Swarm Robots Form Letters for Art Displays

ISEF Category: Technology Enhances the Arts

Subcategory: Engineering Effects  ·  Difficulty: Advanced  ·  Setup: University Lab  ·  Time: Full Year

The Hook

Six tiny robots can act like a moving pen. When they snap into place, a word can appear on a gym floor in seconds. The hard part is not making them move, it is making people read the result fast and clearly. That turns this project into a mix of robotics, design, and human perception.

What Is It?

This project asks how a small robot swarm can arrange itself into letters that people can actually read. Each robot acts like one movable pixel. An overhead camera tracks where each robot is, then a planner decides which robot should move to which spot so the group forms a clean letter shape.

The Hungarian assignment is a matching method. Think of it like a fair seating chart for robots. It pairs each robot with a target point in a way that lowers total travel. The overhead ArUco tags give the system robot IDs, so the planner knows which robot is which.

You are not just testing whether the robots reach the right spots. You are also testing whether the audience can recognize the letter or word from a phone video. That means you get both engineering data and design data, which makes the project stronger than a simple motion demo.

Why This Is a Good Topic

This is a strong science fair topic because you can measure it in clear ways. You can score formation error, travel distance, completion time, and human readability. That gives you real numbers, not just a cool demo. It also connects to real problems in swarm robotics, public displays, rescue systems, and art installations. You can realistically learn control logic, assignment algorithms, computer vision, and user testing while still keeping the project focused.

Research Questions

• How does the initial robot spacing affect final letter accuracy?
• What is the effect of the number of robots used on audience readability?
• Does a Hungarian-assignment planner produce cleaner formations than a nearest-target planner?
• To what extent does camera angle change the measured formation error from phone video?
• Which letter shapes are easiest for viewers to recognize at the same robot count?
• How does adding a collision-avoidance rule change travel distance and completion time?
• What is the effect of background floor color on human readability scores?

Basic Materials

• 6 small differential-drive robots with programmable control
• overhead camera or webcam mounted above the test area
• ArUco tag stickers or printed fiducial markers
• laptop for video capture and control scripts
• taped floor grid or measured test area on gym flooring
• smartphone for recording observer video
• measuring tape or laser distance measurer
• charger and spare batteries for the robots
• colored floor tape for target letter outlines
• consent form or short survey sheet for human readability ratings.

Advanced Materials

• 6 or more programmable differential-drive robots with wheel encoders
• overhead camera with calibrated field of view
• ArUco marker set with printed scale reference
• laptop or mini PC for real-time tracking and assignment planning
• image-processing pipeline for pose estimation
• video-analysis setup for frame-by-frame error measurement
• external battery monitor for repeatability testing
• marked test mat with known dimensions
• survey software for readability scoring
• data logger for robot paths and timing.

Software & Tools

• ImageJ: Measures formation area, spacing error, and frame-by-frame visual clarity from recorded video.
• Python: Runs assignment logic, tracks robot positions, and computes error metrics.
• OpenCV: Detects ArUco tags and extracts robot positions from overhead video.
• Google Forms: Collects readability ratings from viewers in a simple survey.
• R or Python pandas: Organizes trial data and compares accuracy across test conditions.

Experiment Steps

1. Define the letter shapes you want the swarm to form and decide how you will score accuracy.
2. Choose one planner as your baseline, then set up a fair comparison against at least one alternative assignment method.
3. Build a measurement plan for both robot formation error and human readability from phone video.
4. Set the control variables that must stay fixed, such as floor size, camera position, and starting layout.
5. Plan repeated trials so you can compare performance across different letters, spacing patterns, or swarm sizes.
6. Choose the statistics or scoring rules you will use before you start collecting data.

Common Pitfalls

• Using an uncalibrated overhead camera, which makes robot positions and spacing error measurements unreliable.
• Choosing letters with shapes that are too similar, which makes readability scores hard to interpret.
• Letting robots start from different distances each trial, which mixes assignment quality with travel-length effects.
• Measuring only final pose error, which misses collisions, path crossings, and unstable formations.
• Relying on one viewer group for readability ratings, which gives weak evidence for human-perception claims.

What Makes This Competitive

A stronger project would compare more than one planner, more than one letter shape, and more than one way to score success. You could combine geometric error with human readability and ask whether the same formation that looks best on paper also looks best to people. A competitive version would also control camera calibration, robot starting states, and viewer bias carefully. That gives your results more meaning and makes the project feel like real engineering research.

Project Variations

• Test how the swarm performs with numbers instead of letters, then compare readability against typographic shapes.
• Swap the phone video metric for a timed recognition test to measure how fast viewers identify each formation.
• Compare matte and glossy floor surfaces to see how reflections change tracking accuracy and human readability.

Learn More

• OpenCV documentation: Learn how ArUco marker detection and camera tracking work, then find it in the official OpenCV docs.
• NIH PubMed: Search for review articles on swarm robotics, human factors, and visual perception of patterns.
• NASA Open Data Portal: Explore imaging and tracking datasets that can help you think about calibration and measurement.
• MIT OpenCourseWare: Search for robotics, computer vision, and control courses that explain assignment, tracking, and motion planning.
• IEEE Xplore or ACM Digital Library: Read peer-reviewed papers on swarm formation, robot choreography, and readability studies through your school library access.