Changes for page William Reflection
Last modified by William OGrady on 2024/04/09 14:36
From version 12.1
edited by William OGrady
on 2024/04/09 14:36
on 2024/04/09 14:36
Change comment:
There is no comment for this version
To version 5.1
edited by William OGrady
on 2024/04/04 16:12
on 2024/04/04 16:12
Change comment:
There is no comment for this version
Summary
-
Page properties (1 modified, 0 added, 0 removed)
Details
- Page properties
-
- Content
-
... ... @@ -22,27 +22,19 @@ 22 22 23 23 **Week 5** 24 24 25 -More work on the xWiki and started looking at implementation with Diederik. We spent time testing the efficacy of //Robots in de Klas (RidK)//and then investigated alternate ways,like the Social Robotics libraryforpython. We never got it working so decided to hard code a database in //RidK. //I made a rudimentary database in Google Sheets.25 +More work on the xWiki and started looking at implementation with Diederik. We spent time testing the efficacy of //Robots in de Klas //and then investigated alternate ways of 26 26 27 27 28 28 **Week 6** 29 29 30 - This week's biggest decisionwas to abandon the //RidK-//implementation for a custom GPT assistant, usingthevoice function that had been released only a few weeks prior. The decision was made becausealthough this meant separating the prototype intotwo components, the robot and the smartphone, the prototype would be closer to what was envisioned.Hardcodingthe conversations wouldnot demonstrate the natural conversation element in a suitable fashion. Theelement isthe mostimportant in the project. This changemade iteasier touse the database, as the GPT assistant can importexcel files. We planned to have pilot studies done that weekend but this fell through.30 +finalising evaluation plan & database, switching to gpt 31 31 32 32 33 33 **Week 7** 34 34 35 - This week we did the pilot study and the evaluation. My job was documenting the evaluation by taking pictures and recording some videos, mostly for the presentation. I also took time to make every page on the xWiki a bit cleaner and consistent, using the same font size and cutting down paragraphs to avoid excessive repetition. The aim was that reading the xWiki should be no different then reading a regular project report.35 +evaluation 36 36 37 37 38 38 **Week 8** 39 39 40 -Week 8 we had our final presentation, we also planned final xWiki runthrough by creating google document with notes for each page. On Saturday we got together and spent a couple of hours going through each page, completing the xWiki. 41 - 42 - 43 - 44 -**Summary** 45 - 46 -My contributions for this project have been the personas, the problem scenarios, the values and ethical concerns for the HRI. In the second half of the course I worked the implementation, figuring out Robotsindeklas and GPT and some other programs which were never used. I designed the database used for the evaluation. In terms of the xWiki, I took extra effort to make navigation easier. 47 - 48 -Overall, I found the course quite intriguing and fitting for my education. Bear in mind, this was the first course I attended here at TU Delft. I noticed myself behave a little more timidly and be a little more humble when engaging with the course as I was adapting to the material and group dynamic. As the only non-Dutch person in my group, I was welcomed and felt heard at every occasion. That was very important and helpful and something quite invaluable to me, regardless of the work itself. I wish the groups were a little smaller to give each member more substantial work. Most of my day to day work was on the xWiki which to me is not a preferred method of report writing. 40 +Final presentation