Semantic cues improve robotic mapping near ambiguous indoor doors
Semantic-Aware Predictive Mapping for Exploration and Navigation
Robotics
Summary
Mapping unknown indoor spaces using robots can be tricky, especially near doors that look like walls but lead to new rooms. The authors show that adding information about doors helps the robot better predict what lies beyond these tricky areas. They compared two models and found that the one using door information made much more accurate predictions near the doors. This means robots can explore indoor spaces more effectively by knowing where doors are.
What this means in practice
- •For robotics engineers: Improve robot navigation by enhancing map predictions around ambiguous indoor doors using semantic door information.
- •For smart building designers: Create advanced indoor mapping tools that better handle door regions for automated building inspections or maintenance.
Authors
Kenneth J. K. Ong, William W. J. Teo
Abstract
Predictive mapping can support robotic exploration and navigation by estimating unseen geometric layouts from partial occupancy observations. However, occupancy-only representations may fail to distinguish semantically different structures with similar geometry. This is particularly relevant for indoor doors, which may appear as occupied cells like walls but indicate possible connected rooms or corridors beyond the observed region. This work investigates whether semantic door cues improve predictive geometric occupancy mapping around such ambiguous regions. We modify a subset of the CogniPlan dataset by inserting door-induced ambiguities into partial occupancy maps while keeping the ground-truth layouts unchanged. We compare a geometry-only control model with a semantic-cued model trained on the same modified dataset, where the semantic-cued model receives an additional door channel. Evaluation uses L1 error, F1 score, and Intersection over Union (IoU) over both the full map and a 10-pixel door-region mask. Full-map performance remains broadly similar between models, but localized door-region results show a clear qualitative improvement: L1 decreases from 0.004342 to 0.000025, while F1 and IoU improve from 0.031311 and 0.015905 to 1.000000 and 1.000000, respectively. These results suggest that semantic cues can improve predictive occupancy completion in regions where geometric observations alone are ambiguous.