Once a search-and-rescue mission begins, time becomes the most valuable resource. The operation site may be located in mountainous forests, earthquake-affected areas, flood zones, or remote locations where people have gone missing. Complex terrain, building debris, smoke, standing water, changing lighting conditions, and limited communications can all make manual search operations significantly more difficult.

Drones can enter areas that ground personnel cannot reach quickly, expanding the search area from the air. In GPS-denied or weak-signal environments, the drone must also rely on onboard sensors to continuously estimate its position, detect obstacles, autonomously search a designated area within a limited time, detect and locate missing persons, safely approach the target, and accurately deliver simulated rescue supplies.

This complete mission chain forms the recommended challenge topic for this edition of the FlyCore Flying Robot Challenge: “Emergency Search & Rescue and Supply Delivery.” The challenge focuses on autonomous operations by a single drone in complex environments and evaluates the team’s system integration capabilities across perception, localization, planning, and control. Participating teams can develop and validate their projects based on the FlyCore Integrated Control System, bring their completed solutions to the challenge, and compete for an innovation fund worth one million.

01

Challenge Topic Introduction

Emergency Search & Rescue and Supply Delivery

The challenge requires a single drone to autonomously search a designated area, detect and locate missing persons, and then use real-time environmental information to plan a safe route, approach the target steadily, and complete a simulated rescue supply drop. Localization, perception, planning, and control must operate collaboratively on the same aerial platform.

Every step in the mission affects the next. Localization drift may cause deviations in the search route. Unstable environmental perception can increase the risk during target approach, while errors in target position estimation can further affect delivery accuracy. The final deliverable should therefore be a complete system that can operate repeatedly, record the mission process, and remain stable under changing environmental conditions.

Within the defined mission scope, participating teams should focus on the following capabilities:

• Single-drone autonomous operation: Minimize human intervention during the mission and enable the drone to independently perform the main decision-making and control tasks.
• Area coverage search: Scan the designated area according to a planned route while balancing search efficiency and coverage completeness.
• Target detection and localization: After detecting a missing person, provide stable target position information that can be directly used for navigation.
• Safe target approach: Adjust the flight path according to obstacles and terrain changes while maintaining sufficient safety clearance for supply delivery.
• Simulated supply delivery: Control the flight state and release timing so that the supplies land within an effective area near the target.

Challenge Objectives

Participating teams can carry out system design, algorithm development, and scenario testing around the following five areas. Each capability should first be verified independently before full-system integration and testing.

Autonomous Localization

When GPS is unavailable or unstable, the drone must still continuously estimate its own position and attitude, providing a reliable reference for autonomous flight, search coverage, and subsequent target approach. Teams should pay particular attention to localization initialization time, accumulated drift, and estimation stability during rapid motion.

Technical Directions:

Visual Localization | Inertial Navigation | Multi-Sensor Fusion

Environmental Perception

The drone must use sensors to acquire information about its surroundings, understand the spatial structure, detect building debris, trees, slopes, or temporary obstacles, and provide reliable information for obstacle avoidance and safe navigation. Perception results must also be delivered to the planning module in a timely manner to avoid situations where obstacles are detected too late for effective avoidance.

Technical Directions:

Environment Modeling | Obstacle Detection | Spatial Perception

Target Perception

During the search process, the drone must detect missing persons, further estimate their positions and assess their status, and convert targets observed in the camera image into position data that can be used for planning and approach. Complex backgrounds, shadows from trees, changes in body posture, and partial occlusion can all increase the difficulty of target detection.

Technical Directions:

Target Detection | Target Localization | Status Assessment

Path Planning

The drone must follow planned routes to cover the designated search area. After a target is detected, it must also combine environmental and obstacle information to re-plan a safe and efficient approach route. The planning module should minimize search blind spots while also controlling the number of turns, total flight distance, and energy consumption.

During testing, teams can compare coverage rate, search duration, and remaining battery level under different route spacings and flight altitudes to identify parameters that best match the drone’s payload configuration and sensor field of view.

Technical Directions:

Search Planning | Path Optimization | Autonomous Decision-Making

Supply Delivery

After approaching the missing person, the drone must maintain stable flight, continuously track the target, and accurately release simulated rescue supplies at an appropriate position and time. Flight altitude, speed, wind disturbance, and payload variation can all affect the landing point.

It is recommended to begin with fixed-point delivery at low speed and low altitude, then gradually introduce crosswinds, position errors, and payload variations while recording delivery errors.

Technical Directions:

Target Tracking | Motion Control | Precision Delivery

Challenge Application Scenarios

• Post-Disaster Search and Rescue: Search for trapped individuals in environments affected by earthquakes, floods, landslides, and other disasters, while providing timely supply support.
• Mountain and Forest Rescue: Assist rescue personnel in locating missing individuals and improve search and response efficiency in complex terrain.
• Remote Missing-Person Search: Enable unmanned systems to search for and locate targets in areas that are difficult for personnel to cover quickly.
• Emergency Supply Delivery: Deliver essential supplies near the target when roads are blocked or ground rescue teams have not yet arrived.

02

Flying Robot Innovation & Creation Challenge

The development of flying robots requires validation in more real-world scenarios. An innovative idea can only become a truly valuable product after it has undergone development, testing, and practical application validation.

To enable more developers to participate in this process, the 2026–2027 Flying Robot Innovation & Creation Challenge is now officially open.

The competition is open to universities, research institutes, technology companies, developer teams, and flying robot enthusiasts, creating an innovation platform that integrates technical learning, project development, scenario validation, and achievement demonstration.

Participating teams can develop their projects based on the FlyCore Integrated Flying Robot Control System, progressing from solution design and system integration to scenario validation and project demonstration, ultimately creating their own innovative flying robot solutions.

At the same time, flying robot development boot camps will be organized regularly, covering system integration, technology development, and project practice to provide participating teams with opportunities for learning, technical exchange, and project development support.

Continuous Release of Innovation Topics to Provide Directions for Exploration

To help participating teams quickly move into hands-on development, the Challenge will continuously release a series of innovation topics.

Each topic will combine key flying robot technologies with real-world application scenarios to provide project references for developers.

Developers may conduct research based on the recommended topics or extend them with their own ideas and innovations.

From a technical concept to a flying robot that can actually complete a mission—this is exactly the innovation process that the Challenge aims to promote.

03

Build with FlyCore and Compete for an Innovation Fund Worth One Million

The FlyCore Integrated Flying Robot Control System provides developers with an integrated development foundation covering flight control, localization and navigation, and mission development.

FlyCore integrates core capabilities including flight control, mapping and localization, perception and planning, and sensor integration, creating a complete closed loop of perception, localization, planning, and control. It enables developers to rapidly build flying robot platforms and carry out algorithm validation and application exploration.

Based on FlyCore, participating teams can quickly complete solution design, functional development, and scenario validation, transforming innovative ideas into real robotic systems that can fly and accomplish real missions.

With FlyCore, teams can:

✓ Rapidly build flying robot platforms
✓ Conduct algorithm validation and functional development
✓ Complete real-world scenario testing
✓ Participate in Challenge project demonstrations

Turn ideas into flight. Bring innovation into real-world applications.