DeltaSeek: Toward Active Perception in Evolving Construction Environments

2026-09-27

AI Summary

This study addresses the failure of static mapping in dynamic built environments and the constraints imposed by robot embodiment on observational capabilities by proposing an active perception framework. By formally defining the admissible observation set, it distinguishes perceptual infeasibility from acquisition cost and delineates the adaptability boundaries of different embodied configurations to environmental changes. The framework is validated on a Husky A300 platform equipped with a UR5e manipulator and RGB-D sensors, utilizing Industry Foundation Classes (IFC) benchmarks alongside an exhaustive pose evaluation algorithm. Experimental results demonstrate that a wrist-mounted camera effectively resolves the occlusion blind spots inherent to chassis-mounted cameras, significantly reducing the median travel distance from 15.2 m to 6.0 m. These findings substantiate the superiority of embodied perception for adaptive robotic mapping in dynamic architectural settings.

Abstract

Construction environments evolve continuously, causing large geometric changes that degrade static mapping and registration performance. This necessitates active perception, where robots deliberately select sensing configurations to resolve the environment's current state. We present DeltaSeek, an initial framework toward active perception in evolving built environments. While our broader objective is a system that reasons about where, how, and when to observe, this paper addresses a critical prerequisite: how a robot's sensing embodiment constrains the observations it can acquire. We formalize an embodiment's permissible observation set and evaluate with a Husky A300 equipped with a UR5e on an IFC-derived benchmark under chassis-mounted and wrist-mounted RGB-D configurations, scoring observations by geometric visibility and effort by drivable distance. In a room-scale scene with eight controlled changes spanning four observability conditions, exhaustive evaluation over 240 permissible base poses and five arm postures shows that two changes admit no chassis viewpoint whatsoever, while the wrist camera resolves both. For changes observed by both embodiments, the median base travel is $6.0$~m for the wrist camera and $15.2$~m for the chassis camera. These results distinguish sensing limitations from acquisition costs, clarifying whether an observation is impossible or simply requires more travel.

Problem

• active perception
• evolving construction environments
• sensing embodiment
• geometric changes
• observability

Innovation

• Active Perception
• Evolving Environments
• Sensing Embodiment
• Permissible Observation Set
• Construction Robotics