RESUMEN
Hyperconnectivity via modern Internet of Things (IoT) technologies has recently driven us to envision "digital twin", in which physical attributes are all embedded, and their latest updates are synchronized on digital spaces in a timely fashion. From the point of view of cyberphysical system (CPS) architectures, the goals of digital twin include providing common programming abstraction on the same level of databases, thereby facilitating seamless integration of real-world physical objects and digital assets at several different system layers. However, the inherent limitations of sampling and observing physical attributes often pose issues related to data uncertainty in practice. In this paper, we propose a learning-based data management scheme where the implementation is layered between sensors attached to physical attributes and domain-specific applications, thereby mitigating the data uncertainty between them. To do so, we present a sensor data management framework, namely D2WIN, which adopts reinforcement learning (RL) techniques to manage the data quality for CPS applications and autonomous systems. To deal with the scale issue incurred by many physical attributes and sensor streams when adopting RL, we propose an action embedding strategy that exploits their distance-based similarity in the physical space coordination. We introduce two embedding methods, i.e., a user-defined function and a generative model, for different conditions. Through experiments, we demonstrate that the D2WIN framework with the action embedding outperforms several known heuristics in terms of achievable data quality under certain resource restrictions. We also test the framework with an autonomous driving simulator, clearly showing its benefit. For example, with only 30% of updates selectively applied by the learned policy, the driving agent maintains its performance about 96.2%, as compared to the ideal condition with full updates.
RESUMEN
Voice-based interfaces have become one of the most popular device capabilities, recently being regarded as one flagship user experience of smart consumer devices. However, the lack of common coordination mechanisms might often degrade the user experience, especially when interacting with multiple voice-enabled devices located closely. For example, a hotword or wake-up utterance such as "hi Bixby" or "ok Google" frequently triggers redundant responses by several nearby smartphones. Motivated by the problem of uncoordinated react of voice-enabled devices especially in a multiple device environment, in this paper, we discuss the notion of an ephemeral group of consumer devices in which the member devices and the transient lifetime are implicitly determined by an external event (e.g., hotword detection) without any provisioned group structure, and specifically we concentrate on the time-constrained leader election process in such an ephemeral group. To do so: (i) We first present the sound-based multiple device communication framework, namely tailtag, that leverages the isomorphic capability of consumer devices for the tasks of processing hotword events and transmitting data over sound, and thus renders both the tasks confined to the same room area and enables the spontaneous leader election process in a unstructured group upon a hotword event. (ii) To improve the success rate of the leader election with a given time constraint, we then develop the adaptive messaging scheme especially tailored for sound-based data communication that inherently has low data rate. Our adaptive scheme utilizes an application-specific score that is individually calculated by a member device for each event detection, and employs score-based scheduling by which messages of a high score are scheduled first and so unnecessary message transmission can be suppressed during the election process. (iii) Through experiments, we also demonstrate that, when a hotword is detected by multiple smartphones in a room, the framework with the adaptive messaging scheme enables them to successfully achieve a coordinated response under the given latency bound, yielding an insignificant non-consensus probability, no more than 2%.