MyCobot Pro 600 – Curved Path Planning and Kinematics (MATLAB)

This repository contains MATLAB scripts (and optional Python helpers) to enable the MyCobot Pro 600 robotic arm to follow a curved path detected from an image.
It integrates computer vision, coordinate mapping, and kinematic control with optional Simulink models for simulation.

---

🧩 Files

File	Description
cobotLineFollower.m	Main function mycobot_ik(image_path) — extracts a curve from an image, samples waypoints, maps them to robot coordinates, and computes joint angles via IK. Includes local helpers for skeleton-path finding and interpolation.
imageee.m	Script version of the image ➜ path ➜ waypoint pipeline (exploratory/notebook-style).
angle.m	Utility that formats and displays jointAngles_all over a simple time vector.
robott.m	Loads tocheck.urdf, sets a configuration, and visualizes the robot in MATLAB.
Rigiftreedof.m	Builds a multi‑DOF rigidBodyTree (supporting simulation/IK).
Forward.slx, inverse.slx, rigidtree.slx	Simulink models for FK/IK and rigid-body structure (optional).
tocheck.urdf	Robot model (URDF) used by robott.m.
Forward.py	(Optional) Companion Python script, if you prefer prototyping parts of the pipeline in Python.

---

⚙️ Requirements

• MATLAB R2023a or later
• Image Processing Toolbox
• Robotics System Toolbox
• Simscape Multibody Toolbox (for Simulink models)

---

🚀 Quick Start (MATLAB)

1. Place an input image (black curve on white background) at, for example: images/s11.jpeg.
2. From MATLAB, run:

   jointAngles_all = mycobot_ik('images/s11.jpeg');   % in cobotLineFollower.m

3. (Optional) Visualize or simulate:

   robott                   % visualize the URDF model
   open('Forward.slx');     % or open('inverse.slx');

Output: jointAngles_all (N×6) containing joint solutions for sampled waypoints along the curve.

---

🔢 How it Works

• Image ➜ Path: Thresholding → morphology → largest component crop → skeletonization → endpoints → BFS to trace a single path.
• Sampling: Evenly samples points along the traced skeleton length.
• Mapping: Converts pixels to millimeters using a scale factor and remaps to the robot frame.
• IK: Builds a rigidBodyTree (modified DH, MDH transforms) and solves for each waypoint using inverseKinematics with a fixed pose orientation.

---

🧪 Example Snippets

Call the pipeline and inspect angles:

jointAngles_all = mycobot_ik('images/s11.jpeg');
disp(jointAngles_all);

Use the visualization helper:

robott

Simulink (optional):

open('rigidtree.slx');  % structure
open('Forward.slx');    % forward kinematics
open('inverse.slx');    % inverse kinematics

---

🛠 Troubleshooting

• “Fewer than 2 endpoints found”: Adjust thresholdValue or lighting/contrast in the input image.
• “No path found between endpoints”: Ensure a single continuous dark curve; reduce noise with larger morphology (strel('disk', …)).
• Unstable IK: Tweak initial guess in cobotLineFollower.m or reduce waypoint spacing.

---

📜 License

This repository is intended for educational and research use.

---