7COM1032 - Artificial Life with Robotics - Implementation and Analysis of Flocking Behaviour in Thymio Robots - IT and Computer Science Assessment Answer

Assessment Task:

7COM1032 - Artificial Life with Robotics IT and Computer Science Assessment Answer

Tasks 

1. Write the code for instructing a Thymio robot to move around in such a way that it may lead to the formation of a flock.

 A flock is defined as (a group of) more than two entities that move in a way that results in a distance between any two of them not exceeding a stated limit. This limit is usually specified in terms of body-lengths or the perception distance of the entities (in case of the Thymio robot, they are both ~ 10 cm).

The program should be inspired by NetLogo models of fission-fusion dynamics and flocking behaviour (explored in CW1) and topics covered in the lectures and practicals on robotics and behaviour. Before implementation in the robots, it could be run in the Thymio simulator.

Because of the limited sensor capacities of the Thymio robots, a perfect equivalence between the in-silico models (NetLogo, Thymio simulator) and the robot implementation cannot be obtained, so the code should necessarily be an adaptation and simplification of the simulations. This makes it even more important to compare the observed behaviour of the robots acting in “the real world” with that of the agents (“turtles”) in the artificial NetLogo “world”.

2. Make a list of Behaviour Elements From the NetLogo models, the code written for the robot and watching the robots in pilot runs, decide upon the elements of behaviour (= activities) that are going to be observed. End up with at a list of at least five but not more than ten activities. Also, the team should agree on the criterion for calling a robot “solitary” (i.e. decide on n × body-length or max. perception distance, where n is an integer).

Notes about the choice of behavioural elements The activities should be:
a. simple = reliable and unambiguous to identify and observe.
b. “meaningful”, in the sense of being purely descriptive and not interpretative (i.e. not goal-suspected, such as “moving towards another robot / object”).
c. mutually exclusive and defined such that there are no empty slots between activities (an agent is never doing nothing).
d. A useful annotation to the activities is whether (+) or not (-) they are performed by a “solitary” robot (i.e. that is outside the critical distance to any other robot). Activities marked by a (-) may be lumped into a “solitary state” and as such taken up in the analysis.

3. Set up a 
Finite State diagram and a Subsumption diagram for the robot activities based on the code. 
In these diagrams, the behaviour elements are executed behaviour states.

4. Adapt the program so that the execution of activity triggers the emission of a light signal of a chosen colour. All the to be observed activities should be represented by a light signal (colour).

5. Set up the experiment Run the robot with four similarly-programmed other robots in an arena of defined size for 20 minutes without interruption.
At the start of an experiment put the robots in the centre of the arena and not further than one robot-size (or detection-) distance removed from each other. If a robot during a run drops from the table, quickly put it back and continue recording. 
It is advised to run a pilot experiment in advance to be prepared for possible mishaps and complications.

6. Collect data on the behaviour of the robots Each team member should observe one of the five robots during an experimental run of 20 minutes without interruption so that data are collected for each robot. The robot surveyed by a team member is called a focal robot. The behaviour of the focal robot should be recorded and stored in a specially formatted spreadsheet (“Data Analysis CW3”) as the series of successive light signals of different colours. Such a series represents  the sequence of behavioural activities displayed by the focal robot. Each group member should record the behaviour of one focal robot and the sequence of all five robots should be stored in the “Data Analysis CW3” Spreadsheet for further analysis.                                       

A  group member may use the data he/she collected for his/her focal robot for individual purposes, i.e. as the basis for a more sophisticated statistical analysis (see Deliverables 6. below).

You may want to film the run on your phone and use the images for later analysis.

7. Analyse the behaviour data Cast the behaviour sequence of each of the five focal robots into a transition matrix. A transition matrix contains the counts of how often an activity is followed by each of the other activities. Normalise the counts to transition rates by dividing them by the row totals of the matrix and visualise them in a Markov State Space Diagram. 

This 7COM1032: IT and Computer Science Assessment has been solved by our IT and Computer Science Experts at My Uni Paper. Our Assignment Writing Experts are efficient to provide a fresh solution to this question. We are serving more than 10000+ Students in Australia, UK & US by helping them to score HD in their academics. Our experts are well trained to follow all marking rubrics & referencing style.

Be it a used or new solution, the quality of the work submitted by our assignment experts remains unhampered. You may continue to expect the same or even better quality with the used and new assignment solution files respectively. There’s one thing to be noticed that you could choose one between the two and acquire an HD either way. You could choose a new assignment solution file to get yourself an exclusive, plagiarism (with free Turnitin file), expert quality assignment or order an old solution file that was considered worthy of the highest distinction.