Back to Projects

Autonomous
Metro Cleaner

Development of a robust autonomous navigation system using Nav2, Fast-LIO SLAM, and coverage planning for disinfecting Barcelona's metro trains.

ROS 2 Nav2 Fast-LIO
Metro Environment Simulation

01. The Challenge

Disinfecting metro carriages manually is time-consuming and hazardous. This project aimed to automate the process using a mobile robot in the confined, highly symmetrical environment of Barcelona's metro trains.

The main challenges were localization drift in long, uniform corridors (the "hallway problem") and ensuring 100% surface coverage for effective cleaning.

02. System Architecture

Localization & SLAM

After benchmarking 5 state-of-the-art methods (including ORB-SLAM3 and RTAB-Map), Fast-LIO was selected.

  • Zero drift in long corridors
  • Robust to reflection noise

Coverage Planning

Standard planners failed in the narrow metro aisles. A custom ROS 2 Wrapper for a Coverage Planner was implemented.

  • Adapted to robot footprint
  • Complete area coverage

Navigation Stack

Built on Nav2 (ROS 2 Humble), utilizing the Regulated Pure Pursuit controller. This ensured smoother path following and reduced oscillation compared to the default DWB controller, essential for close-quarters maneuvering inside the train.

Real-World Validation

Final experiments were conducted in a real metro carriage in Barcelona.

Obstacle Avoidance SUCCESS
Full Coverage SUCCESS

Internship Details

Company Eurecat - Centre Tecnològic de Catalunya

Role Robotic Software Engineer Intern (Aug 2024 – Nov 2024)