5.CONCEPT EVALUATION AND PRODUCT ARHITECTURE - T-div-Engg-Exploration-2026-27/Repo-2 GitHub Wiki
PUGH CHART
5.1 JUSTIFICATION FOR SCORE GIVEN
5.2 SELECTED DESIGN
Concept 1 is selected as final design because DC motor is preferred over servo due to weight of the machine (portablility and light-weight)
Components as per Concept 1 Table (ED2-CD)
| S.No | Part Name from Concept 1 | Function |
|---|---|---|
| 1 | ESP32-CAM + AI CNN (Kaggle model) | Captures soil image, runs Kaggle CNN model to predict fertility (High/Medium/Low) based on soil color & texture |
| 2 | DC motor | Actuates seed dispensing from box container, also used for locomotion |
| 3 | ESP32 | Main brain - receives data from ESP32-CAM, controls DC motor, sends data to Mobile App |
| 4 | Box container | Stores seeds + fertilizer, low-cost plastic box with partition, releases seed when DC motor rotates |
| 5 | Electric Battery | 12V Rechargeable Li-ion battery, powers ESP32, ESP32-CAM, DC motor. Reusable |
| 6 | Mobile App | Blynk/Custom App - displays live soil image, fertility % from CNN, and controls Start/Stop sowing remotely |
Reason for Selection:
- Low Cost: Only Rs. 2000 (ESP32-CAM Rs.500 vs NPK Sensor Rs.3500)
- Automatic: AI CNN fully automatic, no manual chemical mixing
- Reusable Power: Electric Battery rechargeable + solar charge possible
- IoT Enabled: Mobile App for remote monitoring
5.3 Product Architecture
Architecture Type: Modular Architecture
Module Breakdown (Using Only Concept 1 Parts)
Module 1 - Intelligence Module
- Parts: ESP32-CAM + AI CNN (Kaggle model)
- Function: Fertility Detection via Image Processing
Module 2 - Control & Interface Module
- Parts: ESP32 + Mobile App
- Function: Decision making and User Interface
Module 3 - Actuation & Storage Module
- Parts: DC motor + Box container
- Function: Seed Sowing Action
Module 4 - Power Module
- Parts: Electric Battery
- Function: Power Distribution to all modules