowest infrastructure level to a higher level for interpretation and decision. This multi-layered architecture is typical for Cloud computing hierarchy, which allows to place a middleware platform as part of a Sensor-Cloud interface in the PaaS (Platform as a Service) layer.

The aim of the paper is to integrate these three advanced technologies (IoT, CAS and Cloud) to allow the real-time control of a remote automatic system. For reasons which will be thoroughly grounded as proof of the concept has been chosen an application designed to monitor important parameters in precision agriculture.

Section snippets

Related work

Since our work aims at integrating in the field of control systems three technologies (Internet of Things, Context-awareness, and Cloud computing) we chose to analyze for the state of the art recent works (published in the last three years) in each area but related to at least one of the other two. At the same time, we searched for works that refer to precision agriculture applications, and as far as possible in the field of irrigation and fertilization.

Paradoxically, although Precision

The architecture of the industrial system

Our goal was to ensure control of an irrigation system designed for small and medium size farms (1–5 ha) and monitor parameters related to soil properties, crops’ type and essential environment factors in order to achieve improved results in terms of quality, quantity and costs. In this regard, we rely on combining the specific advantages of several modern technologies. Specifically, the classic solution for automatic control of technological installation for irrigation performed by a central

Context-aware middleware platform

The CMU ensures maintaining the prescribed value (with deviations within acceptable limits) of three process variables: pressure, flow and level. However, changing the confinement values can be done automatically, based on the context information.

In the discussed application the values of the four parameters of context are considered: two for soil (soil pH and soil conductivity EC) and two for the external environment (air temperature and wind speed).

Retrieving context information from sensors,

Implementation of Internet of Things application

At the moment, there are many “Internet of Things” development environments made available by large companies offering Internet and communication solutions (Google, Microsoft, IBM etc.). The appropriate choice of an IoT platform is a challenge for everyone. Above all, it must be known that the chosen environment will not represent the user’s desired IoT application itself. Such an environment represents a set of algorithms, data engines, application programming interfaces (API) and wireless

Experimental results

The irrigation system is in operation since 2016 in an Experimental Research Station. Several tests were performed both in simulated and real environment. Different scenarios based on specific recipes of the water and nutrient mixture were performed in order to determine their efficiency. Unfortunately, the context-aware decision procedure cannot be exemplified (otherwise than by presenting program code lines, which would make the presentation excessively difficult). We chose to present only a

Conclusions

The main conclusions derived from the analysis of experimental results of the application which has been utilized as a proof of concept, for controlling and monitoring an irrigation system connected to an IoT platform. For those tasks were elaborated a hardware Command Unit (CU) and a software application running on IoT platform.

The purpose of this paper was to propose a competitive architecture of a context aware system that allow real-time control and monitoring of an industrial process by

References (25)

  • A. Calera et al.

    Remote sensing for crop water management: from ET modelling to services for the end users

    Sensors

    (2017)
  • K. Fathallah et al.

    PA-RPL: a partition aware IoT routing protocol for precision agriculture

  • Cited by (0)

    View full text