Farm management system.
NOTE: This is an aspirational side project, migration onto github is in progress.
Feed 10 billion people.
- Expensive proprietary hardware and software in machines.
- Complicated unintuitive interfaces.
- Seeding, spraying, harvesting inefficiencies.
- Aging farming expertise.
- Fully autonomous tractors are kinda expensive.
- Some farms here in Australia are bigger than entire countries, mate.
See TODO for specific project goals
- Plug and play displays
- Simple clean interface
- Free software on cheap hardware
- On or offline mapping
- Basic positioning
- Heatmap of spray area
- Peer-to-peer communication
- High precision positioning
- Mapping activities, seeding, harvesting, spraying.
- Machine messaging
- Timing of tasks
- Machine interfacing
- Vision systems
- Navigation
- Machine health and diagnostics
- Self driving autonomous machines
- Individual plant health recognition vision system
- Machine learning optimization of routes
- End-to-end crop cycle management
- High precision farming at the smallest possible cost.
- Intuitive and simple as possible.
- Versatile, work on anything with a screen.
- Primarily developed for RasberyPi.
- Ad-hoc peer-to-peer communication.
- Primarily written in python
- WebUI interface
- Initially using django for web framework
- Seperate backend and frontend
- Backend processes and logs data to sqlite
- Serve live position and other real-time data to simple .json http GET
- Django frontend for local and remote WebUI
- Leaflet mapping with custom tiles and layers.
- Primary buttons and info overlayed ontop of map
- Boot to kiosk mode sort of thing.
- Consideration for multiple GPS receivers, on spray booms and whatever.
- Plug and play add on hardware such as additional sensors / IO / stereo output / additional screens / whatever