POCD: Probabilistic Object-Level Change Detection and Volumetric Mapping in Semi-Static Scenes

TL;DR

This paper introduces POCD, a probabilistic framework for detecting object-level changes and updating volumetric maps in semi-static scenes, improving map accuracy for robotic navigation over time.

Contribution

It proposes a novel probabilistic object state representation and Bayesian update rule that jointly models object pose changes, map consistency, and semantic information.

Findings

Outperforms state-of-the-art in semi-static environment reconstruction.

Effectively tracks object pose changes over time.

Demonstrates robustness on real-world and public datasets.

Abstract

Maintaining an up-to-date map to reflect recent changes in the scene is very important, particularly in situations involving repeated traversals by a robot operating in an environment over an extended period. Undetected changes may cause a deterioration in map quality, leading to poor localization, inefficient operations, and lost robots. Volumetric methods, such as truncated signed distance functions (TSDFs), have quickly gained traction due to their real-time production of a dense and detailed map, though map updating in scenes that change over time remains a challenge. We propose a framework that introduces a novel probabilistic object state representation to track object pose changes in semi-static scenes. The representation jointly models a stationarity score and a TSDF change measure for each object. A Bayesian update rule that incorporates both geometric and semantic information is derived to achieve consistent online map maintenance. To extensively evaluate our approach alongside the state-of-the-art, we release a novel real-world dataset in a warehouse environment. We also evaluate on the public ToyCar dataset. Our method outperforms state-of-the-art methods on the reconstruction quality of semi-static environments.