Skip to content

Week 18: Applications and Implications / Project Management

Linked mechanism

5-bar link Linked mechanism drawing a portrait


Assignment

Application and Implication

  • Plan a final project masterpiece that integrates the range of units covered, answering:
    • What will it do?
    • Who has done what beforehand?
    • What sources will you use?
    • What will you design?
    • What materials and components will be used?
    • Where will they come from?
    • How much will they cost?
    • What parts and systems will be made?
    • What processes will be used?
    • What questions need to be answered?
    • How will it be evaluated?
  • Your project should incorporate
    • 2D and 3D design,
    • additive and subtractive fabrication processes,
    • electronics design and production,
    • embedded microcontroller design,
    • interfacing,
    • programming,
    • system integration and packaging
  • Where possible, you should make rather than buy the parts of your project
  • Projects can be separate or joint, but they need to show individual mastery of the skills and be independently operable

Project Management

  • Prepare drafts of your final project summary slide (presentation.png, 1920x1080) and video clip (presentation.mp4, 1080p HTML5, < ~1 minute, < ~25 MB), put them in your root directory, and check that they are linked in the final presentation schedule

1. Plan a final project masterpiece that integrates the range of units covered, answering:

What will it do?

It is a portrait drawing robot. It is a robot that has a camera eye that can recognize human faces, a brain that converts the image into a line drawing, and hands that draw the image.

The robot idea

Who has done what beforehand?

There are a lot of SCARA-arm drawing robot examples, but I don’t see many that combine them with portrait drawing mechanisms.

Reference SCARA draw bot projects

What sources will you use?

  • Study the above examples, and ask ChatGPT about implementation ideas
  • Seeed Wiki, including ESP-DL, the deep-learning framework that works on the edge with a pre-built face recognition model
  • SCS0009 servo, FE-URT-1 board, and their manuals

What will you design?

System diagram linked mechanism

  • Software application for the portrait drawing pipeline
    • camera
    • Streaming on TFT LCD
    • Switch to freeze a photo frame
    • Face recognition from a photo
    • Eye landmarks detection
    • Edge extraction of the photo, with emphasis on the eyes
    • Converting the edge image to a series of paths on an xy plane
    • Conversion of the paths to servo angles for the right and left shoulders, and pen-up/pen-down motion
  • Electronic circuit
    • Connect Xiao, TFT, and 3 serial servos
    • Recreate and incorporate part of the FE-URT-1 dev board
  • Mechanism
    • 5-bar linked mechanism
    • Pen-up/pen-down mechanism
    • Enclosure of the electronic parts

What materials and components will be used?

  • PETG/PLA 3D prints for the 5-bar mechanism and robot head parts
  • Electronic parts
    • My original PCB board that incorporates the parts below
    • Seeed Studio Xiao ESP32S3 Sense
    • 1.8-inch SPI full-color TFT LCD
    • Button switch
    • 3 SCS0009 servos
    • 6V adapter
    • DC-DC converter from 6V to 5V
    • Schottky barrier diode
    • FE-URT-1 dev board or DIY version
      • PNP transistor
      • 2 x tri-state buffers
  • Wooden board to secure paper
  • Pen and paper

Where will they come from?

How much will they cost?

The following is a draft BOM for one complete robot. The custom PCB replaces the FE-URT-1 board in the final assembly; an FE-URT-1 can still be used during prototyping and debugging.

Category Item / specification Qty. Made or bought Source Est. unit cost Est. subtotal
Controller Seeed Studio XIAO ESP32S3 Sense 1 Bought Akizuki ¥2,660 ¥2,660
Camera OV3660 camera module with 75 mm cable (120°) 1 Bought AliExpress ¥1,273 ¥1,273
Display 1.8-inch SPI TFT LCD, ST7735S 1 Bought Amazon Japan ¥899 ¥899
User input Momentary push button 1 Bought Akizuki ¥90 ¥90
Actuators FeeTech SCS0009 serial bus servo 3 Bought Akizuki ¥1,680 ¥5,040
Power Regulated 5V3A DC adapter 1 Bought Akizuki ¥1,100 ¥1,100
Electronics Copper-clad PCB board 1 Made Yodobashi ¥469 ¥469
Electronics Schottky Diode 1 Assembled Akizuki ¥10 ¥10
Electronics Condensers 470μF 3 Assembled Akizuki ¥10 ¥30
Electronics Condensers 0.1μF 3 Assembled Akizuki ¥25 ¥75
Electronics 0Ω jumper resistors 3 Assembled Akizuki ¥1 ¥3
Electronics Pin socket 1 set Assembled Akizuki ¥80 ¥80
Electronics Pin header 1 set Assembled Akizuki ¥35 ¥35
Electronics wires 1 set Assembled Amazon ¥570 ¥570
Electronics Terminal Block (5mm) 1 set Assembled Akizuki ¥20 ¥20
Electronics heat-shrink tubing 1 set Assembled Akizuki ¥40 ¥40
Mechanism and enclosure PETG filament for the links, pen lift, body and head Approx. 300 g Made Amazon ¥2,565 ¥800
Hardware Chicago Screws for linkage 1 set Bought Amazon Japan ¥1,627 ¥1,627
Drawing surface Wooden base board 3 Cut/engraved Local dollar shop “Seria” ¥110 ¥330
Consumables Pen 1 Bought Local stationery shop ¥100 ¥100
Consumables Drawing paper 1 pack Bought Local stationery shop ¥100 ¥100
Development tool (not installed in the final robot) FEETECH FE-URT-1 servo interface board 1 Bought Akizuki ¥1,660 ¥1,660
Development power supply (not installed in the final robot) Regulated 6 V, 2.8 A DC adapter for the FE-URT-1 development setup 1 Bought Akizuki ¥1,100 ¥1,100
Final robot total ¥15,351
Total including development tools ¥18,111

The FE-URT-1 and its 6 V, 2.8 A adapter are used only to configure, test, and debug the serial servos; they are not installed in the final robot. The estimates exclude shipping and failed prototype parts.

What parts and systems will be made?

  • Milled PCB
  • portrait drawing pipeline software
  • 5-bar link mechanism

What processes will be used?

  • additive fabrication process - 3D printing of the robot body
  • subtractive fabrication process - Laser engraving of the wooden board
  • Electronics design and production
  • Programming
  • Systems integration

What questions need to be answered?

  • Is there an OV3660 camera with a long cable that works with Xiao ESP32S3?
  • If it is not available in time, can I design a board that directly becomes the face of the robot?
  • ❌ Does the SCS0009 servo move with only 5V input from Xiao?
  • ❌ Does the FE-URT-1 dev board for SCS0009 work with only 5V input from Xiao?
  • Can I design and mill a serial signal converter board that can be an alternative to FE-URT-1?
  • Or is giving 6V to the servo and connecting it with Xiao’s TX enough to move the servos?
  • How can I improve the drawing quality of a 5-bar linked SCARA robot?

How will it be evaluated?

  • Cuteness of the robot
  • Quality of the portrait drawing

Reflections

  • Time is limited, and I have to focus on my final project....

Checklist

  • [X] What will it do?
  • [X] Who has done what beforehand?
  • [X] What sources will you use?
  • [X] What will you design?
  • [X] What materials and components will be used?
  • [X] Where will they come from?
  • [X] How much will they cost?
  • [X] What parts and systems will be made?
  • [X] What processes will be used?
  • [X] What questions need to be answered?
  • [X] How will it be evaluated?
  • [X] Uploaded summary slide (placeholder)
  • [X] Uploaded video clip (placeholder)
  • [X] Checked they are linked in the final presentation schedule

AI Usage

  • I used Codex for grammar checking, and made the final review.

Sources

  • References are in the above “Who has done what beforehand?” section.