12. System
1. Overview
This mobile 3D printer robot essentially interprets G-code and navigates while printing. In addition, it maintains print accuracy using motion sensors or visual markers. A microcontroller (e.g., RP2040) runs firmware (e.g., GRBL or Marlin) to control motors and manage temperature, position, and obstacle detection. Sensor fusion is used to compensate for wheel slippage and ensure consistent localization.
System diagram
2. Input devices
The input devices for this machine have the following requirements:
- G-code reading
- End stop (Zeroing)
- Hot end temperature management
- Obstacle detection
- Understanding accumulated travel distance
- Understanding self-position and orientation (to compensate for deviations caused by wheel slippage etc.)
I am considering placing end stops (bumpers) on all four sides as basic sensors, and also finding a way to detect and correct the robot's own position. Also, I need to consider how this machine can actually move without interfering with printed model (especially on the Z axis).
Options for position and orientation detection
For detecting and correcting position, so far, I am considering the following input devices:
- Rotary encorders
- IMU (Inertial Measurement Unit)
- Optical flow sensors
- ArUco markers I briefly asked ChatGPT to compare these options.
Prompt: Compare the following input devices for the project: Rotary Encoder, IMU, Optical Flow Sensor, ArUco Marker
| Feature | Rotary Encoders | IMU (Inertial Measurement Unit) | Optical Flow Sensors | ArUco Markers |
|---|---|---|---|---|
| How It Works | Measures wheel rotation using an optical or magnetic sensor. Converts rotation into position or speed data. | Uses accelerometers and gyroscopes to measure acceleration and angular velocity. Data is integrated over time to estimate movement. | Captures images of the surface and tracks movement by comparing pixel shifts. Works similarly to an optical mouse. | Uses a camera to detect square fiducial markers in the environment and calculates position based on their known size and position. |
| Measures | Wheel rotation (position, speed) | Rotation, acceleration | Surface movement (X, Y) | Global position (X, Y) |
| Accuracy | High (precise for wheels) | Medium (drifts over time) | High on textured surfaces | Very High (camera-dependent) |
| Drift | No drift | Yes (needs correction) | Some drift (depends on surface) | No drift |
| Response Time | Fast (real-time) | Fast (but noisy) | Fast | Slower (depends on camera) |
| Environment Sensitivity | Works in any condition | Affected by vibrations | Fails on uniform surfaces (glass, smooth floors) | Needs good lighting & clear markers |
| Cost | Cheap (~$2-$10) | Affordable (~$5-$15) | Medium (~$10-$30) | Camera + markers (~$30-$100) |
| Setup Complexity | Easy | Easy (with filtering) | Medium (needs surface tuning) | Hard (camera calibration) |
| Use Case | Precise wheel movement | Balance, rotation tracking | Non-wheel-based tracking | Absolute positioning |