Project Narrative:

<aside>

MicrosoftTeams-video (1).mp4

</aside>

<aside>

The goal of this project was to create an end effector system for the Q-Arm that would be used to handle products in a warehouse. This system needs to be able to handle a large variety of shapes and textures of products reliably, something current warehouses struggle with. The specific scenario provided to us was 6 objects of varying shapes and surface textures, including five 3D printed items and a sponge.

Project timeline: October 27th 2025 - December 3th 2025.

</aside>

Unfortuantely I do not have pictures of the final end effector on the Q-arm. However, many of the pictures below represent the end-effector by itself.

Team’s Work and Personal Contributions

<aside>

My role on the team was Subject Matter Expert. Apart from my role in citing all the references for the short summary, I ensured that all the parts of the Final Report were finished by their respective members and completed the short summary alongside my other SME and team members. For the rest of the project, I made contributions outlined below.

</aside>

<aside>

End-effector Mechanism

• We brainstormed together for Milestone-0. We decided on a few mechanisms that appeared promising and assigned designs for each member to sketch
• I was assigned a “claw” mechanism. This mechanism is very well represented online so I took a fair bit of inspiration from online pictures. I did research on gears vs. the worm gear design, and i believe that the worm gear allows for a more “solid” hold of the object but is a much more complex design.

• Milestone-1, we finalized our own designs and refined them. My sketches seemed sufficient so no changed were made.
• For Milestone-2, we cadded our individual design. I ended up cadding the non-worm gear design as I thought the 3D printed items would be fairly light and a less complex design would be more reliable and easy to maintain. My team presented our ideas to each other and we all came to a consensus that a combination of one members un-moving arm and jaw, and another member’s rack and pinion system, would work well together.
• This was my CAD. The curvature on the claw is meant to create a better grip of various objects as the lower edge is able to catch onto the lips and thin edges better. However, this choice is untested and unresearched.

• For the team milestone, I drew the final design.

</aside>

<aside>

Code

• for Milestone-3, I weas responsible for the authenticate() function. The flowchart was what was supposed to be submitted. This was fairly straightforwards for me as the instructions are very clear, however I did ask a TA to check my flowchart for me. He provided feedback for me to split up the password and username check into two parts: checking that the user ID is in the CSV list of valid user IDs and checking if the password the user inputted matches the password associated with the inputted user ID.
• This is the final flowchart I submitted

• Upon completing my work (coding the authenticate() function for Milestone-4), I helped my team mate complete their function and another member put it together as a team. I think this is where I learned a lot of team work and how to understand other people’s code. The sign_up function was completed individually but was troubleshooted as a team, which I think our team really had to work together to figure out the different changes we made and its effects.
• I also found out at this time that I missed a key part where I had to decrypt the password to authenticate the user. I pretty much rushed to finish this, but I likely would have missed this if I had not been looking at the sign_up() function where I saw the encryption implemented.
• Another member put all the code together and helped me complete my commenting as I was working on other things </aside>

<aside>

Iterating on the design

• My teammate then did all the CAD for the rest of the project.
• The first version, I helped him assemble the the slider and pin. The pin was very thin and the slider was split in half, which needed to be glued together. This was a difficult that due to delicate and small components. The design below is the one that needed gluing.

• Alex (CAD team member) made adjustements to the claws and there was a big miscommunication about the design of this claws and where the angle was supposed to be. Upon resolving this miscommunication, we still had further design problems that impeded our testing with the Q-arm.
• One such problem was the tolerances on the gear that connected to the motor shaft. To resolve this I tried to make smaller and tighter by adding tape, or bigger by filing the hole. This was not the best and we reprinted the gears many times because of this.
• Later on, this is some feedback I provided to my teammate that owned and did the CAD.

He made improvements and showed these videos:

MicrosoftTeams-video (1).mp4

MicrosoftTeams-video.mp4

</aside>

Reflection

<aside>

For this project, my team was able to pick up the sponge and drop it off at the drop-off location with little reliability.

This is a great start for our end effector design, but should ideally be further iterated to pick up all objects reliably.

To the right I have listed a few improvements we could make on the current design that I came up with.

Additionally, on a personal level, I think I could be better with time management for my next project.

</aside>

<aside>

One such change to the end-effector design is to add an angle at which it picks objects up at. This was an issue we discovered late in the project after testing with the Q-Arm.

</aside>

<aside>

Another change would be a hard-coded or physical limit that prevents the gears from disengaging from the rack. This is important to keep the end effector reliable.

</aside>

<aside>

Finally, what would have made assembly more successful (many parts broke), would have been to reduce the complexity of the parts and reduce the overall quantity of parts.

</aside>

Summary

<aside>

Our end effector code worked very well, as shown in this video:

ImportedSegment.785901400.162129.mov

On the other hand, our mechanism had some trouble! We were unable to pick up many objects because the angle of the end effector was difficult to work with and could not go perfectly parallel to the table. As a result, many objects slipped out of our grasp. We were able to unreliable pick up the sponge at the end!

Unfortunately, I also do not have a video of the end effector.

</aside>

Skills

<aside>

3D modelling

This project allowed me to apply my knowledge of CAD to create prototypes of the end effector design. Additionally, it allowed me to learn about the design process, with iteration, testing, and motion transmission. This was the first time I 3D modelled something I designed myself and it taught me a lot about 3D modelling itself and the application of certain tools.

</aside>

Design Proccess

<aside>

</aside>