Page 255 - AI for Good Innovate for Impact

AI for Good Innovate for Impact

Traditional SAR relies heavily on visual and audio clues, which can delay rescues. AI models
analyzing RF signal patterns will significantly boost response accuracy and timing.

The machine learning models will use environmental and signal data to predict the missing
persons' location, improving survival chances and operational efficiency. Change 4.2-Climate

Use Case Status: Planning Phase

2�2 Benefits of use case

Sustainable Cities and Communities: Enhancing emergency response infrastructure.
Good Health and Well-Being: Facilitating timely rescues and health outcomes during
emergencies.

• Project Partners Mbeya University of Science and Technology (MUST): Research and
development of AI localization technologies.
• Telecom Company: Provision of RF data and communication infrastructure support.
• Rescue Team: Field testing, feedback, and operational deployment of the solution.

2�3 Future Work

• Define system specifications.
• Collaborate for data acquisition with SAR teams.
• Develop AI models and validate through field tests.
• Expand system for UAV-based aerial searches.

3 Use Case Requirements

• REQ-01: Reliable RF signal triangulation across terrains.
• REQ-02: Real-time predictions with minimal computational resources.
• REQ-03: Portability and SAR kit compatibility.
• REQ-04: Aerial and ground operational integration.

4 Sequence Diagram

Actors:
- Lost Individual (RF Signal Source)
- SAR Teams (Ground and UAV)
- AI RF Locator System
- Command Center

Process Flow:
1. RF signal emitted by lost individual.
2. Signal detection by sensors or UAVs.
3. Signal data sent to AI RF Locator.
4. AI processes data and predicts location.
5. Location shared with Command Center.
6. SAR Teams dispatched based on AI guidance.
7. Search conducted and status updates sent.

219