Built to order

Robot Swarm using ESP-NOW Communication

A three-robot swarm coordinated over ESP-NOW — Espressif's peer-to-peer wireless protocol that needs no router, no cloud and no pairing. One controller ESP32 broadcasts formation commands; three robot ESP32s acknowledge, report position and hold line, triangle, cluster or scatter formations with a proportional follower controller, while ultrasonic sensors provide local collision override. A command console shows the live arena map, battery and link stats, the ESP-NOW packet log and latency telemetry. Suitable for B.E./B.Tech final-year projects in Mechanical, Mechatronics and Robotics.

Robot Swarm using ESP-NOW Communication — project thumbnail preview
More project photos (2)

The problem

Swarm robotics is usually taught as simulation — dozens of agents on a screen, no radios, no dropped packets, no dead batteries. The moment you put radios on real robots, the interesting problems appear: who hears whom, what happens when an acknowledgement never arrives, how formations survive a robot that stops responding. Wi-Fi and Bluetooth both drag in access points, pairing and latency that make multi-robot coordination painful. ESP-NOW, Espressif's connectionless peer-to-peer protocol, is the purpose-built answer: short packets, millisecond latency, no infrastructure. This project builds a real three-robot swarm on it — one controller, three ESP32 differential-drive robots, formation broadcasts with acknowledgements and retries, 10 Hz position telemetry, and ultrasonic collision override that works even if the radio link dies. The console makes the invisible visible: arena positions, packet log, delivery rate and round-trip latency. Formations are design targets measured on the buyer's own floor — range and loss depend on the room, antennas and interference, and the report says so.

How it works

  1. Each robot ESP32 registers the controller's MAC address as an ESP-NOW peer at boot — no Wi-Fi association is ever needed.
  2. The operator picks a formation on the console; the controller broadcasts a formation packet with each robot's target x/y slot (up to 250 bytes).
  3. Robots acknowledge; the controller retries missing acks up to 3 times and the console logs the ack count per broadcast.
  4. Followers run a proportional controller driving their differential-drive motors toward their assigned slots; the leader follows console joystick commands.
  5. Every robot streams position, heading, battery and RSSI back at 10 Hz; the console renders the arena map and link statistics live.
  6. Ultrasonic sensors run a local loop on each robot: an obstacle inside the stop distance halts the motors regardless of what the radio says.
  7. The telemetry view aggregates delivery rate and round-trip latency; the report's test procedure has the buyer measure range, loss and latency in their own room.

Tech stack:

  • 4x ESP32 DevKit (1 controller + 3 robots), Arduino-core firmware
  • 3x differential-drive robot chassis kits (motors, wheels, caster)
  • ESP-NOW peer-to-peer protocol (connectionless, 250-byte frames)
  • Ultrasonic distance sensors (local collision override per robot)
  • Proportional formation-follower controller firmware
  • Li-ion battery packs with charging guidance
  • Single-file HTML/CSS/JS command console (arena, telemetry, packet log)
Parameter Value
Robots 3 ESP32 differential-drive robots + 1 controller
Radio ESP-NOW, peer-to-peer, no router/pairing/cloud
Packet size Up to 250 bytes per frame
Telemetry rate 10 Hz per robot (position, heading, battery, RSSI)
Delivery target ≥ 95% in lab range (design target; room-dependent)
Round-trip ~14 ms typical uncongested (design target)
Formations Line, triangle, cluster, scatter (broadcast + ack)
Retries Up to 3 per broadcast on missing ack
Safety Ultrasonic stop-distance override, radio-independent
Console Arena map, battery/RSSI, packet log, latency chart

Project features

  • [ESP-NOW peer-to-peer link] Controller and robots communicate with connectionless ESP-NOW frames — no router, no pairing, no cloud; up to 250-byte packets with millisecond-scale latency.
  • [Formation control] Broadcast line, triangle, cluster and scatter formations; each follower runs a proportional controller toward its assigned slot while the leader follows the console joystick.
  • [Acknowledged broadcasts] Every formation packet expects per-robot acks; missing acks trigger up to 3 retries, and the console shows exactly who acknowledged and who didn't.
  • [Live arena console] The command console renders robot positions and headings in real time from 10 Hz ESP-NOW telemetry, with per-robot battery bars and RSSI.
  • [Packet + latency telemetry] Packet log, delivery-rate counter and a 60-second round-trip latency chart make the wireless link itself the observable under test — design target ≥ 95% delivery in lab range.
  • [Ultrasonic collision override] Each robot's ultrasonic sensor halts it independently of the radio link — the safety layer that works even during a comms blackout.
  • [Leader-follower roles] SWARM-01 leads (console-driven), SWARM-02/03 follow; roles are firmware constants, so the viva can discuss reassignment and failure handling.
  • [Complete build docs] Chassis assembly guide, ESP-NOW peer-registration procedure (MAC pairing), tuning guide for the follower gains, and commented firmware.

What is included

  • Complete firmware: controller (broadcast, ack/retry, joystick) + robot (peer link, follower control, telemetry, collision override)
  • Chassis assembly and wiring guide for all three robots
  • ESP-NOW peer-registration (MAC pairing) procedure
  • Command console web app (swarm control + telemetry views)
  • Follower-gain tuning guide
  • Project report PDF (background, protocol design, control design, test procedure)
  • PPT presentation for final review
  • Viva Q&A preparation document (ESP-NOW vs Wi-Fi/BLE, formation control, proportional control, acks/retries, link budget)
  • Setup guide (first power-up, peer registration, first formation)

Limitations & prerequisites

  • Delivery rate and latency are design targets measured on the buyer's own robots — walls, Wi-Fi interference and antenna orientation change them, and the report documents the measurement procedure rather than pre-claiming figures.
  • Three robots is the tested configuration; larger swarms need broadcast scheduling work the report flags as future scope.
  • Position tracking in the console comes from each robot's own odometry/telemetry, not an external motion-capture system — absolute arena accuracy is approximate.
  • Ultrasonic sensors see flat obstacles well and soft/angled ones poorly; the collision override is a safety layer, not a mapping system.
  • Battery life limits demo sessions; the report's power notes size expectations honestly.

Frequently Asked Questions

What is ESP-NOW and why use it for a swarm?

ESP-NOW is Espressif's connectionless peer-to-peer wireless protocol: devices send short packets directly to registered MAC addresses with no router, no pairing and no cloud. For a swarm that means millisecond latency and no infrastructure — Wi-Fi and Bluetooth both add association overhead that hurts multi-robot coordination.

How do the robots form shapes?

The controller broadcasts each robot's target slot for the chosen formation (line, triangle, cluster, scatter). Followers run a proportional controller driving their motors toward their slots; the leader follows the console joystick. The arena map shows the shape forming live.

What happens if a robot misses a command?

Every broadcast expects per-robot acknowledgements; missing acks trigger up to 3 retries, and the console shows the ack count (e.g. 2/3 ack) so a deaf robot is visible, not silent.

How is collision handled?

Each robot carries an ultrasonic sensor running a local loop: anything inside the stop distance halts the motors immediately, independent of the radio link — so a comms blackout can't cause a crash.

What range and reliability can we expect?

The design target is ≥ 95% packet delivery in lab range with ~14 ms typical round-trip, but these are measured on your own floor: room layout, interference and antennas all matter, and the report documents the measurement procedure.

Is this project suitable for a final-year project?

Yes — for Mechanical, Mechatronics and Robotics programs. It demonstrates wireless protocol design, multi-agent coordination, closed-loop formation control, telemetry and experimental characterization, all strong viva material. Suitable for B.E./B.Tech final-year projects in Mechanical, Mechatronics and Robotics.

Components & software requirements
  • 4x ESP32 DevKit (1 controller + 3 robots), Arduino-core firmware
  • 3x differential-drive robot chassis kits (motors, wheels, caster)
  • ESP-NOW peer-to-peer protocol (connectionless, 250-byte frames)
  • Ultrasonic distance sensors (local collision override per robot)
  • Proportional formation-follower controller firmware
  • Li-ion battery packs with charging guidance
  • Single-file HTML/CSS/JS command console (arena, telemetry, packet log)
Delivery information

Built-to-order project. Delivery timeline is shared after order confirmation based on current queue.

Support terms

Complete documentation, setup guide, and viva preparation included. Support for setup and explanation provided.

Download abstract (PDF)

Related guides

All guides
Get a quotation