Projects:2020s1-2450 Ambient Intelligence Technology for Assisted Elderly Living
Contents
Ambient Intelligence Technology for Assisted Elderly Living
Introduction
The project will involve the development of a Smart Bed system designed for use by bedridden and elderly patients for monitoring and notification purposes, into a larger connected network. For this Honours project, there will be collaboration with a Mechanical Engineering Honours team that will also use the same smart bed system, but they are still considered separate projects. The team will focus on the development of custom software, making use of commercially-available sensors and components which will need to be integrated into a single system, culminating in a prototype smart bed for proof-of-concept demonstrations.
Project Team
Project Supervisor
- Associate Professor Mathias Baumert
Undergraduate Members
- Martin Gallo Picon: Bachelor of Engineering (Electrical and Electronic) (Biomedical) (Honours)
- Dingbang Lu: Bachelor of Engineering (Electrical and Electronic) (Honours)
Overview
System Description
The project goal is to develop a system capable of providing various patient monitoring functions to assist in elderly living, with comparable accuracy to clinical monitoring methods. Several patient vital signs have been identified as important for health monitoring purposes amongst vulnerable populace, these are heart rate, respiration rate and core body temperature. An additional function was requested by the project sponsor, Uniting Communities, which is a fall detection feature. Fall prevention has been requested because it is one of the more common and damaging problems for bedridden patients in aged care.
Scope: The scope of this project does not include design or implementation of actuators to prevent the falls that are detected, mainly due to time and budget constraints. The included diagram shows a high-level description of project functionality and data connections. Inputs are sent from three different sensor types, and this data is used to complete the expected functionality of the prototype.
Similar IoT health monitoring projects have been done in the past in terms of IoT health monitoring of vital signs, this project being based on several of these examples. The main difference is that the Smart Bed for Assistive Living aims to combine the most accurate monitoring methods across the health monitoring IoT projects that Group 2450 has researched. The result will be a working prototype that incorporates the sensing methods describe previously into a proof-of-concept prototype.
Smart Bed Decomposition
The Smart Bed itself consists of two separate, but connected components. The following section describes their function in high level, then proceeds to elaborate on the implementation methods.
Onboard Systems of the Smart Bed
Data Connections: One of the two main parts of the smart bed are the onboard systems, which consists of the hardware that is included in the bed overlay. These systems include all electronic components which are used to measure respiration rate, which includes the Load Cells and associated circuitry, so the data pertaining to respiration is sent to the Arduino UNO. The onboard computer which is used in all functions of the smart bed (which in this case is the Raspberry Pi 3) is also included in this component, it shares a direct USB connection to the Arduino UNO, so the relevant measurements from the load cells are transferred in real-time, in which case only includes Load Cell 1 and Load Cell 2 Data. The onboard computer calculates the respiration rate based on the data collected by the Arduino UNO, through a Python script. The fall detection system of the bed is also entirely contained within this part of the smart bed (which is based on load cell readings), but in this case, readings from all four load cells are used to determine danger of falling.
Circuitry: The main components in the onboard systems are related to Load Cell measurement acquisition. There are four load cells placed in strategic points around the bed, and these are used for the respiration rate measurement and fall detection features. Each individual load cell must be connected to a linear amplifier so the change in strain gauge resistance (from applied force to a load cell) can be successfully converted to an analogue signal in the Arduino. A few protoboards were used for completing the necessary circuit paths, but will not count as used components in the list.
As such, the following are the associated components in these systems:
- 4 x (50 kg Load Cells)
- 4 x (HX711 Voltage Amplifiers)
- 1 Arduino UNO
- 8 x (10kΩ Resistors)
