Particle-based Instance-aware Semantic Occupancy Mapping in Dynamic Environments

TL;DR

This paper presents a particle-based method for creating detailed, instance-aware 3D maps in dynamic environments, effectively handling noise, segmentation errors, and object motion for improved robotic perception.

Contribution

It introduces a novel particle-based occupancy mapping approach using an augmented state and S$^2$MC-PHD filter to jointly estimate occupancy, semantics, and instances in dynamic scenes.

Findings

Outperforms state-of-the-art methods on Virtual KITTI 2 dataset

Effectively handles sensor noise and segmentation errors

Validated with real-world data

Abstract

Representing the 3D environment with instance-aware semantic and geometric information is crucial for interaction-aware robots in dynamic environments. Nevertheless, creating such a representation poses challenges due to sensor noise, instance segmentation and tracking errors, and the objects' dynamic motion. This paper introduces a novel particle-based instance-aware semantic occupancy map to tackle these challenges. Particles with an augmented instance state are used to estimate the Probability Hypothesis Density (PHD) of the objects and implicitly model the environment. Utilizing a State-augmented Sequential Monte Carlo PHD (S$^2$MC-PHD) filter, these particles are updated to jointly estimate occupancy status, semantic, and instance IDs, mitigating noise. Additionally, a memory module is adopted to enhance the map's responsiveness to previously observed objects. Experimental results on the Virtual KITTI 2 dataset demonstrate that the proposed approach surpasses state-of-the-art methods across multiple metrics under different noise conditions. Subsequent tests using real-world data further validate the effectiveness of the proposed approach.