A conceptual, and practical, di culty with the Continuous-time Generalised Predictive controller is solved by replacing the continuouslymoving horizon by an intermittently moving horizon. This allows slow optimisation to occur concurrently with a fast control action. Some nonlinear simulations illustrate the potential of this approach.

No abstract

No abstract

The development of new biosensor technologies and their active use as wearable devices have offered mobility and flexibility to conventional western medicine and personal fitness tracking. In the development of biosensors, transducers stand out as the main elements converting the signals sourced from a biological event into a detectable output. Combined with the suitable bio-receptors and the miniaturization of readout electronics, the functionality and design of the transducers play a key role in the construction of wearable devices for personal health control. Ever-growing research and industrial interest in new transducer technologies for point-of-care (POC) and wearable bio-detection have gained tremendous acceleration by the pandemic-induced digital health transformation. In this article, we provide a comprehensive review of transducers for biosensors and their wearable applications that empower users for the active tracking of biomarkers and personal health parameters.

Behavior mapping (BM) is a spatial data collection technique in which the locational and behavioral information of a user is noted on a plan layout of the studied environment. Among many indoor positioning technologies, we chose Wi-Fi, BLE beacon and ultra-wide band (UWB) sensor technologies for their popularity and investigated their applicability in BM. We tested three technologies for error ranges and found an average error of 1.39 m for Wi-Fi in a 36 m2 test area (6 m × 6 m), 0.86 m for the BLE beacon in a 37.44 m2 test area (9.6 m × 3.9 m) and 0.24 m for ultra-wide band sensors in a 36 m2 test area (6 m × 6 m). We simulated the applicability of these error ranges for real-time locations by using a behavioral dataset collected from an active learning classroom. We used two UWB tags simultaneously by incorporating a custom-designed ceiling system in a new 39.76 m2 test area (7.35 m × 5.41 m). We considered 26 observation points and collected data for 180 s for each point (total 4680) with an average error of 0.2072 m for 23 points inside the test area. Finally, we demonstrated the use of ultra-wide band sensor technology for BM.

No abstract

ABSTRACT

ABSTRACT