Visual Servoing Implementation for a 2-DOF Robot

This repository contains the implementation of a visual servoing algorithm for a 2-DOF robot.

Steps

Step 1: Object Placement and Image Capture

1. Spawn Object:

   • Spawn the object created in the previous homework on the ground within the robot's workspace.

2. Move Robot:

   • Move the robot via the position controller to ensure the entire object is visible in the image.

3. Image Capture:

   • Capture an image and obtain the coordinates of the 4 circle centers.

4. Output:

   • Present the captured image from the virtual camera.
   • Display the detected circle centers.

Step 2: Robot Relocation and Image Capture

1. Relocate Robot:

   • Move the robot to a different location using the position controller.

2. Ensure Visibility:

   • Ensure the entire object is still visible by the virtual camera.

3. New Image Capture:

   • Capture a new image and obtain the coordinates of the 4 circle centers.

4. Output:

   • Present the new captured image from the virtual camera.
   • Display the detected circle centers in the new location.

Step 3: Visual Servoing Algorithm Implementation

1. Algorithm Implementation:

   • Implement a visual servoing algorithm using the four point features (centers of the circles).

2. Controller Switch:

   • Switch the controller to velocity mode to execute the velocity command.

3. Image Jacobian:

   • In the image Jacobian, use f and Z, both set to 1.

4. Note:

   • All parameters are multiplied with a gain (lambda) in this 2D example and do not have a direct effect.

5. Output:

   • Record the locations (x, y coordinates) of all features over time during visual servoing.

   • Plot the trajectories of the features in the XY plane.

     trial1.webm

Note: Adjust parameters, gains, and other specifics based on your implementation details.