Week 6: Fine Tuning, Early Recruitment & Hitting a Snag

The best part about having a plan is that you can change it

This week marked the end of week 6 and the halfway point of November. That means I’m only here for 10 more weeks. Yipes! Regardless, this week saw some progress and a little bit of a set back. Overall, positive though.

A lot of time this week was spent working on the navigation code for the slow in and slow out movement. At the end of last week, I thought I had cracked things. But it was a short-lived victory, as it seemed that the changes I did were really limited and the robot could never get up to the maximum speed. I found out that it was some issue with how I was sampling the speed. Once I had that solved, it turned out that I was eating up over 30% of my “slow in”. So, I was apparently calculating the minimum speed incorrectly.

I eventually found out that I needed to adjust the parameters for the Fetch robot too. At the end of the week, it seems that I managed to tune the robot parameters so that the Fetch does indeed move as a slow in and slow out (Sorry, no graph this week). Exactly how my original code worked. Hooray! I’m back to where I started.

Of course, that’s not true. Integrating the code in the navigation stack gives me quite a lot. I guess I should also try it out with the Turtlebot3 to see if it can be used there. That’s a definite side project.

Aside from hacking on code. I fixed up my interview guide and started figuring out how things should be set up in the Robot House. I also visited the Cafe Scientifique. These are popular science lectures lead by a researcher targeted at the local public. The first hour is the lecture and then the second half is a question and answer forum. Being, we are in the UK, the cafe is more of a pub, but I had no problem with that. The lecture this time was held by Dr. Phil Porter. He talked about his adventures up in Svalbard, first drilling a hole to implant a sensor at the base of a glacier and then rappelling down and exploring the inside of a glacier. He was a good story teller and it was a classic example of leaving Norway to find out about research being done in Norway.

The Program for the November Cafe Scientifique in Hatfield

Aside from learning about glaciers and drinking yet another type of cider, the main reason I was there was to drum up interest for my experiment at Robot House. Previous researchers had done well with this. I had asked earlier if it was possible to say a few words, and they graciously gave me a couple of minutes. Unfortunately, since I hadn’t received any word on the ethics application, it was not possible to actually recruit anyone. But the prospect of trying to work with some robots was interesting to several people. So, I promised that I would follow up with more information as soon as I could. I hope their enthusiasm is still there when I actually can recruit them. Being a stranger in a country, you are happy to get any sort of interest that you can find.

Having been buoyed by the interest in the experiment, we were at the robot house to try and get Fetch and my code ready to work. Unfortunately, Fetch decided that now was the time for its hard disk to die. It had valiantly worked, but eventually commands started failing and it could not even boot. Being that there is a service agreement with Fetch, we have contacted them. On the one hand, it’s good that the problem was discovered now before anyone was scheduled for experiments. On the other hand, the robot needs to work so we can run the experiments.

I’ll end this with a quick return to the fountain in Welwyn Garden City. The fountain itself has returned to regular water from its pink water in October. I had secretly hopped for some sort of giant mustache that would grow throughout the month to follow with Movember. But it would probably have been ripped off and that would have been painful and embarrassing for the fountain.