Department of Computer Science
CS26020 Robotics and Embedded Systems
Supplementary Assignment 2019-20
Final Submission Date: Friday 14th August 2020 at 13:00 BST (1pm).
Laurence Tyler ([email protected]), MP-1.11
This assignment is deemed to be “of equivalent value” to the weekly practical worksheets.
Assessment of this component will be worth 50% of the marks for this module. Please read all of
this document carefully so that you know exactly what is expected in your submission.
Instructions
This assignment is intended for students who previously failed the practical component of
CS26020. If you are unsure about whether you should do this assignment or not, please contact
[email protected] for advice before continuing.
Your answer to this assignment must be uploaded to the TurnItIn submission point as a SINGLE
PDF file by Friday 14th August 2020, at 13:00 BST (1pm). You may submit multiple versions of
your assignment, but only the last version submitted will be marked.
Introduction
In your practical work with the Formula AllCode robots you have used several different types of
sensors to guide the behaviour of your robot, including:
• eight infrared sensing obstacle detectors
• two downward-pointing line sensors
• one forward-facing light sensor
• left- and right-wheel encoders
In addition to these sensors that you have used, the robots also have an attitude (gravity vector)
sensor, a direction (magnetic vector) sensor, two push-buttons and a microphone.
Based on your knowledge of the robots and their sensors, answer all of the following three
questions. They are designed to be answered as program design/essay questions. Use code
fragments or pseudocode and diagrams as appropriate in your answers. You are not required to
produce actual working code, but rather a sufficiently detailed outline or pseudocode from which
actual code could be written. You are not expected to test and debug code on an actual robot.
Your answers should refer to the specific hardware and software facilities of your robot. You should
write complete and detailed answers for each section, and your whole assignment should be no
less than 2 and no more than 4 sides of A4 (approx. 1000 – 2000 words). The maximum total
mark available for this assignment is 50.
Question 1 of 3 [15 marks]
Sensors in general are only able to give an approximation to the “real world” values of the things
that they measure, and the sensors on your robot are no exception to this. In order to give
meaningful numbers to the repeatability and accuracy of sensors, it is necessary to perform
experiments to collect sensor data under controlled conditions.
1. Explain what is meant by the terms “repeatability” and “accuracy” as applied to the robot
sensors.
2. Describe in detail how you would go about collecting data to assess the repeatability and
accuracy of the robot's infrared obstacle detectors.
3. Describe how you would analyse the collected data mathematically to produce measures of
repeatability and accuracy for the obstacle detectors.
Question 2 of 3 [20 marks]
Imagine your robot is placed in an enclosed rectangular arena with a number of randomly-placed
obstacles. There are also some randomly-placed black stripes on the floor of the arena.
1. How could you use the available sensors together with the motors to program the robot to
drive around the arena in straight lines for one minute while avoiding hitting any obstacles
or the walls?
2. How could you make the robot also count the number of black lines it crosses during that
minute?
In your explanation, give details of how your algorithm could be implemented on the actual robot,
using the hardware and software facilities available. Pay attention to how your behaviours are
triggered and how they interact.
Question 3 of 3 [15 marks]
In order to solve the maze-running problem of Practical Worksheet 5, your robot would need to
keep an internal model of the maze and have a method of tracking its current position and
orientation within that model.
Describe in detail:
1. What program data structures you would use to model the maze, and why you would make
those choices;
2. How you would keep track of the robot's position and orientation in the maze model, and
how you would update these as the robot moves from cell to cell in the maze;
3. How you would detect the "nest" area and how you would record this information in your
internal maze model.
2020-07-08 V1.4 LGT

欢迎咨询51作业君