The Mapping theme is all about building useful representations of workspaces. Sometimes these maps are used by our robots to localise. Sometimes the maps are themselves intrinsically valuable as surveys which contain both spatial and semantic information.

Building maps over large spatial scales and updating them over large temporal scales is both fascinating and challenging.  Large scale workspaces change appearance and geometry over time – in some places this is a creeping or a cyclical process, like day and night lighting; in others it is sudden, like trucks and crowds moving.

We use both cameras and lasers in our mapping work. Sometimes we build purely visual maps whilst in other cases we build dense 3D point clouds from laser sensors. Our most recent work concerns dense mapping from low cost cameras. Expect to see much more on this soon.

 

Our latest research

Building, Curating, and Querying Large-scale Data Repositories for Field Robotics Applications

Abstract—Field robotics applications have some unique and unusual data requirements -- the curating, organisation and management of which are often ...
Read More

Leveraging Experience for Long-Term LIDAR Localisation In Changing Cities

 Successful approaches to autonomous vehicle localisation and navigation typically involve 3D LIDAR scanners and a static, curated 3D map, both ...
Read More

Accelerating FAB-MAP with Concentration Inequalities

Abstract—We outline an approach for using concentration inequalities to perform rapid approximate multi-hypothesis testing. In a scenario where multiple hypotheses ...
Read More

Adaptive compression for 3D laser data

Abstract - This paper concerns the creation of efficient surface representations from laser point clouds created by a push broom laser ...
Read More

Discovering and Mapping Complete Surfaces With Stereo

This paper is about the automated discovery and mapping of surfaces using a stereo pair. We begin with the observation ...
Read More

Efficient Non-Parametric Surface Representations Using Active Sampling for Push Broom Laser Data

Abstract—This paper concerns the creation of efficient surface representations from laser point clouds. We produce a continuous, implicit, non-parametric representation ...
Read More

FAB-MAP 3D: Topological Mapping with Spatial and Visual Appearance

Abstract— This paper describes a probabilistic framework for appearance based navigation and mapping using spatial and visual appearance data. Like ...
Read More

Generating Implicit Surfaces from Lidar Data

Abstract—This paper is concerned with generating a continuous implicit representation of a robot’s workspace using sparse point cloud data. We ...
Read More

FAB-MAP 3D: Topological Mapping with Spatial and Visual Appearance

Abstract— This paper describes a probabilistic framework for appearance based navigation and mapping using spatial and visual appearance data. Like ...
Read More

Semantic Mapping

Autonomous vehicles operating in places like parking lots can leverage of a higher-level understanding of the objects around it. For instance, ...
Read More