ESP-NOW Swarm Search for Lost RFID Objects

ISEF Category: Embedded Systems

Subcategory: Other  ·  Difficulty: Advanced  ·  Setup: University Lab  ·  Time: Full Year

The Hook

A single robot can wander a room for a long time. A swarm can split the work and home in faster. That is the big idea behind distributed search. You can test whether more microbots really find a lost RFID source faster, or whether coordination overhead slows them down.

What Is It?

This project studies how a group of small wireless robots can work together to find a signal source. ESP-NOW is a short-range communication method that lets ESP32 boards send messages directly to each other without a Wi-Fi router. RFID tags emit or reflect a signal that a reader can detect, so the swarm can treat signal strength as a clue about where the object is hiding.

Think of it like a team of people searching a dark room with flashlights. Each robot only sees part of the room, but the group shares updates and keeps moving toward stronger signal readings. A gradient-descent search means each robot tries to move in the direction that improves the signal estimate, step by step, until the swarm converges near the target.

Why This Is a Good Topic

This is a strong science fair topic because you can test a real system with clear numbers. You can measure search time, path length, localization error, and how those change as swarm size grows. The problem connects to warehouse tracking, search-and-rescue, and asset recovery, so the work feels practical. You also get room to build skills in embedded systems, simulation, communication protocols, and data analysis.

Research Questions

• How does swarm size affect the time required to localize an RFID-emitting object? ?
• What is the effect of communication loss on convergence speed in a distributed ESP-NOW swarm? ?
• Does adding noisy signal measurements reduce localization accuracy more in small swarms or large swarms? ?
• To what extent does the choice of movement rule change the final distance from the target? ?
• Which swarm size gives the best tradeoff between search time and communication overhead? ?
• How does obstacle density affect convergence in simulation compared with hardware tests? ?

Basic Materials

• ESP32 development boards, one per robot or agent.
• RFID reader module compatible with your chosen tag type.
• RFID tags or a test object that produces a detectable signal.
• Small wheeled robot chassis or microbot platform.
• Motor driver boards matched to your motors.
• Rechargeable batteries or USB power banks.
• Laptop for code upload and data logging.
• Measuring tape or grid mat for mapping the search area.
• Cardboard, tape, or foam board for simple obstacles.
• Digital stopwatch or phone timer.
• Notebook or spreadsheet for recording trials.

Advanced Materials

• ESP32 development boards with external antennas for longer-range testing.
• Multiple RFID readers or a scanner array for better signal sampling.
• Custom PCB or perfboard for compact robot builds.
• Wheel encoders or an inertial measurement unit for motion tracking.
• Motion capture system or overhead camera for ground-truth position data.
• RF shielding materials or controlled test enclosure.
• Logic analyzer or serial monitor for communication debugging.
• 3D-printed chassis parts for repeatable robot geometry.
• Calibration targets for signal mapping across the test area.
• Breadboard, jumper wires, and soldering tools for prototype integration.

Software & Tools

• Arduino IDE: Uploads firmware to ESP32 boards and helps you debug serial output.
• Python: Cleans trial data, computes convergence metrics, and makes plots.
• MATLAB: Models swarm behavior and compares different search rules.
• ImageJ: Measures robot paths from overhead video when you need visual ground truth.
• Proteus: Simulates embedded circuits if you need a quick hardware sanity check.

Experiment Steps

1. Define the search metric you care about, such as time to converge, final localization error, or total path length.
2. Choose one movement rule and one communication rule, then write them down before you code anything.
3. Build a simulation first, so you can test swarm size, signal noise, and obstacle layout without breaking hardware.
4. Plan a hardware test space with fixed landmarks, repeatable target placement, and a clear ground-truth measurement method.
5. Decide which variables stay constant, including robot speed, battery state, and message frequency.
6. Set up a data table that records every trial in the same way, so you can compare simulation and hardware results directly.

Common Pitfalls

• Using raw signal strength as if it were exact distance, which makes the swarm chase noisy readings instead of the target.
• Letting each robot update at a different rate, which creates unfair comparisons between swarm sizes.
• Testing in a changing room layout, which alters reflections and ruins repeatability.
• Ignoring packet loss in ESP-NOW messages, which hides the real cost of coordination.
• Comparing hardware trials to simulation results without matching the same obstacle map and starting positions.

What Makes This Competitive

A stronger project does more than build a working swarm. It compares at least two search rules, includes a real noise model, and explains why one design wins under certain conditions. You can also raise the level by validating simulation against hardware and using statistics to show when added robots stop helping. A well-chosen failure case, like crowded spaces or message loss, can make the analysis much more impressive.

Project Variations

• Test the same swarm search idea with Bluetooth Low Energy instead of ESP-NOW to compare communication overhead.
• Swap the RFID source for a Wi-Fi signal source or BLE beacon and study whether the search rule still converges.
• Keep the hardware fixed and focus on a pure simulation study of obstacle density, noise, and swarm size.

Learn More

• Arduino ESP32 Documentation: Search the Arduino docs for ESP32 board support and serial debugging basics.
• Espressif ESP-NOW Documentation: Read the official Espressif docs for peer-to-peer messaging on ESP32 devices.
• NIH PubMed: Search for review articles on swarm robotics, distributed search, and signal-based localization.
• NASA Open Science Data Repository: Look for examples of navigation, sensing, and autonomous search data methods.
• MIT OpenCourseWare, Mobile Robot Algorithms: Search MIT OpenCourseWare for robot localization and multi-robot coordination lectures.