SOLVED: CASE STUDY FALL DETECTION, ASSISTIVE TECHNOLOGY

Assessment Task

You are given a brief case study of an application domain, and you are asked to analyse it and produce a number of deliverables for the final requirements and system design documents.

You need to choose to model only a particular viewpoint from the case study and you are required to create two diagrams from each kind: one higher-level diagram for the whole system and one lower-level diagram that corresponds to the viewpoint that you have chosen.

Deliverables and Assessment Criteria / Marking Scheme

Part 1: Requirements identification and modelling

Produce two SysML Requirements diagrams for the given case study:

The models will be marked against the following criteria:

·         Completeness as regards to the chosen viewpoint (conceptual).

·         Appropriate separation of the problem into constituent parts (conceptual).

·         Appropriate separation of the problem and solution (conceptual).

·         Appropriate level of abstraction (conceptual).

·         Correct use of notation, e.g., actors, requirements blocks (structural).

It is important to mention that the models’ assessment criteria contain two parts:

-       Structural – SysML language related aspects: this part assesses the understanding and usage of the modelling formalism.

-       Conceptual – architecture related aspects: this part assesses the conceptual value of your model and requires careful analysis and research into available and proposed technologies.

You are fundamentally assessed on the conceptual – architecture modelling of smart systems and very small part on the technicalities of the modelling language. The use of SysML modelling language is to enable a practical approach that will enhance your future employability skills.

Additional Note for breakdown of marks for diagrams:

·         (x2) SysML Requirements Diagrams (20 marks)

(Total marks for Part 1: 20 marks)

Part 2: Design and propose a system solution for this case study

Propose a technological solution to the case study by:

·         Researching existing assistive technologies (30 marks)

·         Designing, creating and explaining two (x2) SysML block diagrams for structure (10 marks)

·         Designing, creating and explaining one (x1) SysML Sequence diagram for behaviour (10 marks)

·         Designing, creating and explaining one (x1) SysML Activity diagram for behaviour (10 marks)

·         Smart element identification: you need to identify parts of the system from all diagrams (block, sequence and activity) that make it smart.  (10 marks) 

Note that you don’t have to implement your proposed system. You need to perform appropriate research and justify your proposal with references and corresponding documentation.

An appropriate and well-justified proposed architecture solution should be modelled using both structural and behavioural diagrams. Ideas could be drawn from similar systems and solutions that you will explore through research (first bullet point of Part2 - Researching existing assistive technologies). Even if your proposed solution is a completely new idea of your own, you still need to do research and compare it with existing solutions.

It is important not to overlook the smart element in your system design. In this case study as in other similar systems, data are collected and processed at several points offering numerous opportunities for automating parts of the system, adding intelligence by using AI algorithms, etc. Note also that you are not expected to implement automation and AI algorithms but to describe where they would be used in the smart system.  

The marking criteria for the Requirements modelling requested above also apply to the SysML diagrams of Part2. Application of software patterns is highly encouraged.

Your diagrams should be created in the Magic Draw environment and follow model-based design rules for consistency between the diagrams.

Upload a zip file with your Magic Draw project (10 marks).

Also note that the research for existing assistive technologies should not exceed 900 words excluding references and publications.

(Total marks for Part 2: 80 marks)

CASE STUDY – FALL DETECTION, ASSISTIVE TECHNOLOGY

It has been identified that fall detection for the elderly is a major issue. Falls represent a significant threat for the elderly over the age of 65 and the problem is getting worse as the elderly might have to be alone in their homes, care homes or outdoors for a significant period of time. 

Age Concern is a care home in Britain with hundreds of elderly people and they experience 60% of their residents’ deaths due to falls. Recently, there has been a change of management with new strategic directions and one of the improvements has been the funding of an automatic fall detection system. 

The current system that is being used by Age Concern is not based on automatic detection and alarm creation. The residents have to press a telephone button every morning by 11 o’clock to declare that they are fine and taking their medication. This button is connected to a central computing system that monitors the situation for all residents and if it is not pressed by a resident then staff members visit the apartment to identify the reasons. This covers the detection of inactivity status. Also, each resident is being equipped with a pedant that they can press in emergencies and buttons in the apartments and in the lifts are strategically located in easy to access points that can be manually pressed in emergencies. However, all of these mechanisms require some manual action to be triggered by the resident and do not automatically detect problems and raise emergency alarms.

Generally, all the residents in Age Concern, have several medical problems and all the related information needs to be kept in electronic means so that patients, their relatives and the personnel (doctors, nurses, carers) can access/update and consistently maintain information that is related to its resident. Security and privacy issues should be maintained for medical data and each user of the system should have different privileges in using the stored information. For example, the relatives shouldn’t be able to alter the information and shouldn’t be able to read all the information stored. The above should be handled according to the relevant UK data protection act and ethics rules and considerations for medical information (note: the student is expected to extract the corresponding requirements through online research of the corresponding regulations).  

Elderly people that spend time alone represent a large target group for such systems. In case a person tumbles and nobody is around to help independent fault detection and alarming system could automatically call for help by activating an emergency response infrastructure even if the person is unable to trigger an alarm by himself/herself. Also, the person should be able to trigger the alarm himself/herself bypassing the automatic fall detection system.  

Two different types of fall detection systems have been identified: an indoor fall detection system as well as an outdoor one (garden, roads). Age Concern was initially interested only in the indoor fall detection. However, the outdoor and mainly “out of range” fall detection is a new requirement for them that should be taken into account. For example, when the residents are in the garden, mobile technology could assist with setting the alarm in case of fall detection. 

The resulting system should be able to address several different behaviours of elderly people that have been observed through discussions with health care professionals: 

Normal usage: The desired use case is when a person is wearing an emergency wireless transmitter like a wristwatch, a necklace or a pendant permanently during the whole daytime. 

Non-usage: The device being stored away in various drawers from their owners. 

Refused usage: Even if a fall occurs and the person cannot get up on its own, the person might feel reluctant to bother other people and call for help. 

Anxious usage: Anxious persons might wear the device more than required for example during the night leading to false alarms. 

Avoidance behaviour: people decide not to go outdoors at all due to the fear of the device not working outside the house. 

There is a requirement to continuously monitor information regarding the location and the movement of the elderly person that would lead to the immediate detection of abnormal movement in real time.  

The fall detection system will perform real-time and transparent evaluation of the movement situation of the person and will make decisions through related clever algorithms regarding the type of alarm that will be created. For example, the following alarms could be created: 

·         Alarm in case the elderly person will come out of the care home and garden area. 

·         Alarm in case of not movement for example for more than eight hours. 

·         Alarm after detection of sudden acceleration and fall 

The medical professionals (nurses, carers) should have access to the system for observing the level and recharging the batteries of the fall detection system. 

Also, the elderly person should be able to have a choice of communicating directly with corresponding carers, nurses, relatives in addition to the automatic alarms.  

In the case of a false automatic alarm, the elderly person should be able to designate that the alarm was false through an appropriately designed interface.                                                                                            

A major requirement for the future would be that to add an intelligent software agent that would be able to make decisions at the fall detection device and at the information storage level.  

As part of the IoT systems team of your organisation, you have been asked to design a system that provides automatic fall detection in order to improve Age Concern residents’ everyday lives. Your task is to suggest innovative solutions that are cost-effective and could be adopted. You asked to prioritise your solutions according to the level of importance, impact and possibility for adoption.