The cognitive IoT and eldercare

Description

David Martin explores cognitive function in conjunction with edge computing and IoT sensors and actuators for eldercare scenarios—specifically the identification of individuals, daily activity monitoring, and aberration detection performed on-premises using HomeAssistant, the Intu open source project, and IBM’s Watson cognitive services.

The Intu open source project, initiated at IBM, provides a set of service gateways to IBM Watson’s portfolio of AAS cognitive offerings, including conversation and visual recognition. David demonstrates Watson integration with home automation platform HomeAssistant using Intu, detailing the required architecture for HomeAssistant and Intu to interoperate and bring Watson services into the HomeAssistant ecosystem.

Sensing and responding at the edge requires a full stack of capabilities, both in the cloud and at the edge, as well as a full lifecycle for methods and apparatus used to take actions in response to signals detected. This full stack leverages the HomeAssistant open source project for integration with various IoT sensors and actuators, as well as open standard interapplication communications (e.g., REST and MQTT). High-availability capabilities include integration with commercial applications (e.g., Apple, Amazon, and Google) and other open source and DIY projects.

With HomeAssistant as a foundation for the embodiment of a cognitive agent, the capabilities of cognitive applications (e.g., AgeAtHome) could be generalized for broader consumption. Notably, this includes abstraction of the visual recognition capabilities of Watson as a “cognitive camera” and expresses the requisite input and output in the context of HomeAssistant. Leveraging the packaging of HomeAssistant as Linux containers with orchestration through a supervisor, Hass.io, the cognitive camera is packaged as an add-on that exposes new capabilities for HomeAssistant configuration and use in presence detection and activity monitoring.