Original title: A hybrid pipeline for dynamic ontology-based semantic mapping
Konstantinos Dimitropoulos, Ioannis Hatzilygeroudis · September 2026
Robots need to understand what objects are around them and where they sit in a room to navigate properly. Standard systems often lack deep context about their surroundings. To fix this, researchers created a hybrid mapping pipeline that builds an active, meaningful map of the world. The setup uses an external camera to place objects in physical space through geometric projection. It detects items, tracks them continuously, and uses linear regression models to fix errors in estimated real world coordinates. Then, it organizes everything using an ontology, which is a structured web of concepts and rules. This system constantly updates object instances, physical positions, and relationships in real time, giving the robot a clear, changing model of its environment.
Imagine you're tidying a living room while people are still moving things around. Every time you put something away, someone else moves it, so your mental map of where things are keeps getting outdated. Here, the robot faces the same problem with objects in an environment. This system uses a camera to spot changes and a structured list to update the robot's understanding instantly, so it always knows what's where and what it's used for.
Robots need to understand not just where things are, but what they are and how they relate. This system helps them navigate and interact in messy, changing environments like homes or warehouses.
The abstract does not evaluate performance in real-world scenarios or compare with other systems. It focuses on the design and integration of components, not on testing outcomes.
Ask anything about this paper. Answers come from its abstract, and say so when the abstract doesn't cover your question.
A hybrid pipeline for dynamic ontology-based semantic mapping — Konstantinos Dimitropoulos, Ioannis Hatzilygeroudis
Abstract licence: CC0
Explanation generated by Synthesis from the abstract — it is not the original paper. This is a preprint, so it hasn't necessarily been peer reviewed.