Research Analyzer
← Back ICRA 2026

Contact-Driven Localization in a Freeform Robotic Self-Assembled Structure

Mohammadali Rashidioun, Michael Sosa, Petras Swissler

PDF

AI summary

Key figure (auto-extracted from paper)
Modular robots can accurately localize themselves in 3D freeform structures using only local contact sensing and a virtual-force framework, eliminating the need for external tracking infrastructure.
modular robots contact-driven localization self-assembly swarm robotics virtual-force framework freeform structures

Problem

Accurate localization in swarm robotics is hindered by reliance on expensive external tracking systems or high-cost sensors, which limit scalability and deployment in unstructured 3D environments.

Approach

The method uses a virtual-force framework that iteratively refines robot poses by attracting toward physically connected neighbors and repelling from nearby unconnected docks, relying solely on local binary contact cues and onboard IMU data.

Key results

  • Accurate 3D freeform localization achieved using only local contact sensing
  • Tower geometry progressively narrows feasible placement, reducing error variability at greater heights
  • Removing flipping-history Z-cues increases RMSE by ~12% and structural disparity by 13%
  • Eliminating global disconnected repulsion increases RMSE by ~44% and disparity by 47%, proving its necessity for preventing overlap

Why it matters

Enables scalable, low-cost, infrastructure-free 3D self-assembly for swarm robotics in obstructed or outdoor environments.

Abstract

Accurate localization remains a key challenge in swarm robotics, particularly for self-reconfigurable systems that must identify relative positions to form diverse struc- tures. Most existing approaches rely on external tracking infrastructure or high-cost sensors, which limit scalability and deployment in unstructured environments. In this paper, we propose a novel contact-driven localization method for mod- ular robots that leverages only local communication through binary contact information (whether two robots are physically connected or not). To exploit these contact cues, we introduce a virtual-force framework in which robots iteratively refine their poses—attracting toward dock-connected neighbors and repelling from non-connected ones. The method requires no external infrastructure and relies only on minimal onboard sensing. Simulations show effective localization during the assembly of towers and cantilevers, enabling accurate, scalable, free-form self-assembly.

Index terms

Localization Swarm Robotics Multi-Robot Systems

Related papers