Engineering an Autonomous Rescue Drone
Building an Autonomous Rescue Drone for ARC 2026: Engineering, Robotics, and Real World Problem Solving
By Jaiden Chao
Introduction
I joined Team Skyward to compete in the Aerospace Robotics Competition (ARC) 2026 in New Hampshire. Our challenge was to design, build, and test an autonomous rescue drone capable of completing realistic search and rescue missions. What started as a robotics project quickly became one of the most challenging and rewarding engineering experiences I have had.
Working alongside my teammates, I gained hands on experience with robotics, computer vision, embedded systems, and autonomous flight. More importantly, I learned how engineers solve problems through teamwork, testing, and continuous improvement. The experience strengthened my interest in robotics and aerospace engineering while giving me a deeper understanding of how complex systems are developed and refined.
What is the Aerospace Robotics Competition?
The Aerospace Robotics Competition (ARC) is a multidisciplinary engineering competition that challenges student teams to design unmanned aerial systems capable of completing realistic rescue and environmental missions.
The 2026 competition required teams to complete two primary objectives:
Mission 1: Rescue Assistance
The drone was tasked with locating a simulated swimmer in distress and delivering a lifesaving flotation device.
To accomplish this, our system combined autonomous and manual flight operations. A pilot manually navigated the drone toward a designated search area, while onboard computer vision software analyzed camera footage to detect a red rescue target. Once detected, the drone could automatically hold position, allowing the pilot to safely deploy the rescue mechanism.
Mission 2: Environmental Rescue and Object Deployment
The second mission involved multiple tasks:
• Deploying whale tracking tags onto designated targets
• Picking up and transporting turtle rescue objects
• Releasing payloads at specified locations
• Navigating efficiently under strict time constraints
These objectives simulated real world conservation and rescue operations that autonomous aerial systems could support in the future.
Joining Team Skyward
When I joined Team Skyward, I was immediately exposed to the realities of engineering beyond the classroom.
Unlike textbook problems with known answers, every design decision involved tradeoffs. Components failed unexpectedly. Mechanical designs that looked promising on paper did not always perform reliably in testing. Software had to work seamlessly with hardware, and small errors could impact the entire system.
Throughout the season, our team met regularly to plan, build, troubleshoot, and improve our drone. As competition day approached, our focus shifted toward system integration and flight testing.
One of the most valuable lessons I learned was that successful engineering is rarely about finding a perfect solution immediately. Instead, it is about continuously improving a design through experimentation and iteration.
Engineering the Drone System
The final drone represented the integration of multiple engineering disciplines working together as one autonomous system.
Flight Control Architecture
The drone utilized a flight controller integrated with GPS navigation, telemetry communication systems, electronic speed controllers, and onboard computing hardware.
At the center of the autonomous system was a Raspberry Pi 5, which served as the onboard computer responsible for processing visual information and communicating with the flight controller.
This architecture allowed us to combine autonomous decision making with human pilot oversight, creating a safer and more reliable system.
Autonomous Flight Operations
One of the most interesting aspects of the project was implementing autonomous behaviors within the flight system.
Rather than relying entirely on manual control, our drone was capable of performing specific actions automatically after receiving visual input from onboard cameras.
For example:
• Detecting rescue targets
• Holding position when a target was found
• Assisting with mission execution
• Supporting navigation workflows
Developing and testing these autonomous functions introduced me to the practical challenges of robotics and intelligent systems.
Computer Vision and Automation
A major component of the project involved computer vision.
Using Python and OpenCV, our system analyzed live camera footage from an ArduCam camera mounted on the drone.
The software was designed to identify red colored rescue targets within the environment. When the target was detected, the Raspberry Pi communicated with the flight controller using the Pymavlink protocol, triggering an automated response.
Although this was not machine learning in the traditional sense, it represented a real world application of computer vision, image processing, and automation.
Mechanical Design and Prototyping
The drone also required a custom mechanism capable of interacting with mission objects.
Our team explored multiple prototype designs before arriving at a reliable solution. Early concepts included fully enclosed gripping systems and several 3D printed mechanisms. However, these designs introduced challenges related to servo integration, structural reliability, and object retention.
Through repeated testing and redesign, the team eventually developed a functional mechanism that could consistently pick up and release objects during competition missions.
Many of our most important improvements came directly from observing failures during testing.
Systems Integration Challenges
One of the most important engineering lessons from ARC was understanding systems integration.
A successful autonomous drone requires more than individually functional components.
The flight controller, GPS navigation, telemetry communications, Raspberry Pi, camera systems, servo mechanisms, power systems, and flight software all had to work together reliably. A failure in any one subsystem could impact overall mission success.
During testing, our team encountered issues ranging from electronics failures to software communication challenges. Troubleshooting these problems required patience, collaboration, and systematic analysis.
Learning how to diagnose and resolve issues in a complex system was one of the most valuable experiences of the entire project.
Competition Experience in New Hampshire
Traveling to New Hampshire to compete in ARC 2026 was the culmination of our work. The video below captures competition day, when our team put our drone and engineering solutions to the test under real competition conditions.
Competition Day: Putting Months of Engineering to the Test
Standing alongside teams from different regions and watching our system perform under real competition conditions was both exciting and rewarding.
The competition environment highlighted the importance of reliability, preparation, and teamwork. Engineering solutions that work in a laboratory or workshop must also perform consistently in the field.
Seeing our drone successfully complete mission objectives demonstrated how much progress our team had made throughout the season.
More importantly, it showed me how engineering concepts can be transformed into systems capable of solving practical problems.
Technical Skills Developed
Through ARC 2026, I gained hands-on experience in:
Robotics & Aerospace Systems
- Autonomous drone systems
- Flight operations
- Mission planning
- Systems engineering
Software Engineering
- Python programming
- OpenCV computer vision
- Pymavlink communications
- Autonomous flight software
Embedded Systems
- Raspberry Pi integration
- Sensor integration
- Camera systems
- Hardware-software communication
Engineering Design
- Mechanical prototyping
- Design iteration
- Testing and validation
- Failure analysis
Professional Skills
- Team collaboration
- Technical communication
- Project management
- Problem solving under constraints
Reflection
The Aerospace Robotics Competition was far more than a robotics competition. It was an opportunity to experience the complete engineering lifecycle, from initial concepts and prototypes to autonomous systems operating in a real-world environment.
The project strengthened my interests in robotics, aerospace engineering, automation, and intelligent systems. It also taught me that innovation often comes from persistence, teamwork, and a willingness to learn from failure.
As I continue pursuing engineering opportunities, the lessons I learned from Team Skyward and ARC 2026 will continue to influence how I approach technical challenges, leadership, and innovation. Disclaimer: Site built using the RyanCV WordPress theme by Bslthemes and customized for this portfolio. Original theme design credit belongs to the theme authors.