Drones are expected to operate autonomously, yet they will also interact with humans to solve tasks together. To support civilian human-drone teams, we propose a distributed architecture where sophisticated operations such as image recognition, coordination with humans, and flight-control decisions are made, not on-board the drone, but remotely. The benefits of such an architecture are the increased computational power available for image recognition and the possibility to integrate interfaces for humans. On the downside, communication is necessary, resulting in the delayed reception of commands. In this article, we discuss the design considerations of the distributed approach, a sample implementation on a smartphone, and an application to the concrete use case of bookshelf inventory. Further, we report experimentally-derived first insights into messaging and command response delays with a custom drone connected through Wi-Fi.

No abstract

We investigate whether naturalistic emotional human feedback can be directly exploited as a reward signal for training artificial agents via interactive human-in-the-loop reinforcement learning. To answer this question, we devise an experimental setting inspired by animal training, in which human test subjects interactively teach an emulated drone agent their desired command-action-mapping by providing emotional feedback on the drone's action selections. We present a first empirical proof-of-concept study and analysis confirming that human facial emotion expression can be directly exploited as reward signal in such interactive learning settings. Thereby, we contribute empirical findings towards more naturalistic and intuitive forms of reinforcement learning especially designed for non-expert users.

Social Bots are software robots implementing algorithms that autonomously produce content and interact with users of social networks or mimic the behaviour of humans in multiple other forms of digital communication. If they act on behalf of a specific user of party in the communication, this user is sometimes called the "owner" of the bot. Social bots are typically designed to be, benevolent and even useful but some of them are built to manipulate and deceive social users.