Low-Cost Cotton Picking Robot
Budget / Salary₹1,500–12,500
TypeFreelance project
LocationRemote
Posted1 hour ago
I need a complete, low-cost prototype of an AI-powered cotton-harvesting robot for a final-year project demo. The unit must detect ripe bolls with YOLOv8 running in Python/OpenCV, plot a path toward each target, and trigger a vacuum-based suction module that gently removes and stores the cotton. An ESP32-CAM provides onboard vision; motor drivers and sensors handle locomotion and arm movement. FastAPI can expose status and control endpoints so the demo laptop or phone can monitor performance in real time.
The showcase hinges on two points: 1) the system’s observable functionality in a small, controlled test bed and 2) its ability to keep working reliably under repeat runs—our “working condition” highlight. While the final deployment might be in open fields or greenhouses, this build only has to prove itself on a table-top row of cotton stems.
What I’m looking for in your proposal
• A concise architecture explaining how you will fuse YOLOv8 inference, navigation logic, and hardware control on the ESP32 (or companion MCU/Raspberry Pi, if you prefer).
• A bill of materials that meets the low-cost constraint without sacrificing core performance.
• Milestone schedule from initial design through live demo day, including hardware assembly, firmware, computer-vision tuning, motion testing, and demo rehearsal.
• Evidence of prior work with embedded vision, motor control, and hardware-software integration—links to videos or repos are ideal.
Deliverables
• Complete annotated source code (Python, C/C++ for ESP32).
• Wiring diagram, PCB or perf-board layout, and 3-D/laser-cut files if any custom parts are needed.
• Calibration files and a brief guide on retraining YOLOv8 for new cotton images.
• A short video demonstrating autonomous detection, approach, and pick sequence in the demo rig.
Acceptance criteria
The robot must autonomously identify at least three cotton bolls in a row and pick them within one minute, without manual correction, during the final demonstration.
Please include your estimated timeline and total cost in your reply so I can lock in the project schedule.
The showcase hinges on two points: 1) the system’s observable functionality in a small, controlled test bed and 2) its ability to keep working reliably under repeat runs—our “working condition” highlight. While the final deployment might be in open fields or greenhouses, this build only has to prove itself on a table-top row of cotton stems.
What I’m looking for in your proposal
• A concise architecture explaining how you will fuse YOLOv8 inference, navigation logic, and hardware control on the ESP32 (or companion MCU/Raspberry Pi, if you prefer).
• A bill of materials that meets the low-cost constraint without sacrificing core performance.
• Milestone schedule from initial design through live demo day, including hardware assembly, firmware, computer-vision tuning, motion testing, and demo rehearsal.
• Evidence of prior work with embedded vision, motor control, and hardware-software integration—links to videos or repos are ideal.
Deliverables
• Complete annotated source code (Python, C/C++ for ESP32).
• Wiring diagram, PCB or perf-board layout, and 3-D/laser-cut files if any custom parts are needed.
• Calibration files and a brief guide on retraining YOLOv8 for new cotton images.
• A short video demonstrating autonomous detection, approach, and pick sequence in the demo rig.
Acceptance criteria
The robot must autonomously identify at least three cotton bolls in a row and pick them within one minute, without manual correction, during the final demonstration.
Please include your estimated timeline and total cost in your reply so I can lock in the project schedule.
Apply on Freelancer →
Project sourced from Freelancer.com. Applications happen directly on the original platform — we never collect your data.