In the S³ research project, researchers from Fraunhofer IPA worked with project partners to develop safety technologies and recognition algorithms for service robots. They were tested on the Care-O-bot 4 service robot in a nursing home application.
Unlike industrial robots, mobile service robots move in dynamic environments and often among people who are not familiar with them. In the environment of inpatient care, they can also encounter people who are particularly in need of protection. The S³ research project has therefore developed technologies that improve the perception of objects and the environment as well as the handling capabilities of service robots, with a special focus on functional safety. The project received funding from the German Federal Ministry of Education and Research.
The further development of appropriate software technologies for environment detection, the safe grasping of objects and their integration on the service robot 'Care-O-bot 4' were the focus of the work at Fraunhofer IPA. A combination of radar sensor and cameras was used on the robot for this purpose. The radar sensor enables the robot to reliably detect people. The camera data is evaluated using machine learning methods to record what a recognized person is currently doing. By determining body posture based on joint positions, the software can assign actions such as walking or reaching. This enables the service robot to adapt its behavior to its human counterpart in such a way that everything proceeds safely.
Furthermore, the IPA researchers improved the robot's understanding of its environment. Previous algorithms have had problems identifying and localizing transparent objects in space with commercially available 3D cameras. The newly developed algorithms are now able to do so. Neural networks help the robot to better detect water bottles, for example, and determine their exact position. The basis for this was a test scenario in which the robot replaces empty water bottles with full ones in a nursing home. On command, the robot moves to a desired resident's room, detects the bottle it is looking for, picks it up and takes it to the kitchen in the living area. Thanks to an advanced environment model, the robot 'knows' where tables are so that it can specifically search for bottles in the residents' rooms.
Safe gripping was also further developed in the project. IPA experts have developed software for this purpose that dynamically adapts arm movements to the recorded environmental data in order to avoid obstacles. For example, if a nursing home resident spontaneously reaches for a glass while the robot wants to pick up the water bottle next to it, it quickly reschedules its movement.