Summer Courses — Python, CAM toolpaths & a robot-arm simulation

A collection of small, self-contained Python scripts from a summer course: turtle graphics, plotting, computer vision, and a set of scripts that build CAM toolpaths and run them on a simulated 6-DOF robot arm.

Everything here is an offline simulation or a local visual — nothing talks to real hardware.

The point of this repo: travel from syntax (the literal mechanics of code) to intent (describe a goal; let a solver work out the how). Read LEARNING_PATH.md first — it explains how the files are arranged to take you across that transition.

Setup

pip install -r requirements.txt

Requires Python 3.9+. Most scripts open a window (matplotlib or turtle); a few use the webcam. See requirements.txt for what each package is for.

New to git/GitHub? See the GitHub workflow cheat sheet for how to add, commit, and push your changes.

---

Robot arm + CAM toolpaths (the main project)

A UR10e-like 6-DOF arm whose end-effector executes machining toolpaths over a paraboloid part. The toolpath is optimised so every point lands inside the arm's reach and the motion stays gentle.

File	What it does
robot_hand_simulation.py	The original sim: the arm's end-effector follows a face/hand tracked by the webcam. Forward kinematics + a heuristic IK. The starting point for everything below.
cam_toolpath_paraboloid.py	The original CAM demo: draws a spiral toolpath descending over a paraboloid bowl, in 2D turtle with an isometric projection.
robot_spiral_toolpath.py	Combined + optimised. Generates the paraboloid spiral, inverts it into a small dome that sits on the table in front of the robot, auto-places it so the whole path is reachable, and drives the arm along it with a real analytic IK (tool tip lands exactly on each point). Strictly offline (no camera).
robot_zigzag_toolpath.py	Second strategy. A zig-zag (raster / boustrophedon) toolpath over the same dome. Imports the robot model, IK/FK, dome surface, scale, and placement optimiser from robot_spiral_toolpath.py — only the path pattern differs.

Run them

python robot_spiral_toolpath.py     # spiral toolpath on the dome
python robot_zigzag_toolpath.py     # zig-zag toolpath on the dome
python robot_hand_simulation.py     # webcam face-following demo

Each toolpath script prints a reach/optimisation report, then animates the arm machining the part in 3D. Close the window to exit.

How it fits together

cam_toolpath_paraboloid.py  ─┐
                             ├─►  robot_spiral_toolpath.py  ──►  robot_zigzag_toolpath.py
robot_hand_simulation.py    ─┘    (shared robot + dome core)     (imports the core)

robot_spiral_toolpath.py holds the reusable core — solve_ik, forward_kinematics, the inverted-dome surface, the TARGET_SCALE, and optimise_placement(). The zig-zag script and any future pattern just plug a new toolpath generator into it.

Next step (planned): make the toolpath adaptive — let the camera define or warp the dome surface, and re-run optimise_placement() each frame to keep the path reachable. The code is already structured for this; the optimiser and IK are pure functions you can call per-frame.

---

Other scripts

File	What it does
plot_a_parabola.py	Plots a parabola with matplotlib.
image_tracing.py	Traces edges/contours of an image with OpenCV.
smile.py	OpenCV face/smile demo.
Hilbert.py	Hilbert space-filling curve (turtle).
draw_hexagon.py	Hexagon pattern (turtle).
heptagon.py	Heptagon (turtle).
addition.py	Basic arithmetic example.

Teaching materials

teaching_basics/ is a separate mini-course on Python fundamentals (loops, conditions, lists, stacks/queues, dictionaries, debugging), each lesson made visible with turtle graphics. It has its own README.

---

Notes

• Windows: these were developed on Windows 11. Use python (or py) to run.
• Webcam scripts (robot_hand_simulation.py, smile.py) fall back gracefully if no camera or MediaPipe is available.
• The robot scripts are pure simulation — safe to run anywhere, no hardware needed.