Key Competition Challenges

Task:

Create a comprehensive AI system that builds a soil map based on express analysis, determines the need and composition of mineral fertilizer, prepares it, and sprays it on the required areas.

Objective:

To automate the soil mineralization process, reduce fertilizer consumption, and improve the environmental safety of products.

What is required:

1. An express soil analysis system with map generation.
2. A CV model to build a map of plant presence.
3. A data processing system for decision-making.
4. An automatic fertilizer preparation system.
5. An automatic drone refueling system.
6. An automatic field treatment system using drones (with auto-swap/charging of batteries).
7. An algorithm for spot treatment.
8. Algorithms for starting and stopping the system.
9. A short video presentation or a solution presentation.

System Description:

The system collects data from sensors to build a map of mineralization and plant presence, then decides on the need for fertilization. A base station (tanks, drones) prepares the solution, fills the drone, and sends it for spraying. After treatment, a re-analysis is performed, and the system self-learns to improve accuracy. An agronomist sets initial parameters and can stop the system at any time.

Task:

Create a comprehensive AI system that builds an irrigation map and determines the need for watering based on data from moisture sensors, photo/video sensors, and weather forecasts.

Objective:

To automate the irrigation process and reduce water consumption.

What is required:

1. An express soil analysis system with irrigation map generation.
2. A CV model to build a map of plant presence.
3. A data processing system to decide on the necessity and timing of irrigation.
4. An automatic irrigation control system.
5. A short video presentation or a solution presentation.

System Description:

The system collects data from sensors and meteorological services, builds a soil map, and determines the necessity and degree of irrigation. It commands the sprinkler machine and controls its speed. After irrigation, the system analyzes the results and self-learns to optimize future cycles.

Project Goal:

To develop software for an automated laser weed removal system integrated with agricultural machinery. The system must ensure high-precision recognition (> 90%), safe targeting, and integration with on-board electronics.

Functional Requirements:

• CV Module: "Weed vs. crop" recognition, weed localization.
• Laser Targeting Module: Calculation of radiation parameters.
• Safety Module: Prevention of crop damage, laser lock at high speed.
• Operator Interface: System status display, mode control, activity log.
• Reporting: Database of treated areas, report export, visualization.

Technical Parameters:

• Cameras: ≥ 2 units, Full HD, ≥ 30 fps, IP65 protection.
• Laser: Blue diode (≈ 450 nm), 80-320 W power.
• Positioning: GPS/GLONASS, possibly LiDAR.
• Interfaces: CAN bus, Ethernet, WiFi/4G/5G.
• Operating Conditions: -10°C to +40°C, humidity up to 90%.

Project Goal:

To develop an intelligent system to optimize the irrigation of cotton fields using AI. The system should ensure rational use of water resources, increase yields, and reduce costs.

Project Tasks:

Collect and process data from agro-sensors, develop a machine learning model to predict moisture needs, create an optimal irrigation schedule algorithm, and develop a web dashboard or mobile app for visualization and control.

Expected Results:

A working prototype of the system, an irrigation prediction algorithm with at least 85% accuracy, a web interface for agronomists with interactive maps and graphs, and a report on testing and implementation recommendations.