Chris H Portfolio

Home

An Augmented Reality Training Solution for the University of Winchester

Problem Statement

The University of Winchester's Multimedia Centre provides training and short-term loan access to cameras, lighting, and sound equipment. However...

...a reliance on in-person training for camera equipment has resulted in long wait times due to high staff workloads.

Proposed Solutions

Prototype an augmented-reality mobile application which can:

Be operated by students without the presence of a technician

Enable an asynchronous approach to training undergraduate students

Include accessibility features that promote equity within education

Provide a demonstrable proof-of-concept to secure further project funding

Research Method 1: Qualitative Data Analysis

A spike in training demands, especially with new students, could overwhelm staff.

The University did not collect data on camera training bookings but did provide camera loan data. The number of training bookings could not be expected to correlate loan data, but could still inform assumptions:

The quantity of camera loans increases steeply between September and October, while a similar pattern can also be observed between January and February. These follow the University’s semester calendar, indicating a repetitive, seasonal issue.

As training is an antecedent of loan booking, a spike in training demands, especially with new students in Semester 1, could overwhelm staff.

Equipment availability for training could be compromised if loaned to students who have already received training. This may further reduce training session availability.

This data implies that the solution could help by attenuating the demand for staff training during ’spike’ periods.

Site Structure Diagram of the Sober Sauce Website

Research Method 2: Focus Groups and Journey Mapping

Focus groups highlighted several potential benefits of using a simulation for training.

A series of focus groups were held with technicians and training staff, identifying pain points across the in-person training process, from initial booking to post-training support. Pain points were placed together to produce an affinity diagram and identify improvement opportunities.

Key opportunities identified:

Asynchronicity could provide flexibility for the student. They can undertake training when and where suitable for them.

Training simulation will remove set-up and pack-down time. No need for camera equipment or staffing.

Staff could be relieved of administration duties. Assessment activities could be completed in-app and automatically approved upon successful completion.

There needs to be more scope for safely demonstrating what happens to equipment when things go wrong. An AR solution could simulate this without risking students or equipment.

Pre-Training Administration	Preparing for Training	The Training Session	Assessment within Training	Post-Training Administration	After-Support
Lack of online learning resources	Training preparation is time consuming	Students sometimes passive in sessions	Paper forms present a GDPR concern	Time consuming	No revision or refresher materials available
1:1 training is inefficient with large cohorts	Preparation can be wasted due to non-attendance	1:1 training sessions - difficult to demonstrate some equipment	Paper forms / standardisation not always appropriate for checking learning	Extant filing system for attendance and assessment results is not efficient for checking
Admin for training requests is time consuming	Slow student responses to availability enquiries - equipment becomes unavailable	Students are not always happy to attend in-person	Students can feel pressurised to sign forms to agree training has taken place, despite not feeling the have demonstrated this enough
Timetabled training sessions clash with student lecture timetables		Training sessions are taking valuable resources away from students who need them for projects	Assessment via questions and answers can feel awkward for students
Technicians are usually booked 2 weeks in advance

Research Method 3: Qualitative Data Analysis

Four student personas comprised focus group data, identifying varying engagement, confidence, knowledge, and attitude levels.

Data collected from focus groups with staff and students resulted in four personas most engaged with the training provision. These personas varied in confidence levels, pre-existing camera knowledge, and attitude towards training sessions.

Several persona-informed situations were devised to pre-test technology acceptance. Each of these situations would be considered during the ideation and testing processes:

Student Persona	Image	Student Persona	Image
Jack, the ‘Eager Beaver’: Jack is proactive and meticulous about camera gear, investing heavily in training and expecting the university to match his dedication.		Tom, the ‘Crew Member’: Tom is practical and collaborative, focusing on team needs and seeking clarity through questions and online research, but skips extra training.
Sam, the ‘Steady Learner’: Sam enjoys learning but finds equipment challenging, relies on friends for training support, and hesitates to seek staff assistance, affecting project progress.		Marina, an International Student: Marina is assertive and driven, preferring individual training for hands-on equipment guidance to overcome language hurdles and fully utilise university resources (due to paying higher tuition fees)

Research Method 4: Content Audit

An audit of similar AR applications revealed several design patterns which enhanced their usability and accessibility.

An overlaid UI provided instruction and recognisable cues to access functionality. However, translucent buttons limited vision-based accessibility.

Contextual prompts and visible tap targets demonstrate the system’s status.

Tab bars, menus, and navigation elements sometimes mimic the operating system, increasing the opportunity for recognition and affording user control.’

Drawers of placeable objects allow the item inventories to match the functionality of real-world equivalent drawers, providing the user with behavioural expectations.

Error messages usually provide instructions to help users recover from errors.

Research Method 5: Academic Journals

A 'Freeze Frame' function could increase accessibility for students experiencing motor or cognitive impairments.

A small thematic analysis focused on accessibility functions within AR applications was completed. One notable function was ‘Freeze-frame’, which would permit the user to pause the AR experience so that they could inspect their device’s display from a comfortable position before resuming (Herskovitz et al., 2020). Students experiencing unstable walking, involuntary movements, or cognitive impairments may benefit from this function.

“Aiming cameras non-visually is, in general, known to be a hard problem, and we found that the difficulty is only magnified when the position must be held stable while also interacting with the mobile device” (Herskovitz et al., 2020)

Application Task Flows

Two mandatory task flows were identified that would become the basis for the design and testing process.

Ideation, Storyboarding and Sketching

A range of progressively detailed sketches were developed in parallel with regular consultation with training staff and students.

Wireframing and Interaction Designing

A range of progressively detailed sketches were developed in parallel with regular consultation with training staff and students.

Paper Prototyping and Usability Testing

Test participants can easily navigate using the tab bar, object drawers, and toggle switches.

A paper prototype was created, and an initial round of ‘think-aloud protocol’ usability testing covered both task flows:

Viewing a Camera

Creating and Reviewing a Freeze Frame

Test participants were provided with the following scenario:

”You are a student at the University of Winchester and requested camera training. You have downloaded an augmented reality mobile application on your phone, which the University recommended you attain the required training.
In this usability test, you launch the application for the first time and explore its features. Please complete the introduction, view two cameras, and save your first ‘Freeze Frame’”.

Usability Testing Outcomes:

Misunderstanding of Visual Elements: The ‘target’ is often mistaken for an object, likely due to the limitations of paper prototyping and drawing skills.

Onboarding Process Clarity: Users require repeated onboarding prompts to engage with the ‘Freeze Frame’ feature, indicating a need for more precise instructions on when the walkthrough ends.

Storage vs Saving Confusion: There’s confusion about whether a ‘Freeze Frame’ is saved within the app or to the device’s Camera Roll, suggesting a need for distinct terminology like ‘Save’ versus ‘Export’.

Confirmation of Action: Users are unsure if creating a ‘Freeze Frame’ saves the image, indicating a need for explicit confirmation and terminology clarification in the onboarding process.

Prototype Development and Usability Testing

Once wayfinding mechanics had been designed, tested and remediated, a further prototype was created, which would be used to test user responses to animation and graphics. The University's branding guidelines were adhered to when creating the user interface, and testing was completed using an Adobe XD prototype on an iPhone.

Further usability tests focused on the tester’s experience of basic app functionality and accessibility functions.
Each test revealed where the tester would become confused or behave unintendedly. These issues must be rectified to prevent them from being distributed as a final-release candidate for the mobile application.

During these tests, participants repeatedly attempted to swipe objects to rotate them instead of tapping them. Further research indicated that this had become an expected gestural functionality in both AR and VR experiences, so it was implemented in a later iteration.

Final Design Highlights and Lessons Learned

Functionality Highlights

Users can access on-demand training using virtual assets.

Training requirements can be met when the workload is high and technician or asset availability is low.

Design patterns leverage learnt user behaviours from similar AR applications, enhancing recognition and control.

Usability and Accessibility Highlights

The “Freeze Frame” functionality allows users to pause the AR experience while they make observations in comfort.

Gestural behaviours across other touchscreen systems have been integrated following usability test feedback.

A screenshot of the Sober Sauce redesign in Adobe XD

References

Herskovitz, J., Wu, J., White, S., Pavel, A., Reyes, G., Guo, A. and Bigham, J. (2020). Making Mobile Augmented Reality Applications Accessible. ASSETS ’20: International ACM SIGACCESS Conference on Computers and Accessibility. [online] Available at: https://dl.acm.org/doi/10.1145/3373625.3417006 [Accessed 26 Sep. 2021].

WebAIM (2012). WebAIM: Motor Disabilities – Types of Motor Disabilities. [online] Webaim.org. Available at: https://webaim.org/articles/motor/motordisabilities [Accessed 28 Oct. 2021].