Reference no: EM133118337
ENGT5202 Machine Vision, Robotics & Flexible Automation - De Montfort University
Learning outcome 1: "Systematic understanding of machine vision technology and its potential applications, and a critical awareness of current problems and new insights in the design of machine vision systems"
Learning outcome 2: "Systematic understanding of robotics technology and its application, and a critical awareness of current problems and new insights in the design of robotics systems."
Learning outcome 3: "Ability to deal with complex issues related to flexible / intelligent automation systematically and creatively. Demonstrate self-direction and originality in tackling and solving problems, and act autonomously in planning and implementation of systems integrations."
PART A: Discrete Matlab Filter Implementation
Design in Matlab a discrete implementation for a mean (smoothing) and respectively median filter. Discrete implementation means that you are expected to implement the filter algorithms/functionality in Matlab from scratch using basic syntax, rather than using one of the already built-in Matlab filter functions, similar to the example provided here: "Matlab Examples" → "12. Prewitt, Sobel, and Roberts Gradient Operators" → "Gradient_filtering_discrete_example.m" which you can find on Blackboard in the Assignment folder.
Your code should reside in a Matlab script (.m file) which can be easily run by just calling its name from the Matlab's command window. Again, this .m file should be a Matlab script not a function.
Your two filters should be able to take as an input a small matrix provided by the user in the early part of the code, as shown in the example referred to above. Your code should be flexible enough to allow the user to provide as an input a matrix of any size they want, and still run correctly without having to manually adjust any indexes in the remaining code. The filter window size for both filters is 3x3.
Your filter should be able to operate with and respectively without padding, via a user modifiable parameter included in your m-file called "padding". If padding = 0, your filter should operate without padding. When padding = 1 your filter should use padding by zeros, and if padding = 2 your filter should use padding by replication.
The filtered results for both filters should be suitably displayed in the Matlab command window when the code is run. Test your code with the following input matrix and provide the results in your report for all three types of padding, for each of the two filters:
|
2
|
8
|
0
|
2
|
4
|
4
|
4
|
0
|
|
0
|
2
|
2
|
0
|
4
|
6
|
6
|
0
|
|
4
|
8
|
6
|
6
|
6
|
4
|
4
|
9
|
|
1
|
9
|
1
|
2
|
2
|
3
|
4
|
4
|
Make sure that you use good programming practice by suitably commenting the Matlab code from your .m file script where useful to do so, and by using appropriate code indentation in order to make reading your code more legible.
PART B: A review of Sony AIBO ERS-1000
Perform a comprehensive literature review on Sony's AIBO ERS-1000 model, from the perspective of its sensors, actuators, kinematic/locomotion capabilities and user control capabilities and provide the answer to the following questions in PART B or your report:
B1. List all the sensors available in an AIBO ERS-1000 and for each sensor briefly specify their purpose, if they are proprioceptive or exteroceptive and respectively if they are active or passive.
B2. What actuators does the AIBO ERS-1000 use and what exactly are these used for?
B3. With respect to its kinematic/locomotion capabilities, how many total degrees of freedom does an AIBO ERS-1000 have, how are these distributed on its body and where exactly are they located?
B4. List and briefly explain all the possible ways in which users can control an AIBO ERS-1000.
Make sure you provide suitable references linked in the main text of your report to support your answers. You are expected to provide several authoritative references to cover this section.
Attachment:- Flexible Automation.rar