Mechanical 

Adithya Arun:

This weeks focus was still on the suspension sub assembly. We have completed the design for the suspension, track, wheels and side panels which cover the sub system. This design allows for a modular system where the entire sub assembly can be detached from the main body for adjustments or repairs. This process was slightly time consuming as we had to consider how we can make the mold using laser cutting and 3D printing. Corners need to be 3D printed whereas flat surfaces can be laser cut. However, we now have a version that can be produced. In addition we 3D printed a second iteration to test the springs and even with 2mm clearance the printed model was too large for the spring. Even more clearance is needed to get the spring around the model. This week we intend to print out all the necessary parts and assemble the suspension physically. We are also using the side panels and suspension as the reference point for the main body. 

Jonas

 We Finished up the model of the tower which is supposed to contain the different components used in the “shooting” part of the tank. The two bottom parts should be no problem to produce, but the top one might undergo some changes because of its design. The chamfered sides might prove too difficult to 3D-print, and if that is the case then we will either remove them or create another top piece to contain the components.

As far as the cannon goes, we are not yet sure about what the tank is going to fire. An electric softgun was the initial plan, but we might run into some issues with spacing for everything else inside the tower. If the softgun part doesn’t work out, we will simply use a laser pointer to visualize the firing mechanism.

Computer

Rahmat:

In this week I have worked with installation of OpenCV. I have followed some tutorials[1] on the internet how to install this open source on the raspberry pi, but I faced many problems under installation of OpenCV and some of the issue was that “wheels” package didn’t support the “Raspbian Buster”. The main problem was that it took too long to get this open source installed and the other issue I come over was like it didn’t support the current CPU I had in my Raspberry Pi. But in the end, I solved all this problem by switching from “Raspbian Buster” to “Raspbian Stretch”.  I also worked with motion detection on video and on camera. The OpenCV has some inbuilt functions that allows us to use to check the absolute difference between the first and the second image, then we can find the contours to draw it around on the moving object, but to get a better result we need to run iterate over all the contours that we found, then we can choose to drawing shapes around the moving object. The example below shows the result of motion detection with Raspberry Pi. 

reference : [1]https://www.pyimagesearch.com/2018/09/26/install-opencv-4-on-your-raspberry-pi/

Jens 

Worked with tkinter and learned how to use it well enough to create a GUI of buttons where the user could press any one of them and the servos would adjust the angle to match the value shown on the button. 

Kirisan:

Created a script/program in C (arduino) to behave like an SONAR sensor. Components are the same as we use in arduino (ultrasound sensor, jumper cables, breadboard, arduino mega ect.). But to get the screen image of what the sensor is detecting, I had to create another script/program (which is called “Processing”). This SONAR sensor is around 50% complete. This sensor will be an extra material/part of the whole tank, only if we see the need of it. This part will be under C requirements, which means this is not our first priority.

Second part is the openCV where I have been working with Rahmat. This week I have started to create the same openCV which we have been working with till now, on a another raspberry pi. The reason behind this is to have a backup solution in case if the original one is not working, or as we have experienced that one package is missing. So, instead of trying out different type of rasbians, I chose “raspbian sketch” since we found out earlier that the CPU of the Raspberry pi we were working on couldn’t support the the installation of open source. 

Third part this week, I created a script/program to control devices via bluetooth on the phone. Since our prototype isn’t ready, I tested it with simple LED diode bulbs to check if every button on the smartphone bluetooth app gives signal to the arduino, and if the arduino reacts to the signal it received. Pic: https://imgur.com/p4MKiAl