Environment recognition with mobile robots

In the near future, robots will become an increasingly important part of our everyday life. The more important is the ability of a robot to perceive its environment in the best possible way and thus to adapt its actions to events in its environment and to react to them. When using robots for e.g. householdMobile RoboterThe interplay between man and machine always plays an important role in the execution of work, maintenance activities or the monitoring of buildings.

In this project, mobile robots will be used to determine their position in a building, but also the position of surrounding objects such as pedestrians. For this purpose, the robot platforms are equipped with various complementary sensors: laser scanner, camera with infrared depth information, microphones, gyroscope and odometry. In addition, the robots can communicate with each other via WLAN and exchange the information they obtain.

The aim of the project is the development of novel methods for multisensory environment detection. The robots should be able to move and locate themselves in a building independently and to detect other objects such as pedestrians and to follow their position cooperatively on a building plan. For this purpose, probabilistic multi-object tracking methods are to be applied, which fuse the acquired sensor data and thus enable a description of the environment.