Work

Unsupervised Change Detection for Space Habitats Using 3D Point Clouds

Year: 2024

Type: conference-paper

Abstract: This work presents an algorithm for scene change detection from point clouds to enable autonomous robotic caretaking in future space habitats. Autonomous robotic systems will help maintain future deep-space habitats, such as the Gateway space station, which will be uncrewed for extended periods. Existing scene analysis software used on the International Space Station (ISS) relies on manually-labeled images for detecting changes. In contrast, the algorithm presented in this work uses raw, unlabeled point clouds as inputs. The algorithm first applies modified Expectation-Maximization Gaussian Mixture Model (GMM) clustering to two input point clouds. It then performs change detection by comparing the GMMs using the Earth Mover’s Distance. Experiments on data collected in a ground environment replicating the visual features of the ISS with an Astrobee free-flyer at NASA Ames Research Center demonstrate detection of multiple appearing and disappearing objects. The source code is publicly released to promote further development.

Source: AIAA SciTech Forum

DOI: 10.2514/6.2024-1960

Authors: Jamie Santos, Holly Dinkel, Julia Di, Marina Moreira, Brian Coltin, Paulo Borges, Trey Smith, Oleg Alexandrov

Institutions: Chalmers University of Technology, University of Illinois Urbana-Champaign, Stanford University, Ames Research Center, Commonwealth Scientific and Industrial Research Organisation

Language: English

Topic: Robotics and Sensor-Based Localization

Subfield: Aerospace Engineering

Field: Engineering

Domain: Physical Sciences

Open Access status: green

Funders: National Aeronautics and Space Administration, Commonwealth Scientific and Industrial Research Organisation, Chalmers Tekniska Högskola, Nuclear Safety and Security Commission

Award: THE CURRENT PLANETARY ROVER PARADIGM CONSISTS OF LARGE BOXY SIX-WHEELED ROBOTS BUT SMALL TO MICRO-SIZED ROBOTS HAVE SIGNIFICANT POTENTIAL IN SPACE EXPLORATION. AN ORIGAMI-INSPIRED FOLDING SMALL ROBOT HAS MANY BENEFITS FOR TERRAIN EXPLORATION DUE TO ITS FOLDING CHASSIS: A FOLDED CONFIGURATION LOWERS THE CENTER OF GRAVITY FOR TACKLING STEEP INCLINES AND SAVES SPACE FOR TIGHT CREVICES AND CAVERNS. THE COMMON APPROACH FOR FOLDING ELECTRONICS IS TO SIMPLY ADAPT TRADITIONAL ELECTRONICS FOR FOLDING. IN CONTRAST A SENSOR THAT IS ENTIRELY FLEXIBLE AND STRETCHABLE AND THUS ABLE TO BE ARBITRARILY SHAPED ALLOWS FOR A GREATER VARIETY OF MECHANICAL DESIGNS SENSOR PLACEMENTS AND SENSING CAPABILITIES OF THE ROBOT. A CAPACITIVE FLEXIBLE AND STRETCHABLE SENSOR CAN BE MADE BY SANDWICHING A PLIABLE DIELECTRIC SUCH AS SILICONE IN BETWEEN TWO FLEXIBLE ELECTRODE LAYERS. THE RESEARCH PROPOSED AIMS TO DEMONSTRATE THE FEASIBILITY OF SOFT STRETCHABLE/ FLEXIBLE TEXTILE-SILICONE COMPOSITE SENSORS AND EVALUATE THEIR PERFORMANCE FOR NOVEL SPACE ROBOTICS. THIS WOULD FULFILL RESEARCH IN TECHNICAL AREA 04: ROBOTICS AND AUTONOMOUS SYSTEMS SECTION 1 ITEM 5 (4.1.5). BY LEVERAGING TEXTILES DURING THE CONSTRUCTION PROCESS A TEXTILE COMPOSITE SENSOR IS INHERENTLY SUITABLE FOR FOLDING AND STRETCHING. CREATING A NONRIGID AND FULLY FOLDABLE SENSOR SUITE OPENS OPPORTUNITIES FOR FOLDING ROBOTS AND OTHER NOVEL ROBOTIC CONFIGURATIONS.

OpenAlex ID: W4391300995

FWCI: 1.637

Cites: 26

Cited by: 2

Related to: 10