Interaction Spaces Between Human and Mobile Autonomous Systems

Autonomous mobile systems (from German: sabeS) have long ceased to exist exclusively in an industrial context, and are playing an increasingly important role in everyday situations. Not only explicit interactions between humans and sabeS increase, but also random encounters. The subject of this research proposal is the exploration of plannable motion sequences, which a sabeS should show to solve spatial conflict situations with human actors efficiently and satisfactorily. Such conflict situations are a common problem, considering, for example, the use of autonomous delivery robots in public spaces. Based on findings from research in proxemics, the research project presented here aims to identify interaction spaces surrounding the human in which a sabeS should show either long-term predictable or short-term avoidance manoeuvres to enhance safety, efficiency and satisfaction.A first question relates to the description of these short-term and long-term interaction spaces and the question of whether they can be better described by a temporal or spatial dependence. The long-term interaction space is characterized by greater distances between the actors and thus a higher time budget. A sabeS should accordingly adapt the intended trajectory as a function of the probable trajectory of the human being. By contrast, the short-term interaction space is characterized by shorter distances and a correspondingly shorter time budget. While currently a simple stop of the sabeS is being implemented to prevent collisions, recent findings show that efficiency and satisfaction can be increased by more meaningful movement patterns. Such short-term movement patterns are subject to further development and verification in this research project.The geometric shape and extent of the postulated long and short-term interaction spaces of humans will be investigated in dependence on the characteristics of the approaching sabeS. Size, shape and speed can play a role, as well as the angle from which a sabeS approaches the human. Also, human characteristics shall be included. Personal characteristics such as age, mental model of the approaching sabeS, and possible distractions in the environment could affect the extension and shape of the interaction spaces.The planned research project envisages the inclusion of simulation studies (virtual reality) as well as real-world and field experiments. For this purpose, a mobile robot is to be built on a modular platform that is suitable for field experiments. The results obtained will allow to draw conclusions on a novel movement design for sabeS.