Studied task

In this section, we bring detailed explanations for the 5 AI-enabled CPS in our experiment.

1. Point Reaching

The robotic manipulator needs to reach a designated point with its end-effector.  This basic skill is fundamental for completing more complex manipulation tasks.

2. Cube Stacking

The robot must grasp a cube and place it on top of a target cube.  This task requires the AI controller to manage the spatial relationships between the two cubes accurately.

3. Peg In Hole

The robot is required to insert a peg into a small hole at the center of a pillar.  This task demands precise control of the pose of the object.

4. Ball Balancing

The robotic manipulator should hold a tray and keep a ball centred on the tray's top surface.  Since the robot has no direct contact with the ball, this task introduces increased complexity in controller design.

4. Ball Catching

The robot aims to catch a ball thrown at it using a box-shaped tool. This task requires the ability to interact with a moving object.