Team : ESD
Name: Ian Murphy
Section: Pi Communication
For this sprint I was tasked with figuring out how to make 2 raspberry pis communicate with each other.
This sprint went way better personally than the last one. The task itself was quite challenging in that there are a few options to choose from to get the pis communicating. I got that task completed but wasn't really happy with the outcome as it was incredibly slow. In this sprint I was more comfortable with everyone as a good span of time had passed and I found it a lot easier to communicate with the lads on my team as well as the other one.
TeamworkTeamwork was a big improvement on the last sprint. The team as a whole communicated a lot better than what it had in the first sprint. A number of the lads would always be around to give a hand and I always tried to be helpful to other team members if they needed the help. One member of the team who gave plenty of time was Kamil, not only was he helpful with any question I had in regards to his threading system but was also willing to put in time with developing a number of functions in his code to help make the sending and receiving a lot easier for me. Moving on to the next sprint should be enjoyable and the team working like a well oiled machine is a big contributor to that.
Project management was a bit messy at the start of the sprint with deciding how the best way to approach this sprint would. It was fairly quickly decided however that the class would be split in two, one team would take care of the mechanical side of the project and the other would focus on the software side, I chose the software side as I have more of an interest in it. After the teams were decided we split up and had a discussion about what tasks there were and who would be assigned each task. This worked really well and we all got in with it relatively quickly.
There was some trouble when it came to combining everyone's code at the end of the sprint so there should have been a date set well in advance so that people had a proper chance to try integrate everything instead of panicking on the day it was all meant to go together, a small mistake however in comparison to the rest of the positive experience I had in this area.
I gave a fairly detailed technical explanation of what I was doing in this sprint here: http://lampbotics.blogspot.ie/2016/03/pi-communication.html . As is stated at the end of this post using the Wi-Fi dongles and then sockets to communicate between the pis did not perform as expected, instead of real-time communication there was a ridiculous delay between one pi send data and the other receiving it. For the next sprint I think using the pi's built in serial communication would be a possible solution for getting over the large delay with the Wi-Fi option pursued in this sprint.
That's the end of my rambling. To sum it up, much better all round than the first sprint and the next one looks like it'll be the same.