Shaun Barlow, Josh Dunkley-Smith, Pierre Nicolas, Hemavani Sayakobi, Michael Vernon 0
READ THIS BEFORE ADDING IN YOUR WORK 2
Problem Formulation 3
Sensing – THIS IS WORTH 1/6 6
Feedback Control Philosophy – THIS IS WORTH 1/6 7
Output Control – THIS IS WORTH 1/6 8
Plant Modelling – THIS IS WORTH 1/6 9
Figures title: 13
READ THIS BEFORE ADDING IN YOUR WORK
Student groups are required to submit a single written Design Proposal according
to the assessment specification and will include detailed appendices, showing
drawings, simulations, etc. of proposed solutions. The Design Proposal will
include the proposed project design, discussions on progress to date in
completing the required project and a safety plan relevant to the underlying
(2500-3000 words + figures) (10 pages)
Our word (&page) limit is relatively low, so try to incorporate as many figures as you can. I’m going to ask someone(Josh) to make a base inventor assembly so we can all be working from it to keep things relatively similar.
Right now every section is its own page with a page break. This will mean that we’re going to be a little higher in page count, but that’s fine as it will help us to figure out how well-balanced all of our sections are
The cyan text is the pasted info from the rubric, keep it mind when you do your work, and delete it when you’re done.
BY THE END OF THE WEEK (11/08) WE NEED COMPONENTS LOCKED IN
Everybody is expected to have completed their work by the end of Friday (16/08) next week (uni break). It’s like 2 pages max, so just get it done.
For our project, we will be expanding on the first suggested project, “a tennis ball tracking system”. We will be creating a system that will be able to dynamically project a laser onto a randomly bouncing ping-pong ball.
Based on the guidance of our lecturer we will be breaking up our initial goal into three major milestones to ensure that we are able to successfully complete it,
Stage 1 – A system to project a laser onto a ball moving in a pendular motion.
Stage 2 – A system to project a laser onto a bouncing ball at a fixed distance.
Stage 3 – A system to project a laser onto a randomly bouncing and located ball.
Our first stage is analogous to the first suggested project, and we believe will be a good starting goal to ensure that everything is working as expected. Our second stage will allow the ball to move in a much less controlled manner, and will require our system to react much quicker and with greater accuracy. Our third, and final, proposed stage will require the resolution of depth and again place more emphasis on the accuracy and precision of our system.
With these goals in mind we will be able to compose our list of expected problems that we will need to overcome in order for our system to be deemed a “success”. For our initial stage we will need:
• Accurately resolve the location of the ball
• Accurately place a laser onto the located ball
For our second stage:
• Quickly resolve the ball’s location
• A laser guidance system that is able to handle the ball’s rapid speed changes
For our third stage
• Resolve the ball’s distance from the system
• A large enough range of motion of the laser guidance to match our parameters
Further problems presented by the geologic separation of our team and other unique considerations:
• Every team member must be able to easily sourced components
• Our designs must be in a format that all team members have the tools to work with
• Our designs must be able to be assembled by all team members
• Our ball must be regular, standardised, and of a unique colour
• Our camera must be able to resolve the colour of our ball
• Our camera must be of sufficient resolution to allow useful data to be collected
At this time we are making several key assumptions:
• That the monocular camera will be able to effectively resolve depth
• That the camera will be able to reliably distinguish the ball in our standard lighting conditions
• That our mechanical system will be able to accelerate fast enough to match the ball’s movements
• That our mirrors will be able to withstand the potential forces without cracking, warping, or becoming compromised in any other fashion
• That our laser source will be strong enough to enable our located mark to be seen in standard laboratory conditions, or daylight
Our first problem will be what specific components we will be using. Due to the isolated nature of our team members, a high priority needs to be attached to ease of acquisition and standardisation. With this in mind we will be using:
• “Dragonfly Competition Table Tennis Balls – Orange”, available nationally from Rebel Sport
• A CAMERA
• RASPBERRY PI VERSION?
• OTHER BITS?
Akin to our initial problem breakdown, we will be splitting up our solutions into clear stages to better guide our development and research efforts.
Sensing – THIS IS WORTH 1/6
Camera Module v2
Net price $25
Still resolution 8 Megapixels
Video modes 1080p30, 720p60 and 640 × 480p60/90
Linux integration V4L2 driver available
C programming API OpenMAX IL and others available
Sensor Sony IMX219
Sensor resolution 3280 × 2464 pixels
Sensor image area 3.68 x 2.76 mm (4.6 mm diagonal)
Pixel size 1.12 µm x 1.12 µm
Optical size 1/4-
Full-frame SLR lens equivalent
Focal length 3.04 mm
Horizontal field of view 62.2 degrees
Vertical field of view 48.8 degrees
Focal ratio (F-Stop)
What we’re looking for is name, model, brand, details! Your other team members need to be able to read this and understand wtf we’re using, and why
Definitely throw in some other stuff we would never actually use. This is the most expansive rubric, but also the least “he made this up” so should be ok
An elaborate description of the sensing aspects
A supporting argument to why certain sensor types are proposed against others
Reasoning into sampling frequencies and associated complexities in real time applications
Pre-filtering techniques and associated complexities in engaging different techniques. A discussion on appropriate technologies.
A discussion in Image/Video processing complexities
Feedback Control Philosophy – THIS IS WORTH 1/6
This is where we’ll explain the how and why of our system’s operational stuff.
Key point for max grades is the “supporting examples” bit so definitely throw in our original inspiration (Japanese pingo pongo bois) here
They also make a key reference to results so yeah
Elaborate description of control design, i.e. model-based and rule-based with supporting arguments. Examples to establish what is suitable for this application.
Plant modelling and feedback modelling and implementation aspects.
Sensing, output as well as hardware in the loop.
Considerations for design
Control of laser pointer: we could use a controller on the laser pointer where calibration of pointer is not required. Detect pointer’s position using vision and use the x and y error between laser and ball centroid to control gimbal. This may not work where distance to background is significantly larger than distance to ball.
Alternatively, laser gimbal has a fixed origin WRT the camera. Imaging feedback for laser position is not used.
Examples of controls
Ping Pong Ball Tracker
Uses an image processing-only approach to determining position of object.
Analyses each frame for contiguous colours using bwlabel().
Filters returned regions by area and ratio of height to width using predetermined thresholds.
Accesses Centroid property of selected array element.
This approach should be tested for efficiency. Can it run fast enough to produce results in real time?
It follows a rule-based approach and does not take into account any predictions for future position of target.
Computer Vision Toolbox
A single function in the Computer Vision Toolbox that returns an estimate of motion between images or video frames. Can return velocity vectors or matrix of motion between images.
Kalman Filter for Object Tracking
Allows tracking of motion when object is hidden for some frames. Predicts the motion using trajectory of object.
Example of detecting the position of a laser pointer given a fixed background.
The example uses a colour filter to find high intensity red spot. May need to consider our choice of colours of both laser and ping pong ball to make it easier to differentiate the two in image processing.
Output Control – THIS IS WORTH 1/6
Here we need to go into the what and how of our output signals. You will need to cover what PWM the raspberry is capable of, what a standard motor needs to operate, what signal conditioning we need to do to ensure interoperability of our components (voltage level switches, timing signals, etc.) how our choice of PWM parameters will limit our potential reaction speeds. For example a 10 Hz PWM might be perfectly functional but that means we will have to 0.1s for our motors to begin to move which would obviously be far too much latency
Output device analysis i.e. PWM signal use, etc.
Output signal conditioning
Limitations and areas of concern in-terms of performance.
Plant Modelling – THIS IS WORTH 1/6
This one seems basic and is about (I think) how we will be making our CAD designs with foresight into how it will be used within the simulation (Simulink) environment.
The “closed loop” thing is a key point for a higher score, but I don’t really get it.
Also this will be attached to the production of the base CAD design for all members to print to allow for assembly of our system. Obviously will need to wait for final component selection but having robust, portable CAD models for printing and simulation will be absolutely essential to our ability to actually complete this project as nothing else can really happen without them.
As mentioned, I’m happy to help but obv limited as nothing I have on hand (F360, NX) is accepted by Simulink or I would volunteer for this myself.
A sound approach to appropriate plant modelling and underlying assumptions.
Evidence of familiarity in the modelling aspect relevant to this project particularly in terms of first principles.
Plant interactions in the closed loop
Last Name, F. M. (Year). Article Title. Journal Title, Pages From - To.
Last Name, F. M. (Year). Book Title. City Name: Publisher Name.
1[Add footnotes, if any, on their own page following references. For APA formatting requirements, it’s easy to just type your own footnote references and notes. To format a footnote reference, select the number and then, on the Home tab, in the Styles gallery, click Footnote Reference. The body of a footnote, such as this example, uses the Normal text style. (Note: If you delete this sample footnote, don’t forget to delete its in-text reference as well. That’s at the end of the sample Heading 2 paragraph on the first page of body content in this template.)]
Note: [Place all tables for your paper in a tables section, following references (and, if applicable, footnotes). Start a new page for each table, include a table number and table title for each, as shown on this page. All explanatory text appears in a table note that follows the table, such as this one. Use the Table/Figure style, available on the Home tab, in the Styles gallery, to get the spacing between table and note. Tables in APA format can use single or 1.5 line spacing. Include a heading for every row and column, even if the content seems obvious. A default table style has been setup for this template that fits APA guidelines. To insert a table, on the Insert tab, click Table.]
Figure 1. [Include all figures in their own section, following references (and footnotes and tables, if applicable). Include a numbered caption for each figure. Use the Table/Figure style for easy spacing between figure and caption.]
For more information about all elements of APA formatting, please consult the APA Style Manual, 6th Edition.