Week 6: Paper reviews, duckiebot simulation, continuation of eDx course, science center symposium

07/15/2024

Over the weekend, I read a paper – Resolution Optimal Motion Planning for Medical Needle Steering from Insertion Surfaces – and had to do a peer review for it. Today we discussed it with Diane and it was pretty great. The paper introduced a new algorithm Backward RCS* which inverts the start pose and target to find the most optimal path from the target to multiple insertion points and not just one. A simple example of that would be thinking about how there are specific points for which we can draw blood from the vein from the outside. It can also be done by finding the most potent vein inside and then planning its path to the skin.

The paper review took more time than expected but it was good to understand what makes a research paper a good one and knowing when it should be accepted or not.

07/16/2024

The first target today was to use the physical map configuration in Heather’s lab to go through the next step in the DuckieTown manual – Lane Following(LN) – which basically has to do with putting the duckiebot in a lane and allowing it to move around the lane having the knowledge of staying in the lane because the camera captures its path ahead. It is supposed to detect the white line and stay within that. Skyduck was successful in the lane following process. I would say that wheel calibrations really help in this case because if it is not set right, the duckiebot will not move in the right direction or might steer off the mat.

The second target of today was to continue with the edX course(self-driving cars with duckietown) and go through chapter 3 – towards autonomy – to complete the Braitenberg Vehicles exercise. This involved learning about image manipulation and image filtering. I ran into a problem with cloning the GitHub in the command line using the way I was instructed to during the exercise and later found out that it was a problem with the DuckieTown GitHub. Instead of running git clone -b mooc2022 git@github.com:Okailey/duckietown-lx in the command line which did not work, I used the old fashioned way to clone the repository by copying the https link and by running git clone https://github.com/Okailey/duckietown-lx.git and this worked well.

07/17/2024

Continuing with the edX course(braitenberg activity), I completed the first two activities and moved on to the third activity and I ran into some problems with that and it turns out it was a general error for everyone trying to complete that activity so we contacted people from duckietown and alerted them of the issue so they said they will get back to us!

Instead of our usual weekly departmental meetings where we give updates, we went to batch to grab some ice cream and talk about the poster presentation and how we were going to go about it. Heading back to the lab after an hour, we spoke about what our plan will be for tomorrow and how we were going to navigate the bots and answering questions anyone may have.

07/18/2024

We decided on having a little slideshow to act as our “poster” which turned out pretty great. We set up around 11 and got going around 12 with the talks. We had leafduck(Yu Wati’s duckiebot) and skyduck running in the physical setup in two different lanes while horizon(Farzana’s duckiebot) was given to anyone who came by to talk to us to have a go at navigating them while the other two bots went round without crashing into them. Overall, the poster session was pretty great and we got to practice how to give detailed information regarding our research to others.

After the session, we packed up and returned to our lab to set everything down and charge our bots and laptops. I decided to keep working on the braitenberg exercise where I started with exercise 3 but unfortunately ran into a problem – when I ran a command on the command line and It produced a link which would leave me moving forward with the activity, I kept getting page not found error and tried running the command multiple times using different browsers and WiFi. Nothing seemed to work. I will give it a try once more tomorrow and if it does not work, figure out the reason behind it.

07/19/2024

For today, the target was to read the second research paper – A Complete Algorithm for a Moving Target – Traveling Salesman Problem with Obstacles, complete my peer review of it and go back to the braitenberg exercise . Reading the paper took a bit more time than I anticipated but I was able to flesh out the important bits which made it a bit easier to understand but I unfortunately was not able to get to the braitenberg exercise so that will be for next week.