08 Apr

Bryan Bednarski - April 8, 2021 This repo serves as a jumping-off point for system integration of semantic instance segmentation and multi-object tracking in real-time applications. The initial goal of this work was the explore possible optimizations in semantic segmentation through tried-and-true...

07 Apr

In our previous paper, we investigate the consensus problem with von Mises distribution. We want to extend our analysis to more general distributions, in particular the von Mises-Fisher (vMF) distribution over hyperspheres. In fact, vMF distributions are used in some machine learning algorithms. I...

06 Apr

There are three steps I will take in order to implement this design tool: Step 1: Verify I operate correctly in FEA (Solidworks Simulation). In order to do this, I will use some existing numerical solution and/or analytical solutions to find out critical buckling load. Then I will apply this criti...

31 Mar

Vertical control on the newly design balloon: We will start testing horizontal control shortly.

31 Mar

Old custom balloon: Newly design balloon: Compared with previous balloon, the newly design balloon has better shape for our capture mechanism. And the leakage situation is also better after we imporve and iterate our manufacturing process.

31 Mar

After testing our previous active caputure mechanism, we found that one is pretty unstable, so this week, we tried another passive capture mechanism. And it worked nice. During the test, we did find some bugs that we need to fix. We'll try to fix that this week.

31 Mar

We tried one of our designs on the fish balloon. Tested vertical and horizonal control under Opti-track system:

29 Mar

As mentioned in the last blog post, there are two steps in solving buckling problems The first step is to calculate plane stress in the system. This step converts external load (in-plane load) into stress distribution in the plate, so as to be used later in the second step (buclking analysis) T...

25 Mar

We've tested the different thrust using Crazyflie blimp setup. The 80% thrust and 100% thrust: It is noticed that 100% thrust will generate more swing motion on the spherical balloon which doesn't have the stable mechanism (fin for example)