Board v1

Swarm Robot Platform

A low-cost, scalable swarm robotic platform designed for research and educational purposes, capable of infrared (IR)-based communication and basic cooperative behaviors.

Playlist of videos showcasing robots and algorithms: https://www.youtube.com/playlist?list=PLNTM9zCJujTltedypfpow2rPiPkVAqfGI

---

🧠 About the Project

This project focuses on the design of a cooperative robotic platform built around the ESP32 microcontroller. The primary goal is to enable simple, local communication and decentralized coordination among identical robots, forming a swarm.

Key design constraints:

• Minimal production cost (~24 EUR/robot, parts only)
• Compact and robust mechanical design
• Infrared-based full-duplex communication (360°)
• Modular and scalable C firmware using ESP-IDF
• Developed as part of a Bachelor’s Thesis at Brno University of Technology

Thesis Title:
Návrh a realizace kooperativní robotické platformy s jednoduchou komunikací
(Design and Implementation of a Cooperative Robotic Platform with Simple Communication)

---

⚙️ Features

🔁 360° IR Communication

6 pairs of IR LEDs and phototransistors evenly placed around the robot enable full directional communication.

🚧 Obstacle Detection

2 front-facing IR LED + phototransistor pairs detect obstacles.

🤖 Cooperative Algorithms

With Leader

• Simulates a robot "calling" others to complete a task.

• Modified CSMA/CA protocol ensures robots check for ongoing transmissions before broadcasting.

• Robots walk randomly looking for signals:

  • If signal is found in a randomly chosen time window → the robot becomes a follower.
  • If not → the robot becomes a leader.

• Leader can be pre-set (WITH_LEADER 1)

• Flowchart of system with leader:

• Implemented algorithms:

  • Follow
    Followers track the leader, who moves randomly.
    

  • Spread Out
    Leader stays in place and acts as a beacon. Followers spread and return.
    

Without Leader

• No signal detection or leader election.

• Algorithms run immediately after power-on.

• Implemented algorithms:

  • Follow Chain
    Robots follow each other in a line. First robot walks randomly. Predefined or dynamic ID (order) assignment.
    

  • Chain
    Robots stay still in a line. The last robot moves forward to the front.
    

🧠 FSM-Controlled Behavior

Each robot uses a finite state machine (FSM) for autonomous control and decision-making. The diagram shows the flow of states of algorithms with leader.

🔧 Firmware Modules

• io_define – I/O pin mappings, robot ID
• servo_driver – movement (servo) control
• led_driver – IR LED driver
• hwtimer – Hardware timer driver
• adc_lib – ADC driver
• dm_comm – IR communication (Manchester encoding, CSMA/CA)
• coop – Cooperative algorithm logic
• state_machine – FSM and behavior transitions

---

🛠️ Setting Up

Requirements

• ESP-IDF
• PlatformIO (recommended; VS Code plugin)
  Target board: ESP32DOIT-DEVKIT-V1
• USB-to-UART bridge (e.g. CP2102)

🔧 Calibration

• Movement Calibration
  The robot doesn’t use encoders, so the motors may spin at different speeds, causing drift.
  Adjust SERVO_FORWARD_LEFT_MOD and SERVO_BACKWARDS_RIGHT_MOD in io_define.h to straighten movement.

• IR Signal Thresholds
  Tune SIG_THRESHOLD in dm_comm.h and DIS_THRESHOLD in coop.h to adjust signal sensitivity.

• Algorithm Selection
  Some algorithms (like Chain or Follow Chain) should start immediately on power-up.
  Define the startup behavior by setting flags to "1" (CHAIN or FOLLOW_CHAIN).

⚡ Flashing Instructions

1. Connect the robot to your computer via USB.
2. Open io_define.h and set:
   • ROBOT_ID – unique number per robot
   • WITH_LEADER – 1 (for pre-set leader - ID1) or 0
   • CHAIN or FOLLOW_CHAIN – chosen algorithm
3. Build and flash the firmware