Digital Society value chain vision: Digital society requires the closer integration of technological systems and human users. This requires those systems to accurately interpret real world stimuli, inferring activity and intention. Such stimuli are poorly differentiated, indistinct, and contradictory. AI’s promise is to provide interpretation of stimuli in these situations that cannot be adequately modelled in the traditional way, thus allowing reactive processes to be better informed and enabling a closer integration of humans and digital infrastructure. The value chain integrates the edge AI technological developments into three demonstrators.
VCD 5.1: Machine Room demonstrator (lead: TECHNO, NXP-NL, TUE) – aims to showcase in a real-world environment that prognostic monitoring of infrastructure is a realistic proposition, with largely decreased prevalence of false positives. This use case utilizes audio and other low data rate sources with time constants in the order of hours, experimentally deployed in the Gaasperdammertunnel (Nl). Core characteristics are low information rate, single detection location, long term trend detection, low security/privacy concerns, fixed reference frame, multiple disjoint sensors.
VCD 5.2: Human Activity Inference in Office/Home/Industrial Facilities demonstrator (lead: SIG, ALMENDE, TUE, NXP-NL) – aims to extend state-of-the-art by allowing building management to automatically support its occupants with environmental adaptations for atmosphere setting and working context support through among other optimised light schedules, using embedded AI on heterogenous HW/SW platforms. The building would also support optimized cleaning, energy usage (sustainability), and maintenance as a side benefit. This use case requires inferring human behaviour, context, comfort, and intention, being notoriously difficult and certainly when doing it in a way that maintains privacy.
VCD 5.3: Shared Human and Robot Operating Spaces demonstrator (lead: TUE, NXP-NL, TECHNO) – aims to move beyond state-of-the-art, targeting mobile robots and humans working together collaboratively in a shared space, based on a high degree of flexibility, adaptability, and safety in the robot fleet. It further requires the robots to understand and anticipate the activities, motions, and intentions of humans. This use case is characterized by on-platform processing of multiple high bandwidth sensor streams obtained within a moving reference frame.