{"id":4293,"date":"2020-10-07T12:32:33","date_gmt":"2020-10-07T11:32:33","guid":{"rendered":"http:\/\/dronesonen.usn.no\/?p=4293"},"modified":"2020-10-07T12:32:33","modified_gmt":"2020-10-07T11:32:33","slug":"fifth-sprint-puck-buddies","status":"publish","type":"post","link":"https:\/\/dronesonen.usn.no\/?p=4293","title":{"rendered":"Fifth sprint – Puck Buddies"},"content":{"rendered":"\n

Anders<\/p>\n\n\n\n

This week there has been no work on this project from my side, due to my hospitalization as mentioned in the previous blog post. Anyhow, stay tuned as the others most certainly have something to show off!\u00a0 <\/p>\n\n\n\n

Halvard<\/p>\n\n\n\n

I have mostly worked on producing parts and testing them so that we will have some numbers for our budget of forces\u00a0acting on\u00a0the system.\u00a0 Gaute and I started testing the materials\u00a0we are going to use\u00a0this week.\u00a0\u00a0I have also started\u00a0modelling the subsystem for the magazine, firing\u00a0mechanism\u00a0and the plates for\u00a0mounting\u00a0the electronics.\u00a0 <\/p>\n\n\n\n

\"\"
The subassembly of the\u00a0magazine and the housing for the firing mechanism<\/em> <\/figcaption><\/figure><\/div>\n\n\n\n

This is our proposed firing mechanism and the magazine assembly, the hockey-pucks will be stacked in a\u00a0tube (PVC, etc.) and\u00a0then they will be pushed\u00a0into the spinning wheels\u00a0with a servo. Most of the parts in this sub-assembly are intended to be produced with\u00a0a laser cutter that\u2019s why they have a rather\u00a0simple\u00a0look.\u00a0\u00a0 <\/p>\n\n\n\n

The first series of\u00a0material\u00a0tests\u00a0(ABS, PLA, PET-G, Nylon, etc.)\u00a0 <\/h3>\n\n\n\n

This is where most of my time has went this week. The test parameters are as following;\u00a0each material is being tested 5 times and the value will be the average,\u00a0in some cases I will also list the spread. As well as\u00a0finding\u00a0the\u00a0anisotropic\u00a0properties\u00a0of the\u00a03D printed parts. <\/p>\n\n\n\n

\"\"
<\/figcaption><\/figure><\/div>\n\n\n\n
\"\"
The test parts and the test machine<\/em> <\/figcaption><\/figure><\/div>\n\n\n\n
\"\"
The X1 and X2 are the same material, just different temperatures, treatments, spools, etc.\u00a0\u00a0<\/em> <\/figcaption><\/figure><\/div>\n\n\n\n
\"\"
One of the results from the tests.\u00a0<\/em> <\/figcaption><\/figure><\/div>\n\n\n\n

Her is the formula for the tensile strength for the regular PLA horizontally printed at 210 C @ 25 C:\u00a0<\/p>\n\n\n\n

\"\"<\/figure><\/div>\n\n\n\n

This is right in the expected range, for the one printed vertically we had a \u03c3 = 18.342Mpa. This is what is known as anisotropy. This shows us that we also must take the orientation of the print into consideration during the design-process  <\/p>\n\n\n\n

This research will lead to a more thorough publication next week when we have tested all the materials and treatments for the materials.  <\/p>\n\n\n\n

Gaute<\/p>\n\n\n\n

This week I worked on several things. I made some adjustments to the\u00a0web-cam module\u00a0that I started on last week.\u00a0So now the load from the\u00a0web-cam\u00a0isn\u2019t directly on the step-motor shaft, but rather on\u00a0an axial bearing, this will\u00a0reduce the wear and tear on the step-motor to a minimum.\u00a0\u00a0<\/p>\n\n\n\n

I also made some\u00a0mounting guides for the frame. They\u00a0will make the assembly of the frame easier for us since we won\u2019t\u00a0have to weld anything. Makes it faster to build, and easier to perform any maintenance\u00a0and disassemble.\u00a0\u00a0<\/p>\n\n\n\n

\"\"<\/figure><\/div>\n\n\n\n

But what took up most of the work\u00a0hours this week was to adjust and fit the\u00a0launch-mechanism-assembly to the frame. I took quite some time since the measurements\u00a0wasn\u2019t in the same\u00a0scale\u00a0as the frame-assembly. <\/p>\n\n\n\n

Nicholas<\/p>\n\n\n\n

Electrical<\/h3>\n\n\n\n

I finally sat down an learnt how to make components in Eagle and was therefore able to create the schematic draft for the project. <\/p>\n\n\n\n

\"\"
Schematic draft as there is still some missing connections<\/figcaption><\/figure><\/div>\n\n\n\n

Software<\/h3>\n\n\n\n

So this week I spent\nmost of my time researching and reading up on TensorFlow and machine learning. Therefor\nI don\u2019t have too much to show for this week. I spent a lot of time reading and writing\nexample code seeing the effects of overfitting the model and how changing the\nnumber of layers and the layers data affected the model. Our goal is to use our\nown trained model for the object detection to improve the performance as it\nwould only have to identify 1 class instead of 80 as YoloV3 does. <\/p>\n\n\n\n

One thing I could\nmention is that since I have a Nvidia GPU installed in my computer I could\nleverage CUDA to run model learning on the GPU instead of on the CPU. After\nmany unsuccessful attempts at installing the necessary drivers for CUDA I found\nout that I had to uncheck the \u201cInstall Visual Studio Integration\u201d part. More\ninformation about CUDA can be found here: https:\/\/developer.nvidia.com\/cuda-zone<\/a>  <\/p>\n\n\n\n

The goal for next week\nis to get some work done on the custom machine learning model and see how that\ngoes. I will also probably start to look into running the model on a Raspberry\nPi with TensorFlow Lite.<\/p>\n","protected":false},"excerpt":{"rendered":"

Anders This week there has been no work on this project from my side, due to my hospitalization as mentioned in the previous blog post. Anyhow, stay tuned as the others most certainly have something to show off!\u00a0 Halvard I have mostly worked on producing parts and testing them so that we will have some […]<\/p>\n","protected":false},"author":77,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[1],"tags":[],"class_list":["post-4293","post","type-post","status-publish","format-standard","hentry","category-uncategorized"],"_links":{"self":[{"href":"https:\/\/dronesonen.usn.no\/index.php?rest_route=\/wp\/v2\/posts\/4293","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/dronesonen.usn.no\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/dronesonen.usn.no\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/dronesonen.usn.no\/index.php?rest_route=\/wp\/v2\/users\/77"}],"replies":[{"embeddable":true,"href":"https:\/\/dronesonen.usn.no\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=4293"}],"version-history":[{"count":6,"href":"https:\/\/dronesonen.usn.no\/index.php?rest_route=\/wp\/v2\/posts\/4293\/revisions"}],"predecessor-version":[{"id":4307,"href":"https:\/\/dronesonen.usn.no\/index.php?rest_route=\/wp\/v2\/posts\/4293\/revisions\/4307"}],"wp:attachment":[{"href":"https:\/\/dronesonen.usn.no\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=4293"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/dronesonen.usn.no\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=4293"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/dronesonen.usn.no\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=4293"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}