Sign in Subscribe
Robotics industry

Robotics Researchers Publish New Report on People Detection Using 2D Range Finders

Strategic Insights

— 1 min read

Robots that navigate environments and detect humans in their vicinity require accurate knowledge of their surroundings, especially when it comes to human presence. Traditionally, computer vision has been the primary approach for detecting humans, but robotics researchers have now highlighted the potential of 2D range finders as an often-overlooked hardware feature for people detection. A new study published by researchers at the University Pablo de Olavide in Seville, Spain, introduces a new dataset, FROG, for people detection using knee-high 2D range finders, which offers greater laser resolution, scanning frequency, and annotation data compared to existing datasets.

Key Takeaways:

  • The researchers highlighted the potential of 2D range finders for people detection, which is often overlooked despite their convenience and field of view.
  • The new dataset, FROG, offers greater laser resolution, scanning frequency, and annotation data compared to existing datasets such as DROW.
  • FROG includes annotations for 100% of its laser scans, 17x more annotated scans, 100x more people annotations, and over twice the distance traveled by the robot.
  • The proposed benchmark based on the FROG dataset analyzes the performance of state-of-the-art people detectors using 2D range finder data.
  • The new end-to-end deep learning approach for people detection achieves results comparable to the state of the art while avoiding CPU preprocessing.

Statistics:

  • The FROG dataset contains 100% annotated laser scans, with 17x more annotated scans and 100x more people annotations compared to DROW.
  • The proposed people detector attains results comparable to the state of the art, while an optimized implementation for ROS can operate at more than 500 Hz.

Sources:

  • Beyer et al. (2018): DROW dataset
  • Frontiers Media S.A.: Publisher for Frontiers in Robotics and AI
  • Roediger, J. H. (2025). FROG: a new people detection dataset for knee-high 2D range finders. Frontiers in Robotics and AI, 2025, 12. (Frontiers in Robotics and AI - http://www.frontiersin.org/Robotics_and_AI)
  • NewsRx LLC: Copyright 2025, NewsRx LLC (verticalnews.com)
  • Journal of Engineering: Report - NewsRx. New Robotics Study Findings Recently Were Published by Researchers at University Pablo de Olavide (FROG: a new people detection dataset for knee-high 2D range finders). Journal of Engineering. November 3, 2025; p 2241.

Read more