Part C – Amazing Escape solved

$35.00

Category: You will receive a download link of the .ZIP file upon Payment

Description

5/5 - (1 vote)

Overview
After seeing your review of their metro simulation design, Shoddy Software Development Company were
outraged and sought revenge. They organised to have you kidnapped, and abandoned in a dangerous location,
with the expectation that you would not survive your attempt to return to civilisation.
Awaking in a strange vehicle in an unfamiliar location surrounded by dangerous traps, you quickly realised
that attempting to drive out manually would soon prove fatal. However, Shoddy did not account for your
software design and development skills. You quickly connected your laptop (conveniently left with you) to
the vehicle, coupled it with the vehicles sensors and actuators, and integrated your tailor-designed vehicle
auto-drive system. In no time at all (well, before the due date at least) you were on your way, the vehicle
automatically navigating its way out of the maze, as quickly as possible*, while avoiding+ the traps.
The Maze
The maze you find yourself in is a ‘perfect maze’ (no loops) constructed in the form of a simple grid of tiles
consisting only of:
• Roads
• Walls
• Start (one only)
• Exit (one only)
• Traps
As your would expect in a maze, some roads will lead to dead-ends. As the roads will not all be the same
width, you’ll need to be on the lookout for narrow dead-ends where it is hard to get back out.
The Vehicle
The car you find yourself in includes a range of sensors for detecting obvious properties such as the car’s speed
and direction, and actuators for accelerating, braking and turning. It has a maximum speed (which is slower
in reverse) and a limited turning rate. The car also includes a sensor that can detect/see three tiles in all
directions (that’s right, it can see through walls). The car can be damaged by events such as running into
walls, or through traps (see below). It has limited health; if the car’s health hits zero, Shoddy wins and you
lose. Similarly, if the car is stuck in a trap and can’t continue, you lose. If however, your software takes the
car from the start to the exit, you win and live on to write more damning software design reviews.
*The time it takes the car to reach the exit and the final health of the car will be combined to provide a score
for your amazing escape.
The Traps
There are different types of maze traps which have different effects on your vehicle; the effect continues as
long as the vehicle is in the trap:
1
• Lava: damages your vehicle.
• Mud: slows your vehicle down.
• Grass: causes your vehicle to slide (no change in direction possible).
A lava trap may destroy your vehicle and a mud trap may stop and hold your vehicle, either way resulting in
failure. A grass trap is not a problem in its own right, but where you end up when you leave the grass trap
may be a problem.
+ Some traps will be avoidable (i.e. can be by-passed), but others will have to be traversed with some
consideration as to the resulting damage and to where the car ends up.
The Task
Your task is to design, implement, integrate and test a car autocontroller that is able to successfully traverse
the maze and its traps.
It must also be able to choose between and apply three different approaches to handling a dead-end:
1. U-turn
2. 3-point turn
3. Reverse out
A key element is that your design should be extensible to allow for it to deal with other trap types and
additional or faster traversal strategies. You will not be assessed++ on how sophisticated or fast-traversing
your algorithms are, beyond whether they work to get the car to the exit (++other than for the bonus mark
for the top 5% based on total score).
The package you will start with contains:
• The simulation of the world and the car
• The car controller interface
• A car manual controller that you can use to drive around a maze
• A working but basic car autocontroller that can navigate a maze but ignores traps and can only u-turn
in dead-ends
• Sample mazes (created using the maze editor Tiled)
If, at any point, you have any questions about how the simulation should operate or the interface specifications
are not clear enough to you, it is your responsibility to seek clarification from your tutor, or via the LMS
forum. Please endeavour to do this earlier rather than later so that you are not held up in completing the
project in the final stages.
To achieve your task, you will be completing two submissions. The first will be your design. The second will
be your implementation. The detailed task break down for these submissions follows.
System Design
You have been provided with a design class diagram for the interface to the car controller and other simulation
elements on which it depends. Your first task is to understand the provided interface and specify your own
design for a car autocontroller. As always, you should carefully consider the responsibilities you are assigning
to your classes.
This design should consider all relevant software design principles used in the subject thus far. You are
required to provide formal software documentation in the form of Static and Behavioural Models. Note that
you are free to use UML frames to help manage the complexity of any of the design diagrams.
You may build on the existing autocontroller design or start afresh; either way, you will need to justify the
end result (see below).
2
Static Models
You must provide a Design Class Diagram for your implementation of the car autocontroller subsystem,
including the interface. This design diagram must include all classes required to implement your design
including all associations, instance attributes, and methods. It should not include elements of the simulation,
other than your autocontroller and the interface.
Behavioural Models
In order to fully specify your software, you must specify the behaviour along with the static components. To
do this, you must produce sequence diagrams and a communication diagram showing how elements of your
subsystem interact. For your sequence diagrams, you must present the following two use-cases.
Controller Functionality
Detecting and progressing from a dead-end
Detecting and avoiding/traversing a trap
Your sequence diagrams must be low level sequence diagrams (i.e. show methods and arguments) which are
consistent with your static models. Further to this, you must provide a communication diagram that shows
all communication between your components within your subsystem.
Design Rationale
Finally, you must provide a design rationale, which details the choices made when designing your autocontroller
and, most critically, why you made those decisions. You may want to apply GRASP patterns, or any other
techniques that you have learnt in the subject, to explain your reasoning. Keep your design rationale succinct;
you must keep your entire rational to between 1000 and 2000 words
Design Submission
You will submit a final design which includes all elements as specified in the submission checklist.
Implementation Submission
Your final task is to provide a working implementation of your subsystem. You must implement the interfaces
provided within the simulation package. This implementation must be consistent with your submitted final
design; if this is not possible due to a flaw in your final design, your implementation may vary from the design
to correct this flaw and this inconsistency with your design must be clearly noted in comments in your source
code. Please also note that the simulation and the car controller interface must not be modified in your
submission (though you may wish to add trace to it or make other similar changes to support your testing).
You should ensure that your system provided is well commented and follows good Object Oriented principles.
Gradle
Your first implementation task should be to import the project, and build and run the application, to ensure
that you have the existing simulation running without issue. The provided simulation makes use of Gradle
as a build and dependency management tool. Instructions for installing the Gradle plugin in Eclipse were
provided for Project Part B; if you have already completed the installation of the gradle plugin for Part B,
you will not need to do it again.
3
Version Control
It is strongly recommended that you use Git or a similar system for version control. Using a
version control system makes it much easier to work as a team on a complex project. The Melbourne School of Engineering runs servers for vanilla Git (http://git.eng.unimelb.edu.au/) and for
bitbucket (https://bitbucket.cis.unimelb.edu.au:8445/), and there are cloud-based servers freely available
(https://bitbucket.org/).
If you do use version control, please ensure you have set your repository to private so that other students in
the subject cannot find and copy your work.
Building and Running Your Program
Your program must be importable using the same setup we have provided you to run the project. We will be
testing your application using the desktop environment.
Submission Checklist
This checklist provides a comprehensive list of all items required for submission for the each part of this
project. Please ensure you have reviewed your submissions for completeness against this list.
Design
1. Ensure you have added your group number to all documents
2. Create a ZIP file including the following design elements for your Car AutoController:
a. 1 Design Class Diagram
b. 2 Sequence Diagrams
c. 1 Communication Diagram
d. 1 Design Rationale (1000-2000 words)
Implementation
1. Ensure you have added your group number to all documents
2. Zip your package folder including all required src files.
Marking Criteria
This project will account for 15 marks out of the total 100 available for this subject. These will be broken
down as follows:
Final Design
The Part I Design Submission counts for 10 of the 15 marks available for this project, broken down as follows:
Criterion Mark
Class Diagram 2 Marks
2 Sequence Diagrams 4 Marks
Communication Diagram 1 Mark
Design Rationale 3 Marks
4
We also reserve the right to deduct marks for incorrect UML syntax and inconsistencies within your diagrams.
Implementation
The Implementation Submission counts for 5 of the 15 marks available for this project, broken down as follows:
Criterion Mark
Functional Correctness 2 Marks
Design Compliance and Code Quality 3 Mark
We also reserve the right to award or deduct marks for clever or poor implementations on a case by case basis
outside of the prescribed marking scheme.
Further, we expect to see good variable names, well commented functions, inline comments for complicated
code. We also expect good object oriented design principles and functional decomposition.
If we find any significant issues with code quality we may deduct further marks.
On Plagiarism
We take plagiarism very seriously in this subject. You are not permitted to submit the work of
others under your own name. This is a group project. More information can be found here:
(https://academichonesty.unimelb.edu.au/advice.html).
Submission
All document submissions must be made to the LMS using the provided links on the project page. Instructions
for submitting the implementation will be provided seperately. Your document submissions for each part
must include all items as required in the submission checklist. Your document submissions (whether they are
design documents or reports) must be pdf documents. Documents in any format other than PDF will not
be considered during marking.
You must submit one zip file for each submission containing all required items. You must not use any
compression format other than zip for your submission. Other archive formats will not be considered for
marking.
Every item you submit must contain your LMS Group number for identification purposes.
If you have any questions at all about submission, please contact your tutor before the submission due date.
Only one member from your group should submit your project.
Submission Date
• Design is due at 11:59 p.m. on the 23rd of May..
• Implementation is due at 11:59 p.m. on the 28th of May..
Any late submissions will incur a 1 mark penalty per day unless you have supporting documents. If you have
any issues with submission, please email William at tiow@unimelb.edu.au, before the submission date.
5