Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Quality-guided UAV Surface Exploration for 3D Reconstruction

Benjamin Sportich, Kenza Boubakri, Olivier Simonin, Alessandro Renzaglia

TL;DR

<3-5 sentence high-level summary> This paper addresses efficient unmanned aerial vehicle (UAV) exploration for 3D surface reconstruction under user-defined quality constraints. It introduces a modular Next-Best-View framework that uses a TSDF-based map and a reconstruction-quality objective to guide both view generation and view selection, enabling adaptive trade-offs between speed and fidelity. The method combines frontier-based view generation with an information-gain plus navigation-cost evaluation, all grounded in a quality function and supported by a Voxblox-based local planner. Experimental results in indoor and outdoor simulations show superior coverage and competitive map accuracy compared with baselines, while demonstrating the ability to tune exploration behavior to match prescribed quality targets.</endofparagraph>

Abstract

Reasons for mapping an unknown environment with autonomous robots are wide-ranging, but in practice, they are often overlooked when developing planning strategies. Rapid information gathering and comprehensive structural assessment of buildings have different requirements and therefore necessitate distinct methodologies. In this paper, we propose a novel modular Next-Best-View (NBV) planning framework for aerial robots that explicitly uses a reconstruction quality objective to guide the exploration planning. In particular, our approach introduces new and efficient methods for view generation and selection of viewpoint candidates that are adaptive to the user-defined quality requirements, fully exploiting the uncertainty encoded in a Truncated Signed Distance field (TSDF) representation of the environment. This results in informed and efficient exploration decisions tailored towards the predetermined objective. Finally, we validate our method via extensive simulations in realistic environments. We demonstrate that it successfully adjusts its behavior to the user goal while consistently outperforming conventional NBV strategies in terms of coverage, quality of the final 3D map and path efficiency.

Quality-guided UAV Surface Exploration for 3D Reconstruction

TL;DR

<3-5 sentence high-level summary> This paper addresses efficient unmanned aerial vehicle (UAV) exploration for 3D surface reconstruction under user-defined quality constraints. It introduces a modular Next-Best-View framework that uses a TSDF-based map and a reconstruction-quality objective to guide both view generation and view selection, enabling adaptive trade-offs between speed and fidelity. The method combines frontier-based view generation with an information-gain plus navigation-cost evaluation, all grounded in a quality function and supported by a Voxblox-based local planner. Experimental results in indoor and outdoor simulations show superior coverage and competitive map accuracy compared with baselines, while demonstrating the ability to tune exploration behavior to match prescribed quality targets.</endofparagraph>

Abstract

Reasons for mapping an unknown environment with autonomous robots are wide-ranging, but in practice, they are often overlooked when developing planning strategies. Rapid information gathering and comprehensive structural assessment of buildings have different requirements and therefore necessitate distinct methodologies. In this paper, we propose a novel modular Next-Best-View (NBV) planning framework for aerial robots that explicitly uses a reconstruction quality objective to guide the exploration planning. In particular, our approach introduces new and efficient methods for view generation and selection of viewpoint candidates that are adaptive to the user-defined quality requirements, fully exploiting the uncertainty encoded in a Truncated Signed Distance field (TSDF) representation of the environment. This results in informed and efficient exploration decisions tailored towards the predetermined objective. Finally, we validate our method via extensive simulations in realistic environments. We demonstrate that it successfully adjusts its behavior to the user goal while consistently outperforming conventional NBV strategies in terms of coverage, quality of the final 3D map and path efficiency.

Paper Structure

This paper contains 20 sections, 12 equations, 4 figures, 4 tables.

Table of Contents

  1. Introduction
  2. Related Work
  3. Entropy & heuristics
  4. 3D representation
  5. Problem Formulation
  6. Reconstruction quality in TSDF
  7. Proposed Approach
  8. Use of the quality function to guide the exploration
  9. View Generation
  10. View Evaluation
  11. Information Gain Component
  12. Navigation component
  13. Local Planning
  14. Numerical Experiments
  15. Simulation setup
  16. ...and 5 more sections

Figures (4)

  • Figure 1: Illustration of an autonomous exploration and 3D reconstruction task with an aerial robot. Red points in the free space represent candidate viewpoints generated to select the next robot's position, black ones are surface frontiers within the sensor's range.
  • Figure 2: Vertical and horizontal rotation for the view generation (coordinate system centered on the frontier $i)$.
  • Figure 3: House and Tunnel environments (ground-truth mesh) with 15-minute trajectories.
  • Figure 4: Software stack