Sensor Data Fusion for Precision Agriculture

Project Title

Sensor Data Fusion for Precision Agriculture

Project Area of Specialization Artificial IntelligenceProject Summary

The purpose of this project is to develop a system for multi-sensor fusion, and integration of soil/crop properties using AI techniques (e.g., Fuzzy logic / ML, EKF). The data will be collected using static sensors and/or mobile robots including temperature, moisture, humidity, and/or resistivity. This data will be transferred to a base station which can be either static or mobile, where it will be pre-processed. After pre-processing, the data will be fused using AI techniques, possibly in multiple layers. The result, pre-processed, and raw data will be displayed to a human user (using android app), and it could be input to another system.  ?

Project Objectives

1. To design and develop automatic data acquisition system using static sensors, as well
2. as cooperative mobile robots equipped with IoT sensors.
3. To develop data pre-processing, data reduction in order to make data valuable and to
4. only transfer useful data.
5. Utilize information from spatially separated sensors of different kind, including temperature, humidity, soil resistivity and moisture.
6. Fusing data from different sensors for predication. ?
7.  To design the system to handle and store data streams and historical data efficient database;
8. ?To design android-based interface to provide visualization of real-time data and data analysis

Project Implementation Method

1. Requirement Engineering: To conduct systematic literature review to analyze and outline functional and non-functional requirements, output will be presented as UML diagrams.
2.  Analysis of the system: Modules of the system will be identified, Detailed Data model, ERD and other UML diagram will be developed for each module based on detailed analysis using software engineering principles.

3. System Design:  All modules of the system including hardware and software will be designed respective engineering design principles. Based on our preliminary study following module are expected to be designed.

1. Data collection: Data collection module includes Static IoT node and mobile robotic
   platforms equipped with advanced IoT sensors, including camera, moisture, temperature etc.
2. Transferring of data from sensor to end user via base station?
   • To transfer data collected from sensor to base station. ?
   • To transfer data from base station to server. ?
   • Transferring of data from sensor to end user via base station
   • To Transfer data from server to end user. ?
   • Transfer data from server to end user through push or pull
3. Contextualization, Data Reduction and Transmission module: It will append the contextual information to the sensor data and delete redundant and noisy data. It will also transfer data from sensor to the Fog and Cloud.
4. Data Pre-Processing: Quality of data is a critical factor in the success of data analytics and decision making. The data collected by moving and static sensors may contain discrepancies and inconsistencies.
5. Data Fusion?

•  Increasing degree of truth by generating output based on rate of input occurring rather than binary patterns. ?
• Prediction will be given for future course of action. ?
• Predictions will be based on sensor data. ?
• Non-linear data will be converted into linear data. ?
• Noise from data will be removed. ?
• End-User Module?
• Real-time data will be displayed to end user.?
• Update soil properties to end user on android app.?
• Soil data will be displayed according to categories.?
• Soil data will be displayed on graphs.?
• Soil data will be displayed in tables according to categories. 

      6. Visualization: The visualization of the output is very important because the famers and other stakeholder must be able to view results

4. System Evaluation: The module will be tested using comprehensive Test plan which includes unit and integration testing.
5. Documentation and dissemination: All the modules will be documented in light of
software engineering techniques. The project progress will be shared at university and national level.

Benefits of the Project

1. To improving crop monitoring by efficient fusion of data from different source and sensors.
2. To collect and combine data from multiple sources
3. The sensors fusion may increase in crop yield
4. It may sustain the economic viability of farm operations;
5. satisfy and improve human food and fiber needs;
6. it may improve in help in better crop management
7. helps in decisions to be made when, where and what to cultivate

Technical Details of Final Deliverable

1. IoT and robots based data acquisition system which includes various IoT agricultural sensors and aerial / ground robots equipped with IoT
2. sensors and static IoT nodes. The main sensor we plan to use in camera, though we will acquire data form other sensors as well.
3. Communication Module: Mechanism for Data Transfer suing Wifi/ Bluetooth/ LoRA. It will be implemented using Raspberry-pi
4. Sensor Fusion: This module will fuse data from various sensors and output could be useful for predictions and data analytics
5. Visualization: Aneroid application to visualize sensors data and data analysis
6. Project documentations, including SRS, Design documents, testing and deployment, presentation and/or posters.

Final Deliverable of the Project HW/SW integrated systemCore Industry AgricultureOther Industries IT , Food , Manufacturing , Telecommunication Core Technology Artificial Intelligence(AI)Other Technologies Internet of Things (IoT), Robotics, Cloud Infrastructure, Big DataSustainable Development Goals Zero Hunger, Good Health and Well-Being for People, Sustainable Cities and Communities, Partnerships to achieve the GoalRequired Resources

Item Name	Type	No. of Units	Per Unit Cost (in Rs)	Total (in Rs)
			Total in (Rs)	80000
Raspberry pi	Equipment	2	14000	28000
Motors	Equipment	2	1000	2000
Temp & humidity Sensor	Equipment	4	150	600
Servo Motor	Equipment	4	700	2800
camera	Equipment	1	6000	6000
Battaries	Equipment	2	2500	5000
ultrasonic sensor	Equipment	2	100	200
traveling/presentation/publication	Miscellaneous	4	500	2000
printing	Miscellaneous	4	1500	6000
4	Miscellaneous	4	500	2000
Pixhawk Flight Controller	Equipment	2	9600	19200
BLDC 1000KV A2212	Equipment	4	750	3000
BLDC RACE SPEC 2300KV	Equipment	2	1600	3200