Experiential Design / Creative Experience Design Project Prototype
5.10.2023 - 7.11.2023 (Week 6 - Week 11)
Takuto Hozumi / 0354047 / Bachelor Of Design (Hons) In Creative Media
Experiential Design
Creative Experience Design Project Prototype
INSTRUCTION
This project is a continuous project from Experiential Design Project Proposal.
To revisit the concept and design, conduct testing to determine if the project
is feasible, and create a prototype.
UI Design Brush-up
Fig.1: First Design in Proposal
The prototype was re-created in Figma to prepare the actual screen design.
Fig.2: Revised Interface Design
Fig.3: First Page
Fig.4: Menu
Fig.5: Car Setting Camera
Fig.6: Confirmation
Fig.7: Saved the Location
Fig.8: Navigation (Target Far Away)
Fig.9: When Closer (Target in Screen)
Fig.10: When You Arrived At the Car
Technical Attempt of Geospatial in Unity
Geospatial is a plug-in that reads 3D models from Googlemap.
Fig.11: Implementation Guide
I followed the installation guide above.
- Introducing Android build support
- AR foundation, session setting
- API acquisition and configuration
- Introducing AR core
This is roughly what I should do.
Fig.12: The Beginning of Hell
During the installation process, Unity crashed. And since all files became
unstartable and almost affected the progress of the other module (Games
Development) assignment that was using Unity, I reset everything and
reinstalled it.
While installing android build support again, an unexplained error occurred
and one file could not complete the installation. I tried to find a way to
remove and reinstall this one, but could not find it, so I hope it will not
affect me later.
Fig.13: Plugin Install 1 AR Foundation
Fig.14: Plugin Install 2 AR Core
Fig.15: Plugin Install 3 XR
Fig.16: Setting
Fig.16: XR Error
I have come this far and encountered a situation where an item that should
appear on the XR list is not present. I did not know the cause or solution,
nor did I know what was affecting it, until the very end. I could not have
anticipated before I started that simply following the instructions in the
introduction guide would be so difficult.
I had already spent an enormous amount of time working toward the first
plan, but I had no choice but to start thinking about changing the plan,
because I could not allow the final outcome to not be completed. I then
spent two full days over the weekend trying to improve the problem areas,
but no success at all.
For these reasons, the project was determined to be infeasible. The next
section will redefine a new concept and explain the process of creating
another application.
Re-define New App Concept & Process
I get back into the swing of things and come up with a new plan. After
receiving advice from the instructor, I began to think about an
application to check the condition of car parts.
Fig.17: Technical Reference
Since the technology actually exists and the implementation instructions are
concrete, it would not be as unpleasant as Geospatial's.
Since I had three days to submit the assignment, instead of coming up with a
visual design from scratch, created a modified version of the abandoned plan
to streamline the design process.
.png)
Fig.18: Start Page
Fig.19: Main Menu
Fig.20: Scanning
Fig.21: Status Windows
The structure of the application was further simplified, with minimal
functionality.
The concept is that the gauge is full (green) immediately after the engine,
coolant, oil, etc. are replaced, but as time passes, the gauge decreases and
turns from yellow to red. The timing of replacement/maintenance can be
determined.
The application begins by scanning the hood of a car. Since the goal is to
create a prototype, there is no back-end in this case, but I believe that an
AI-equipped car or other vehicle would qualify for actual commercialization.
Technical Attempt with Vuforia ModelTarget
First, use Polycam to obtain a 3D model.
Fig.22: Polycam App
Fig.22: Polycam Register
I attempted to test the scan using an object at home.
Fig.23: Scanning
Fig.24: Completed 3D Data
Perhaps the color caused the tile joint material to be judged as an
object. The excess can be trimmed off later. Anyway, now that we know it
can scan objects, I go with scan a car.
Fig.25: Hood Open
Fig.26: Scan Done
The internal structure of Hood is very complex. Therefore, I recorded it
from various angles to eliminate as many invisible parts as
possible.
Fig.27: Data Rotation
Fig.28: Data Trimming
Fig.29: Completed Data
From here, the completed 3D model is set to be recognized as the target.
Fig.30: Model Target Generator
Model Target Generator was installed.
Fig.31: Target Setting
Fig.32: Exporting
The model was exported as a target after setting the size of the object, the
angle at which it is to be detected, and its color. Although the model was
somewhat complex there, it took about an hour to complete.
Back to work in Unity.
Fig.33: Import the Package to Unity File
Fig.34: ARCamera Setting
ARCamera, ModelTarget was added, and the sprites to be placed in its lower
hierarchy were set.
Fig.35: Library
Fig.36: Library
I downloaded additional libraries to increase the customizability of
textures, etc. However, it is too difficult to use and has not yet been
applied.
It is said to be able to add effects such as covering an object.
Fig.37: Testing
The window design used images exported from the Figma prototype.
WALK-THROUGH
Fig.38: Figma App Design Data
Fig.39: Figma New Prototype
Fig.40: Technical Attempt (ModelTarget)
Fig.41: Video Presentation (8/11/2023)
Fig.42: New Design (Figma Board)
REFLECTIONS
In this task, I started by looking into the feasibility of my previously proposed plan. I had planned a parking record & navigation app with Geospatial, but despite spending an enormous amount of time on it, I could not even get Geospatial into Unity. I was extremely seething and at the same time, I was so furious from the bottom of my heart at my own inability and the software not working the way I wanted it to that I could not sleep because of my frustration. I almost punched my laptop so many times. Unlike design software, which is intuitive and easy to work with, actually making it work programmatically is a very complex process, requiring not only C# programming but also learning about Unity's functionality and linking it with external plug-ins such as Vuforia and Geospatial. And when I ran into problems that I couldn't solve, rarely were I able to solve them, neither Googling nor ChatGPI. Maybe my search and questioning methods were wrong. But I thought I lacked the prerequisite knowledge to begin with. I decided that it would be impossible to complete the first plan in the remaining three weeks, and after seeking advice from the instructor, I unfortunately came to the decision to change the plan. The new plan is not really going well, but I have managed to make some progress on my own. Even while working on the second plan, which seems easier than the first one, I am having trouble like creating targets for the 3D models but not being able to export them, hiding the 3D models, and spending 3 hours trying to stop the demo video eternally playing if I press play button in Unity. I was reminded of my temperamental nature through the task. I will reflect on my temper tantrums and try to solve problems little by little, researching what I don't know one by one when I am faced with any problem.
Comments
Post a Comment