Category: Alex’s Status Reports

For the past 2 weeks I prototyped and finished making the launching subsystem. It was very rewarding to finally be able to shoot a ping pong ball out of the cannon (even though at the beginning it was going at a very high speed). I don’t really see many possible issues with our final bit of integration, something I was worried about at one point was how difficult it would be to move the pipe around, but after talking with Simon I am confident it won’t be an issue.

No major design changes this week. It is however, worth noting that we ended up only needing a thicker tube to fire the ping pong ball, so the problem we had firing the ping pong ball was indeed due to the resistance of the tube.

I also finished the cup detection. I take several 3d coordinate pictures (with the LIDAR), project these onto the plane that is perpendicular to gravity and then convolve a circle onto that image to get out the cup locations. Visually this works very well but I am having trouble extracting the precise cup location and will be finishing that up later tonight.

I’ve worked a bit on collecting data for ball detection and cup detection and am going to continue doing that next week. Next week I also plan on collecting data for the launching subsystem.

I had to learn a lot of stuff about piping and air pressure for this capstone project and I found a lot of useful websites explaining how to do projects that are similar to ours. But ultimately, I found a blog post explaining pressure and flow in terms of voltage and current and that proved to be very intuitive for me. I also learned a lot through trial and error of different computer vision algorithms for cup and ball detection. Something else I learned was how to apply my mathematical skills to write code that necessitates it (the ball dynamics and cup projection were especially enjoyable for me to finish).

No schedule changes this week and we will be focused on testing and integration for the rest of the semester.

This week I worked on finalizing the LIDAR for ball detection and cup detection. I have finished ball detection and I created a neat visualizer for it (from what I can tell the ball detection and trajectory detection work very well, but I am waiting to test it to make any statements about the accuracy). I am also almost done making the cup detection module, but I am struggling to figure out how to make debugging output for it.

I have assembled the launching mechanism and plan to do some math next week to see if we will need larger tubes to launch the ball. If time permits, I plan on testing it, but I am unsure about how much time I will have as I am trying to finish the CV subsystem as we have our interim presentation next next week.

I also helped Simon debug the raspberry pi’s wifi issue and I looked into how to get it working on CMU wifi (as I had to do this with my laptop). To get it working on CMU wifi I believe we just have to register it as a personal device and add the cmu wifi certificate to it.

Next week I plan on finishing CV, helping to get the system integration going and (if time permits) writing software for the launching subsystem. I am currently behind schedule. I also plan on helping mike setup the DDNS.

This week I worked more on the trajectory prediction, with the ping pong balls, I finished the filtering of unrealistic trajectories and am now considering working on getting the trajectory that traveled the most distance. I don’t think that this component will be strictly necessary since filtering out unrealistic trajectories already works quite well. I also need to think about how to test this component of the system as well.

I thought a lot about cup detection this week because it feels harder to do than ball detection. I am going to try mapping the 3d point cloud into a 3d “pixelated” space and then go layer by layer searching for circles. Alternatively, if that doesn’t work I could use a computer vision model to detect the cups based on the rgb vision rather than LIDAR vision, but I’m trying to get LIDAR vision to work properly.

I’m almost done with the first iteration of the launching subsystem, I only need to obtain a 2 liter soda bottle to use as a pressure chamber and hook that up to our current pressure system.

Next week I think I’m just going to focus on trying to get as much as I can done with the computer vision subsystem and I might ask one of my groupmates to look for/buy the pressure container (a 2 liter soda bottle).

This puts me around where I expect to be on the schedule.

Over the last 2 weeks I have spent considerable time working on the ball detection part of the computer vision subsystem. As seen on the design report, we can now pretty reliably detect ping pong balls that are flying through the air. I have also finished the trajectory prediction with the ping pong balls, I still need to filter out unrealistic trajectories, but that shouldn’t be too bad and I am looking forward to testing that component soon.

I have also spent considerable time working on the design document.

Over spring break I was able to retrieve some of the parts for the launching system and I began assembling the launching subsystem with the parts that I had available. I am looking forward to receiving the remaining parts and then I want to finish building the first iteration of the launching mechanism.

This leaves me on track with the schedule.

Next week I plan on finishing construction of the launching mechanism and getting some progress on the computer vision subsystem (I will probably focus on trying to finish the computer vision subsystem).

Time Breakdown:

1. Class time (4 hours)
2. Ordering Parts (2 hours)
3. Design Document (2 hours)
4. LIDAR (4 hours)

This week I was in class for 4 hours watching presentations.

This week I ordered a good portion of the parts I need for the pressure subsystem. Additionally, I’ve looked into some alternative parts that we can buy and am considering buying them just to see if they can be used to improve our base design. These include buying a simpler solenoid valve for the firing mechanism and weighing some options on different PVC pipe radii/thicknesses that I can buy.

This week I also worked on writing up some of the back of the napkin math that was used for our design process to include in our design document. This mainly involves the pressures necessary to launch the ping pong ball with sufficient velocity.

Finally, this week I worked on the LIDAR system. First I verified that it can indeed track a ping pong ball at the distance we require (it can). Then I began working on the software to interface with the LIDAR, I was having some trouble on this front, but I will be finalizing the software for this subsystem over the coming week.

This leaves me behind schedule on ordering parts, but ahead of schedule on getting the LIDAR to work.

Next week I plan on working on ball detection with the LIDAR as well as beginning assembly of the pressure subsystem.

Time breakdown for this week:

1. Class (4 hours)
2. Design (4 hours)
3. Setup RPI (2 hours)
4. Tinker with camera (3 hours)

This week I worked in class on the design for 4 hours, I mainly finalized the design for the pressure subsystem. The main ideas all stayed the same (i.e. we have an electronically controlled proportional valve to decrease the pressure, an air pump to increase the pressure, and a release valve to fire the ping pong ball.

Additionally, I was working to see if we could use an Intel Realsense Depth Camera D455 instead of an Intel Realsense LIDAR Camera L515 as the one that 18500 has was being used by group B1. This did not work, but luckily for us group B1 didn’t end up using the L515, so we can use it now. That unfortunately means that what I was working on with the D455 is not useful anymore, but that is good because I didn’t really feel like we could ultimately get anywhere with it given its lighting requirements.

This week Mike and I also worked on setting up the Raspberry Pi, it was harder than I remember, but it wasn’t too bad. Mike and I were having a lot of trouble getting the internet to work properly. We got the RPi to connect to the internet, but for some reason we couldn’t connect to websites.

We plan on ordering some parts this coming week so I can work on the pressure subsystem. This week I also plan on using the L515 when I receive it and I hope to get some progress on that.

As of right now we are ahead of schedule as I have already tinkered with a camera and stereo range detector to see how well they work for our use case.

Time breakdown for this week:

1. Mandatory in class time (4h)
2. Design Slides (4h)
3. Basic Ball Detection (4h)

This week I worked on design slides for a couple hours with my group on Saturday. We started working on finalizing all the designs and we began making some of the slides for the design review.

On Friday night, I worked on doing basic ball detection. I have a couple videos that I could upload, but wordpress says “Sorry, this file type is not permitted for security reasons.” I found that the ball detection is fairly difficult to do on its own (even when I am assuming that the ball will be the only thing moving in the frame) and have decided to get a LIDAR+Camera to help with this. The LIDAR+Camera combination will allow us to get a depthmap that the camera can use and should greatly simplify the detection code while making it simple to estimate the ball position.