Artificial intelligence (AI) can be used in a variety of ways to control robotic devices, such as by managing their motion, helping them to select and grasp objects, or even allowing them to interact naturally with humans.
One way AI can be used to control robotic devices is through a technique called inverse kinematics, which calculates the joint angles a robotic arm would need to be in to reach a specific goal point. Inverse kinematics can create fluid, natural-looking movements but is very computationally demanding, so it has not seen widespread adoption in industry yet.
Another way AI can be used in controlling robotic devices is known as robotic grasping, wherein a robot arm is used to pick an object up off a shelf or table. Although achieving precise grasping can be difficult, this method has potential for widespread industrial use.
Finally, some AI research has been focused on developing techniques for human-robot interaction (HRI). This involves creating robots that can interact with humans in a natural way, such as asking for help with a task. HRI is still in its early stages but promises to be significant in the future.
References:
https://en.wikipedia.org/wiki/Artificial_intelligence
https://en.wikipedia.org/wiki/Inverse_kinematics
https://en.wikipedia.org/wiki/Robotic_grasping
https://en.wikipedia.org/wiki/Human-robot_interaction