A new generation of wireless sensors for integrated precise agriculture

Final Report Summary - AGRISENSACT (A new generation of wireless sensors for integrated precise agriculture)

Executive Summary:
The growing demand for more agricultural products forced farmers, in the beginning of the last century, to adopt resource-intensive and unsustainable practices which increased economic and environmental costs. Precision Agriculture (PA), one of the most significant advancement in agriculture since the advent of mechanization, allows taking into account the spatial and temporal needs of soil and crop to maximize production and profitability while minimizing risk. The AgriSENSact project results in the development of an integrated PA system which will allow precise management of the crops taking by integrating several of the main practices of PA. The proposed system is based on the AgriProbe concept, a modular device including specific modules for sensors, energy supply and communications, that can be tailored for each specific agricultural application. These allow the SMEs of the consortium to gain a competitive edge in their respective markets and to expand/consolidate their core businesses.
The main achievements of the project can be summarized as follow:
1) The targeted outputs were established at the beginning of the project (selection of application and product performance) and both the initial technical specifications for the AgriProbe and its components and the initial product concept and design were established. (WP1);
2) Three Agriprobe prototypes were assembled at different stages of the project for validation and testing. (WP2); The Agriprobe specifications and concept were continuously refined and optimized up to the production of the third prototype which was ready for field testing.
3) The Soil Sensor Module (SSM) and Atmospheric Sensor Module (ASM) were produced taking into account the project specifications and outputs (WP3); The soil nitrate thin film based sensor was produced and tested. However the lack of sensitivity of the produced sensor did not allow for reliable measurements. This sensor was replaced with a commercial PH sensor in agreement with the contingency plan.
4) A system to acquire data from the sensors embedded on the device, process and transmit them to an external Information System over a robust, secure and easy to configure Wireless Sensor Network was developed. The system also manages the energy harvesting module and power consumption, ensuring self-sustainability of the device. (WP4);
5) The thermoelectric generator for the self-sustainable AgriProbe device was developed. However a solar panel was used in the Agriprobes for field testing as the energy output of the thermoelectric generator was not enough for long term field testing (as foreseen in the contingency plan) (WP5);
7) The several Agriprobe prototypes developed in the aim of the project were tested in laboratory conditions, under controlled conditions in the field and at the Matarromera vineyards in Spain.
8) Dissemination, Exploitation and management activities were carried along the project according the the description of work. The Intellectual Property Right Plan for the use of results by the partners in the Consortium was established.

Project Context and Objectives:
With the financial crisis still continue with pressure on the European economy, the primary sector is no exception, particularly agriculture, due to the strong dependence of Europe on food imports, matters such as increase profitability of crop production, sustainable use of resources, improve product quality, ensure food security, protect the environment and maximize the use of land assume preponderance in scientific and technological development. The European agricultural sector can potentially save €20 500 million per year in energy and fertilizers and soil improvers, through the adoption of the AgriSensAct solution. These savings do not include the additional advantages of reducing the Europe’s dependency on food imports, the water scarcity, the emissions of COs, the prevention of soil erosion and deterioration of habitats, the improvement of the products quality and the contribution to the employment.

The growing demand for more agricultural products forced farmers, in the beginning of the last century, to adopt resource-intensive and unsustainable practices which increased economic and environmental costs. Precision Agriculture (PA), one of the most significant advancement in agriculture since the advent of mechanization, allows taking into account the spatial and temporal needs of soil and crop to maximize production and profitability while minimizing risk. Wireless sensors networks have recently emerged as one of the key technologies to implement precision agriculture and have already been applied to precision irrigation, the application of water to a given site in a volume and at a time needed for optimum crop production. However, up to now, the lack of adequate sensors for soil fertilizers sensing prevented the implementation of WSNs for site specific fertilization, the application of the right amount of fertilizers at the right time in the right place. The AgriSENSact project results in the development of an integrated PA system which will allow precise management of the crops taking by integrating several of the main practices of PA. The proposed system is based on the AgriProbe concept, a modular device including specific modules for sensors, energy supply and communications, that can be tailored for each specific agricultural application. These allow the SMEs of the consortium to gain a competitive edge in their respective markets and to expand/consolidate their core businesses.
The project ultimately aims to bring a new generation of cost effective and affordable modular probe for the precise agriculture with highly efficient results. To reach this goal the outcomes of the project will include 4 main groups of products, namely:
1) Integrated sensorial probe (AgriProbe product) with energy harvesting and data acquisition and transmission capacities to be integrated in WSN for the simultaneous management of precise irrigation and site specific fertilization.
2) Energy harvesting device based on thermoelectric materials technology to be implemented in the AgriProbe energy module
3) Soil nitrate sensor based on thin film technology to be incorporated in the AgriProbe soil sensorial module.
4) Contact-connection device for electricity and data transmission between the different modules of the AgriProbe product.

Project Results:
The project created an interdisciplinary platform bringing together industries, SMEs and RTD performers with diverse but complimentary background.

The Work Package 1 aimed at obtaining the technical specifications and requirements for the AgriSENSact system to be used, as guidelines, during the AgriProbe development. The initial end-users specifications established in WP1 were mainly about the geometrical requirements of the AgriProbe. Taking into account that the concept of the AgriProbe was to produce low cost probes incorporating mainly commercially available sensors (excepting the nutrient sensor) the sensors were selected from the commercially available solutions taking into account their costs, durability in service and intended use in the field and performance as described by the manufacturers. The only end-user in the project is dedicated to vineyards and it was not possible to define requirements for other crops. However, the modular design of the AgriProbe will ensure that modules specific to other crops can easily be incorporated in the probe in the future. The final consumer, especially the wine producers, which are used to spent thousands of Euros in vineyards monetarization, trough sample collection, weather stations control, individual probes, manpower, and so on, could have an excellent alternative for overcome, at least, part of those issues with less costs. From the system requirements specification the AgriProbe functional and non-functional requirements were specified and presented in a detailed and comprehensive way in the MS8 – AgriProbe Requirements report.

Regarding WP2, the development of the AgriProbe comprised 3 different prototypes that were produced sequentially. First prototype was produced in order to ensure the functional capabilities of the prototype. This prototype was produced according to the specifications established in WP1. The initial technical specifications and product concept and design established in WP1 were continuously revised and updated taking into account the results of the first prototype tests. The final integration of the AgriProbe modules and the production of the prototype for validation and testing have been achieved in the second prototype which integrated all the modules of the AgriProbe. The second prototype was carried out with two versions of the Energy module (TEG) in order to ensure successful integrations of the AgriProbe components. While the TEG (Energy module) developed at KTH was being tested in the field at Matarromera vineyards under IPN supervision an alternative TEG was developed at IPN in order to ensure and optimize the physical, electrical and communicational capabilities of the second prototype. The second prototype of the AgriProbe was tested in the laboratory and in the soil under controlled conditions and the testing results pointed out several deficiencies of the second prototype including the long term insolation of the joins between modules, the air circulation in the temperature sensor module and the long term performance of the PH sensor. On the other hand, the temperature difference achieved by the TEG (energy module) was not enough to generate and adequate amount of energy. A third prototype of the AgriProbe was then conceived and developed in order to overcome the problems detected in the second prototype. As the TEG energy output was not adequate for field testing, a solar energy panel was integrated the third prototype instead of the TEG in agreement with the contingency plan. However, the TEG development proceeded at KTH as the results obtained with the second prototype were sufficient for further development.
Under the scope of WP3, the Soil Sensor Module (SSM) was successfully developed regarding the requirements needed as electrical/electronic components for signal acquisition, processing and delivery to the communication module described in WP4. Also, three different types of soil sensors were integrated in the soil module and consequently added to the probe for field tests in the Matarromera vineyards, allowing measure four different parameters such as: soil moisture, soil temperature, pH and soil electrical conductivity. Regarding Atmospheric Sensor Module (ASM) the air humidity plus air temperature sensor was successfully placed inside the module, ready for sending data in real time. The module used was modified by drilling holes along the module surface and were tested regarding the number of holes, angle of drilling, position and diameter, which provided a helpfully understanding of which system had better results under adverse weather conditions, mostly under rain. The model chosen has been integrated in the final probe and tested in the field with success, taking into account the accuracy and reliability of the data as well the lifetime of the module and its electrical components. Taking into account both mechanical bonding and electrical (wires and contacts) needs, and subsequently the modules conception and development proceeds, the design of the modules was a success. This type of innovative modules assembling by using a secondary module as an integration module, promoted, not only that the components inside the modules didn’t suffer any mechanical solicitation (torsion, tension, etc) but also, increased the stability, mechanical resistance and waterproof resistance of the probe.
In the scope of work package 4, a detailed analysis of the existing technology and the AgriSENSact network requirements was carried and focused on the following topics:
-radio signal propagation in agricultural environments;
-network topology requirements;
-network protocols;
-antennas characteristics;
-battery technology;
-typical serial buses;
-typical board construction approach.
The information obtained in the survey of the state of the art provided the means for a hardware selection with the assurance that it would maximize the, overall, network performance.
The radio signal propagation in agricultural environments study defined, from a theoretical point of view, the system characteristics that maximized the communication performance when using SRD (short range devices) frequencies. From the conclusions it was clear that the best compromise between, communication range and the AgriProbe dimensions and antenna position, was obtained when a frequency near the 868MHz was selected, which stemmed from the fact that reducing the frequency increases the communication range but also the distance between the antenna and the ground.
A detailed assessment of the AgriSENSact communication needs provided important information regarding the network topology requirements when in many installation situations it is necessary to overcome various types of obstacles. From the analysed topologies it was clear that the only one that could provide a reliable network connection and at the same time simple installation procedures, was a network topology with mesh or similar characteristics.
The most common protocols were analysed, and in all of them discouraging limitations were found that prevented their use. From the study it was possible to see that some didn’t allow mesh networks and others were so complex that routing nodes were unable to enter in a low power mode, which from an energy stand of view, was unacceptable. In order to overcome this situation, a specifically designed protocol for the AgriSENSact network was developed and can be characterized as follows:
for allowing low power modes on all nodes and by providing a true mesh network, i.e. all nodes can route information;
the existence of synchronization mechanisms to provide constant network reference;
and for having several mechanisms that simplify the installation.
A survey of typical antennas provided valuable information regarding the irradiation graphs and communication ranges and with the result list of advantages and disadvantages of each type of antenna it was possible to conclude that, the only one that could provide a good communication range and, an in almost all directions coverage, was the omnidirectional antenna.
A detailed analysis of the most common battery technology used in systems with harvesting capabilities showed that the only type of technology suitable for this type of application was the one based on Lithium, not only because of the high energy density, low self-discharge and no memory effect but also because of the low maintenance requirements.
A comprehensive study of many serial protocols was conducted and provided valuable information regarding the requirements necessary to develop the AgriProbe bus. Of the analysed buses, only the I2C provided easy integration of commercial off the shelf digital sensors while at the same time required a reduced number of communication lines.
With the system requirements in mind a simple and versatile AgriProbe bus was developed, where the data communication with the acquisition boards was assured by the I2C, but also with the inclusion of two power lines (Vbat and 3.3V) that provide power to the sensorial elements of the AgriProbe. This type of solution provided the necessary bases for implementing the characteristics identified during the system requirements survey, where it was defined that the AgriProbe should be capable of providing power and control over bus and have hot-swap and plug-and-play capabilities with minimal hardware requirements.
A study comparing the advantages and disadvantages of developing the data process and transmission board using a typical SOC (system-on-chip) IC (integrated circuit) and a separated processing and transmission solution, was conducted.
Both approaches were exhaustively tested in laboratory conditions and the modular approach showed the best communication results.
A power management board capable of collecting energy from both photovoltaic and Peltier cells at the same time or from each one separately was developed. In order to guarantee a correct recharging process of the secondary cells, numerous security mechanisms that control the charging speed, the maximum charging value, the cell temperature and load balance, were implemented.
Using the information gathered during the batteries study, the power management for self-sustainability board, was developed and prepared to use typical Lithium batteries and is able to provided more than the AgriProbe required energy to operate correctly independently of the type of energy source used (TEG or/and solar panels. The developed board can be seen in Figure 1.

Figure 1-Power management board.
A versatile data acquisition board capable of acquiring sensorial information from any type of analogue sensor was developed and successfully integrated in the AgriProbe bus. Developed with the AgriProbe characteristics and features in mind, this board can be characterized for having a wide range of addresses, which allows, for up to, 16 analogue sensors to be connected, to the AgriProbe bus, at the same time. In Figure 3, it is possible to see the final version of the data acquisition board, where an impedance matching circuit was included in order to provide support for high impedance sensors.

Figure 2-Data acquisition board.

In order to address the need to integrate digital serial sensors, a flexible serial data acquisition interface board was developed and used to integrate electrical conductivity sensors. Figure 2 depicts the serial interface board layout.

Figure 3-Data acquisition PCB layout (top and bottom layers).

Following the conclusions obtained during preliminary tests, a modular approach was used to develop a data process and transmission board with outstanding communication characteristics. Using the knowledge obtained during laboratorial tests, the transmission section of the board used a special type of PCB (printed circuit board) with very specific characteristics which minimized the radio interference and increased the overall communication performance. In Figure 3 it is possible to see the final version of the data process and transmission board.

Figure 4-Data process and transmission board.

Using the conclusions of the network protocol study, a versatile WSN communication protocol was developed. The developed protocol was divided in two distinct levels, the low and the high level. The low level part of the protocol provides a communication abstraction layer between the radio and the high level part of the protocol and is responsible for encapsulating high level messages and maintaining a reliable point-to-point link between two nodes. Using radio specific interruptions and binary message type checks, it was possible to achieve a very high speed data transferring with reliable packet deliverable. The high level part of the protocol consists of a set of messages used to control the AgriSENSact network operation.
The high level part of the WSN communication protocol consists in the following messages:
• Synchronization message;
• Configuration message;
• Authorization message;
• Data message;
• First connection message.

The synchronization message provides the means necessary to maintain the network synchronized by compensating eventual node drifts.
The WSN configuration message allows for a complete node configuration where parameters such the acquisition period, sleep time, max number of failed communications, deep sleeps time, node ID, and others can be changed.
Authorization messages provide the means for new AgriProbes to correctly join a specific AgriSENSact network and are used to transfer the network encryption key.
The first connection message is used by AgriProbe during the installation to notify the coordinator that a new node has joined the network and require configurations, but also to establish a valid network communication routing path.
In order to simplify the network installation a simplification mechanism was developed where the entry point for given network can be any node in the vicinity of the node been installed. When the node been installed connects to a standard node in his vicinity it forces the addition of a new routing entry in all the subsequent nodes until the coordinator is reached. This type of mechanisms of automatically adding new routing paths and auto-negotiate with the coordinator to the network configurations means, that no technical user intervention on the installation site is required.
Within the WP5, regarding the power generation scheme Thermoelectricity has been considered a promising candidate for powering remote sensors systems. In this project the geothermal and ambient heat was used to create a thermal gradient across a Peltier device to generate power, that can be used either to directly power up the AgriProbe communication with the communication center, or re-charge a battery that can provide the energy for the desired communication. Several prototypes were developed for the TEG power generator module, where the heat exchange plays a very significant role, yet has not been considered as a significant task at the beginning of the project.
During the project, with the use of developed prototypes we have monitored the temperatures in air and soil for various seasons and could demonstrate that 5-10 oC difference was successfully generated using an Al heat exchange casing around the Peltier device.
Under the scope of WP6, the AgriSENSact validation process followed a typical two testing stages where the first stage focused on the AgriProbe device and the second on the AgriSENSact wireless sensor network.
The first testing stage was performed in laboratory conditions and focused in the hardware validation regarding the reliability and overall performance.
The second testing stage was performed in real field conditions and focused mainly on the software developed and provided valuable information regarding the overall system performance, efficiency and network reliability. Field tests were performed in two different places, one at IPN (Figure 5) and the other at Matarromera vineyard in Valbuena de Duero (Figure 6), Spain.

Figure 5-IPN AgriProbe testing.

Figure 6-Matarromera installation.
At IPN, a detailed assessment of the AgriProbe and the AgriSENSact WSN behaviour in real field conditions was conducted and provided reliable information concerning the developed solution. At Matarromera vineyard, a long term pilot of the AgriSENSact WSN was installed and is expected to provide valuable information regarding the network behaviour over time, where not only is expected to endure different weather conditions (winter, summer) but also some landscaping changes with vines growth.

Potential Impact:
The WP7, Dissemination and Exploitation activities, was during this period in launching all the dissemination activities as well as providing the Intellectual Property Right Plan for the use of results by the partners in the Consortium, both objectives of the DoW. Many outcomes of the WP have been achieved during this reporting period, some of main activities are outlined as follows in
• Project results have been identified and developed;
• Strategy for IPR protection (patents and others) are being up-dated – Deliverable D7.2 and Deliverable 7.4;
• Policies in regard the IPR protection and disclosure of any relevant information have been included in the new Consortium Agreement.
• Final version of the exploitation plan has been developed;
• Future actions needed to achieve TRL 9;
• Consortium plan for future commercialization;
• Future funding was developed – D7.4.
• Future dissemination strategy was defined (see D7.4);
• Webpage of the Project;
• Agrisensact Video;
• Flyers and Roll-up banners;
• Articles and magazines;
• Prototypes for demonstration and dissemination of results;
• Appropriate and quality materials for dissemination have been elaborated – Deliverable 7.3 – 65 page document.

List of Websites:
Website: www.agrisensact.eu
Contact details:
Grandesign: Rui Curado: r.curado@cadflow.pt
Matarromera: Ángela García Álvarez : olivocultura@emina.es;
KTH: Muhammet Toprak : toprak@kth.se;
VPS: Rodrigo Ferreira : rferreira@vps.energy; Luisa Matos: lmatos@isaenergy.pt;
Advamat: Martin Daněk : danekma2@fel.cvut.cz
IPN: Ana Manaia: ana.manaia@ipn.pt

Final Report Summary - AGRISENSACT (A new generation of wireless sensors for integrated precise agriculture)

Condividi questa pagina

Scarica