There is an industrial need for robust and reliable localisation solution relying as little as possible on external absolute positioning sensors in the environment. This is needed because currently all external localisation solutions for mobile systems have some shortcomings: SLAM-based localisation using only on-board sensors, such as LIDAR or 2D/3D camera, fails in highly dynamic or repetitive environments, and absolute localisation technologies are not sufficiently robust: GPS link can fail, UWB is inaccurate in non-line of sight conditions, fiducial markers require direct line of sight.
The goal of the project is to create a robust and cost-efficient localisation framework that can be used to combine all types of localisation, overcoming the shortcomings of its subcomponents. For this, a graphSLAM based localisation framework will be developed that will combine both relative and absolute localisation sensors. Potential use cases that can use this framework are:
- Overcome jamming of GPS signals or signal loss on drones by combining GPS with visual SLAM
- Realise accurate localisation of an AGV when docking by combining SLAM with fiducial markers placed at the docking stations
- Combination laser scanner-based SLAM and Wi-Fi fingerprinting in long repetitive corridors
MARLOC_ICON is an Industrial Research ICON project. We are looking for companies to join the User Group and work with us on the valorisation of the project.
Interested? Complete the form below and we will contact you as soon as possible.