Mobile Robot Algorithms Laboratory

Project domain

Simultaneous Localization and Mapping (SLAM) methods are essential for mobile robots which are supposed to act in an unknown environment. Despite various algorithms have already been proposed, an algorithm that would robustly solve the problem in the general case and satisfy performance constraints is still a subject of research. Unfortunately, there is no publicly available framework that would provide a common set of components needed to speed up SLAM research (frameworks and toolkits that simplify the development of particular SLAM parts are not taken into account).

Project goals

• Creation of a framework that acts as a constructor of SLAM algorithms (researchers are supposed to connect available components by themselves and add necessary modifications).
• Implementation of a full set of basic components that can be assembled into the most common (fundamental) SLAM algorithms.
• Creation of the infrastructure and tools for debugging and analysis of SLAM algorithms.
• Creation of service for managing SLAM datasets (convertion, storage, provision and so on).

Environment

All software we develop is supposed to be executed on Linux and in Robot Operating System environment. We performed preliminary research on a wide spectrum of existing tools and environments like MTK, MRPT and others, which showed that ROS was the most promising choice.

Roadmap

• Support for components for graph-based SLAM methods.
• Implementation of additional scan matchers for Laser-based SLAMs.
• Support for an extended set of sensors and measurements (3D scans, monocular/stereo cameras).
• Implementation of vergent stereo vision.
• A service for SLAM datasets.
• ROS SLAM Testing Farm: a virtualized, container-based environment for semi-autonomous testing of SLAM algorithms.