Sean Cleary Introduction Blog

 1.0 About Myself

Hello, my name is Seán Cleary and I am a final year student in SETU studying electronic engineering. In my free time I enjoy baking, cycling, and board games.

2.0 Previous Blog Post 

When looking through the previous blogs, this one in particular really stood out to me. I really like how the information is neatly laid out in this blog. The use of images, video, and additional reference links really adds to this blog and allows for information to be more easily conveyed.

 https://lampbotics.blogspot.com/2016/03/robert-burke-sprint-2-extended-face.html

The information in this blog is split up into smaller, easy to digest sections, and is written in very clear and concise language. This makes it very easy for anyone reading this blog entry to parse what is happening, and follow along with the work being carried out.

The blog flows very clearly, with the different headings being clearly defined. The images present are also clearly numbered, and all of them are captioned. 

3.0 Interesting Videos and Ideas for Binbot

When looking for information on a bin robot or "binbot" previous, I came across this video. In this video, a man modifies his bin to turn it into a "binbot". This binbot is able to navigate to a person upon voice command.

 I thought this was an interesting implementation of a bin robot, as it was able to freely move to wherever it was needed. While our binbot has a much more limited range of movement, it could still be interesting to see what could be done with that, perhaps our binbot could react to voice commands and perform movements based on different commands.

3.1 Binbot and AI 

Analytical AI is already likely to be used in conjunction with this binbot, as in my own project I am using OpenCV with a dataset (which consists of thousands of annotated images) to allow the binbot to identify different types of rubbish. This form of analytical machine vision AI will be integrated into the binbot itself.

 Beyond that a further idea could be to allow the binbot to be able to recognise and categorise types of rubbish that it is seeing for the first time. In this way the binbot would have a sort of machine learning that would allow it to become more accurate the more it is used. This is however likely beyond the scope of this project. 

4.0 State Machines 

Modern state machines first came about in the 1950s, when George Mealy and Edward Moore created papers on the usage of state machines. These two scientists give there names to the concepts of the "Mealy" and "Moore" state machines, which are the two main types of state machine.

• Mealy Machine: Outputs determined by current state and inputs.
• Moore Machine: Outputs are solely determined by the machine's current state.

 

Figure 1: Example of a Mealy Model (Image from Wikimedia Commons)

 

 

Figure 2: Example of a Moore Model (Image from Wikimedia Commons) 

 

Figure 4

 An example of a Mealy and Moore model can be seen above in figures one and two. It can be instantly seen from the figures that in general Moore model state machines have more states than Mealy model machines. These Moore model machines tend to be simpler to design, but requires more hardware than the Mealy machine.

Figure 3: Drawio state machine iteration 1

 

 

Figure 4: Drawio state machine iteration 2

In the first class we used Drawio to modify state machines for a DRS (Drag Reduction System) system. Figures three and four above show two examples of those state machines.

Figure three was the initial iteration, however after realising that it wouldn work correctly, I added in the additional "DRS Warning" state as an intermediary state between ready, on and pause. This allows for an in between state when a vehicle is more than fifty metres away from another vehicle.

While in DRS warning, DRS can still be activated. However after five seconds in warning the system goes into DRS pause, and DRS can no longer be activated.