Localization algorithms are necessary for a robot to know where it is. These usually work with sensor data from which the position can be calculated. In many cases, it is still very time-consuming for engineers and researchers to find out how a sensor is constructed, in which format the sensor data is communicated and which position calibration this sensor has on a robot. Christian Brommer and his colleagues in the Control of Networked Systems research group at the University of Klagenfurt have developed a new method that eliminates the need for all of this: The algorithm can automatically recognize the sensor model and use it to calculate important data for localization
Modern robotic systems - such as drones or autonomous vehicles - use a variety of sensors, from cameras and accelerometers to GPS modules. However, their correct integration has so far required expert knowledge and complex calibration
Christian Brommer, Alessandro Fornasier, Jan Steinbrener and Stephan Weiss, all members of the Control of Networked Systems research group at the time of the research work, have now developed a new method and published it in the renowned journal IEEE Transactions on Robotics (T-RO). This method enables robots to automatically identify the type of a newly added sensor, estimate its position and orientation and correctly integrate it into the existing navigation system
-With the method we present, it is no longer necessary to know which sensor is being used. Whether GPS, magnetometer/compass or speedometer - the data can simply be passed on to the algorithm and the sensor model is automatically recognized," says Christian Brommer. However, the researchers need some movement for the recognition, as he goes on to explain: -This can be done hand-held in a lab, for example, or, as we showed in the paper, while flying a quadcopter and while driving a car.-
There is undoubtedly a need for the method: on GitHub, a platform for open source projects, there are more than 14,000 requests from developers for the keywords -sensor model integration-. -Our work aims to make the integration of sensors into localization solutions such as filters easier, faster and more robust," says Christian Brommer
Christian Brommer, Alessandro Fornasier, Jan Steinbrener & Stephan Weiss (2025). Sensor Model Identification via Simultaneous Model Selection and State Variable Determination. IEEE Transactions on Robotics, https://ieeexplore.ieee.org/document/11078000
Drone Hall News Control of Networked Systems News Intelligent System Technologies News TeWi
https://www.aau.at/wp-content/uploads/2025/08/AdobeStock_1630812078.jpeg 2448 4368 Research Communication https://www.aau.at/wp-content/uploads/2015/09/aau-logo-300x110-300x110without-background3.png Research Communication 2025-08-12 08:26:11 2025-08-12 08:26:11