Indoor-Robots

MARVIN

ARTOS

Since the early days of service robotics mainly indoor robot applications have been examined, such as cleaning robot in private households, transport systems in factories or logistics centers. Meanwhile, there is an extensive range of products, which can be used in different applications. In research, the focus is mainly in the exploration of unknown areas, environmental modeling and identification of persons, whereas the intention of a person in the surrounding of the robot has to be taken into account for interaction and for a sequence of actions of the robot.

The research in the field of indoor robot focuses mainly on the exploration of unknown areas of buildings. For this purpose, different basic methods have been developed that allow time-efficient exploration. Thus, for example, a larger floor in a typical office building can be detected and mapped for subsequent transport or monitoring tasks completely automatically with sufficient accuracy. For the validation of these approaches, the service robot MARVIN is used. With the help of MARVIN, also questions about logistics are investigated within the SmartFactory-project.

With the help of the robot system ARTOS, two different research areas are discussed. Firstly, the question, how sensor networks that provide incomplete descriptions of the state of the environment or of persons in the vicinity of the robot can be used for the development of cost-effective and reliable transportation systems. Probabilistic methods are investigated in order to provide security assessments that take into account measurement errors or failures of simple sensors. The second line of research examines how mobile robot systems for the care of the elderly can be used, e.g. to detect emergency situations on time or to support simple household tasks. Besides a fast navigation in a typical apartment environment,  the identification and modeling of the behavior of persons over an extended period of time is a fundamental problem. With the help of these behavior patterns, on one hand search strategies of the robot can be optimized and, second, the implementation of the service tasks can be supported.

For student education, small forklift robots are used to introduce the basic methods such as localization, mapping and navigation for mobile robots in an experimental environment.