NATURE RELATED SCIENCE PROJECT EMPOWERED BY AI
Goals: Crea ng a science project related to nature and enhanced by AI. AI or
ar ficial intelligence has mordernized many fields of worklife and has eased
workflow for many sector such as deep research, office work, spread sheet
management etc. This project ventures through a mostly unpopular field of AI
helping research or other works related to Nature.
To keep in mind:1. Project should solve or help solve a real life problem.
2. Method of research or work should be enhanced by AI.
3. Components or the way of func oning should not
effect local biodiversity or plants.
Project-01: Smart Agri-Watch Drone
Capabilities: 1. Cropdusting
2. Detecting plants with diseases by VISUAL RECOGNATION.
3.Detecting invasive species.
Components: Drone(Pre-made)
Raspberry Pi Camera module
Single Board Computer: Raspberry Pi 4 (4gb)
2200-5200mAh battery
Wireless data transmitter-receiver
12V mini water pump
500-1000mL bottle
servo-triggered valve or relay switch
AI Components: Language: Python Or C++.
Autonomous workflow: Mission Planner or QGroundControl
Training dataset: PlantVillage, Kaggle plant leaves
Training: Google Collab or Teachable Machine
Model deployment: Tensor flow or Pytorch.
How It Works: Autonomous workflow makes watering, spraying pesticides etc done
in a more efficient way. Visual detection detection helps detect plant diseases while
doing crop dusting by constantly scanning the plant for visual symptoms of diseases,
also the visual detection system helps detect and count Invasive Species such as water
hyacinth, parthenium etc. AI helps with visual recognition, points out plants with
diseases such as Ring spot, Necrosis etc while also recommending steps to prevent
the diseases from spreading. Also The visual detection of invasive species also helps
farmers take action to prevent a disaster from happening.
Estimated cost: 22-25,000 BDT. Ditching the Crop dusting capability can reduce
the cost.
Project-02: Smart Plant Monitoring System(SPMS)
Capabilities: 1. Predicts plant diseases
2. Points out watering needs.
Components: Soul moisture sensor
DHT11 or DHT12 (temp and humidity sensor)
LDR
ESP8266 or HC-05 (for wifi/bluetooth)
Arduino UNO R3 board (motor/sensor functions)
How It Works: Sensors picks up activity of soil like pH, percentage of water etc and
relay it to the Adruino UNO R3 board which collects and relays the data to Mobile
phone or other external systems where Pre-existing software or AI predicts plant
diseases and watering needs for the plant. Many open source resources are also
available to develop independent and new external AI system.
Estimated cost: 4-6,000 BDT.
Caution: Unsure of fulfilling project goals as the main hardware system lacks actual
AI capabilities, external device (If made by group) can drive up costs dractically.
Project-03: Smart Native GreenCare System
Capabilities: 1.Monitors plant moisture and waters when needed.
2. Alerts when watering is necessary.
3.Recommends native plants for the location/garden type.
Components: Arduino UNO R3
LDR
Soil Moisture Sensor
Relay Module
5V Mini Water Pump
Jumper Wires + Breadboard
Plastic tubing
Water container
LCD or OLED Display
ESP8266
How it works: Soil Moisture Sensor reads analog moisture value, Arduino checks if
it’s below a threshold (Example: Reading<300 = dry) If dry, triggers relay →
activates pump → waters plant Displays or logs a message: “Watering Plant…”. If
moisture is fine, message = “No watering needed”. Furthermore uses the data from
soil mosture sensors and LDR to recommend plants that can grow better In current
environment.
Estimated cost: 6-8,000 BDT
Caution: Basic AI functions, Native plant recommendations can sometimes be wrong
due to sudden weather changes and lack of integrated or externally inputed GPS data.
Project-04: Water life monitoring system
Capabilities: 1. Monitors water quality
2. Tracks fish movement
Components: Arduino UNO or Nano
DS18B20 Waterproof Temp Sensor
pH Sensor Kit (with probe)
Turbidity Sensor
Ultrasonic Sensor (HC-SR04)
IR Break Beam Sensors (x2)
LCD (16x2) + I2C Module
SD Card Module + Card
Jumper Wires + Breadboard
9V Battery/5V USB
How it wors: Sensors pick up water purity and temprature to predict fish movement,
ultrasonic sensor help count fish population and helps study biodiversity.
Estimated cost: 6-8,000 BDT
Caution: Needs enclosed environment for concept proofing, real- world scalling can
be
more costly, lacks the feel of AI and More like autonomous computing.
Conclusion
Pros; Each Project solves or helps solve a real life problems.
Project-01: Helps farmers to farm more efficiently and prevents plant diseases and
attacks from invasive species.
Project-02: Helps with gardening.
Project-03: Also helps with gardening but with more autonomy
Project-04: Helps study sea-life
Cons: While fulfilling their role each have their own problems
Project-01: Costly
Project-02: Hardware system lacks actual AI capability.
Project-03: plant recommendations can sometimes be off.
Project-04: Does more autonomous computing than AI workflow.
Cost overrun potential: Cost may over-run due to Lack of affordable or
designated components and work essentials such as SBC, Hot Glue kit, Soldering
Iron, Sensors etc.
