Wireless Sensor Network for AgriTech Applications
1 like348 views
The document presents an overview of wireless sensor applications in agritech, focusing on the Internet of Things (IoT) enablers and wireless sensor networks (WSN) for agriculture. It discusses various sensors and technologies for precision farming, automated irrigation, and environmental monitoring, alongside case studies demonstrating practical implementations. Furthermore, the document highlights advancements in agricultural methodologies, such as soil nutrient estimation and automated fertilization systems using IoT solutions.
Wireless Sensor Network for AgriTech Applications
- 1. Wireless Sensor Applications for AgriTech Applications 9th Feb. 2019 Presented at TiE IoT Forum – AGRITECH@2019 Rashtreeya Sikshana Samithi Trust R V College of Engineering, Bengaluru-59 (Autonomous Institution affiliated to Visvesvaraya Technological University-Belagavi) Dr. K N Subramanya Principal, RV College of Engineering, Bengaluru subramanyakn@rvce.edu.in
- 2. Presentation Outline 1) IoT Enablers 2) Wireless Sensor Networks in Agriculture 3) Case Studies
- 3. IoT Enablers RFID Transceiver EnablingTechnology Implementation Connectivity Deep Learning IIoT Agriculture Transportation HealthWater Recourse Power Bluetooth Wifi Smart City
- 4. Connectivity Layers Services Local Connectivity Global Connectivity Service Providers IoT Management Gate way Gate way Gate way Power Water Recourse Internet
- 6. Technologies Enabling IoT • Embedded Systems • Cloud Computing • Communication Protocols • Wireless Sensor Networks • Integrating Applications • Analytics and Visualization
- 7. Types of WSN Terrestrial WSNs Underground WSNs Underwater WSNs Multimedia WSNs Mobile WSNs Hybrid WSNs
- 8. WSN in Agriculture W(Wireless)--- S(Sensor) --- Air, Soil & Water + Others (Experts, Market, Information, Technologies and Infrastructure ) N(Networks)---
- 9. Wireless Technologies in WSN for IoT Communication through radio frequency (RF), WiFi, ZigBee, SigFox , LoRaWAN etc Optical Wireless Communication Light ( LiFi ) --- Future Bluetooth Communication InfraRed Communication
- 10. Wireless Technologies in WSN
- 11. Sensors in WSN for Agriculture Can we have sensors that help in providing data for predictive analytics like knowing the possible price of a vegetable or a crop even before sowing the seed? Can we have technology support for Automated Urban Farming based on health condition of a Family Hydropinics Acquaponics Aeroponics Can we have sensor that help in precision farming? How can we measure force exerted by roots to absorb water How can we sense Light Reflectiveness of a soil How can we sense the Fertigation precision to roots and leafs How can measure soil air permeability ( singular and dynamic ) Optical sensors for measuring soil properties Location Sensor (Precise positioning is the cornerstone of precision agriculture) Output from sensor will help in Yield Monitoring, Weed Mapping, Variable Spraying, Variable Rate Fertilizers, Salinity Mapping and Guidance System with Topography and Boundaries
- 12. CoE in Macroelectronics Activities under IDRC: 38 Training/ workshops/seminars/ conferences 400 faculty of RVCE/ other institute 3 International Conferences 18 Faculty are pursuing PhD 16 Elective courses are added in the curriculum related to this domain 32 UG/PG student projects 6 external Researchers used facility 20 Projects – Seed money grant 3 Ongoing projects Under technology transfer : Provided Solar lighting to the old age home in the adopted village Commercialization( under process) • Non-Invasive B.P Monitoring Device • Design and development of electrospinning system for fabrication of nanofibres Facilities IDRC: Software tools like Material studio, Silvaco, Comsol etc.. Fabrication : PECVD, Sputtering, Laser writer etcc Characterization: XRD, SEM, AFM etc…
- 13. Networks in WSN for IoT based Agriculture Connected Farming -Farmers, Farms and Consumers -Open Market -Fair Pricing Building Local Networks which helps to build a ecosystem Could it be in range of 5kms or 5000 kms ( Techniques are same except the communication part)
- 14. Case Study -1 • Check the air quality relative to standards or limit values • Detect the importance of individual sources • Collect data for the air quality management • Observe trends (related to emissions) • Develop warning systems for the prevention of undesired air pollution episode.
- 15. Design Methodology • The proposed architecture consists of sensors which is connected to Arduino Board. • The sensors that will be used for analysing the air quality pertain to the following gasses: Carbon Dioxide (CO2), Carbon Monoxide (CO), Hydrogen Sulphide (H2S), Nitrogen Dioxide (NO2), Oxygen (O2), Sulphur Dioxide (SO2) etc. • Each sensor is industry grade and is capable of measuring the air quality with maximum accuracy. • The data is sent to the thingspeak cloud using wifi module. • Based on the analysed data for contamination level a notification will be sent in the form of E-mail. Architecture Diagram Arduino Pin diagram
- 16. Thing Speak – Cloud platform High Level design System Architecture High Level design Hardware Configuration
- 17. Detailed Design Programming Language & Platform Selection • ThingSpeak enables sensors, instruments, and websites to send data to the cloud to store in a channel • Once data is in a ThingSpeak channel, we can analyze and visualize it, calculate new data, or interact with social media, web services, and other devices • Combine data from multiple channels to build a more sophisticated analysis • Pushingbox is a cloud that can send notifiction based on API calls • From one request , we can send several notifications
- 18. Experimental results and analysis
- 19. Graphical analysis of Humidity Graphical analysis of Temperature Graphical analysis of Carbon Dioxide
- 20. Graphical analysis of Carbon monoxide Graphical analysis of Pressure
- 21. To water plants with the use of devices like raspberry pi, Moisture sensor, Relay. While Python programming language is used for automation purpose. • To get Weather data predicted based on data set that is received from the metrological site. • To do Machine learning code execution on edge device to analyse the parameters. • To enable this system to contributes an efficient and cheap automation irrigation system. • To Implement Monitoring system based on wireless communication technology has been developed to control remotely, which realizes the measurement of rain fall, soil parameters. Case Study – 2 Irrigation System using Weather prediction
- 22. Methodology Weather data Analysis Algorithm: • Irrigation System uses Decision tree algorithm to predict weather. • Decision tree is one of the many machine learning algorithms and it is like a decision tool for prediction. The classifier has these steps: • collect data • train classifier • make predictions
- 23. Implementation System Architecture Models 1. Raspberry Pi Module 2. Firebase Module 3. Android Application Module
- 24. Experimental Analysis Experimental Dataset • Data set include the date and precipitation and rain status of previous days. • The precipitation identifies the chances of rain of that day. The Rain columns denotes binary value that is for rain coming or not. • If the value is 1 that means the rain will come on that day. These data can be obtained from the metrological site of specific location. • After these data will given to the raspberry pi as in the form of the excel sheet. Results
- 25. • To develop an interface module which mixes the Rapitest capsules with water. • To develop an apparatus which mixes the soil samples with water. • To display the completion of mixing Rapitest capsules, soil and water on LCD. • To obtain different colour patterns from Rapitest reactions. • To capture the image in smartphone camera after Rapitest reactions and to know the estimation of NPK in soil using an Android app. Case Study – 3 Development of IoT based Soil NPK Estimation Module
- 26. Block diagram of Soil NPK estimation module Fig. 1: Block diagram of Proposed system
- 27. Methodology • DC motors are used to control the capsule mixing, soil mixing and for stirring in the transparent beaker. • Solenoid valve is used to control the flow of water. • LCD is used to display the time required for capsule, soil and water mixing in the beaker. • LEDs are used to display the process completion. • The capture of images of Rapitest reactions using Smartphone camera. • An Android app namely Know Your Soil is used to give the details about the estimation of macro-nutrients present in soil for the captured images. Results :Developed Module Fig 2: Developed Module
- 28. NPK Estimation Step 1 Fig 3: NPK Estimation Fig 4: Soil ,Capsule and Water Mixing in the Module Step 2 Fig 5: Detection of Nitrogen in the Soil Sample Fig 6: Detection of Phosphorous in the Soil Sample Fig 7: Detection of Potassium in the Soil Sample
- 29. Android App-Know Your Soil Recommendation Table Fig 8:Android App Recommendation for Soil Sample. Validation Results
- 30. At present, robots are used for agriculture for the following processes Case Study - 4 Automatic Fertilizer Dispensary System Objectives: To determine the available nutrients (Nitrogen, Potassium, Phosphorous) in the soil. To identify the deficient nutrients in the soil. To determine the right amount of fertilizers to be added to maintain the soil fertility. To automate the entire process of addition of fertilizers.
- 32. Soil Test Colour Charts Results after Soil Test
- 33. COLOR SENSOR Individual RGB color detected Vcc = 5V Serial 25 byte data output for complete RGB values COLOR SENSOR SPECIFICATIONS
- 34. Factors affecting output obtained: Distance of the object from the sensor affects the quality of output Enclosure for the sensor Size of the object TESTING OF COLOR SENSOR
- 35. INTERFACING OF SOLENOID VALVE Relay connection with arduino Interfacing of solenoid valve with relay and arduino