Why is the observed/desired position and velocity of the robot joints single values?

[rey-allan]

Hello!

This might be my lack of robotics background, so apologize if this is a dumb question. But, I'm having a hard time understanding why the values for the position and velocity of the robot joints are single values. For example, for the 3DOF robot, the observations are 3 values for position and 3 values for velocity.

1. Shouldn't the values for each joint be posX, posY and posTheta (or vel for velocity), like with the puck's values?
2. What does each value represent? For example, if the position of the 1st joint is -1.15570723, what does that mean?

Thanks!

[rey-allan]

Okay, I found this article on forward kinematics, and I think I got it now. The values are the parameters of the joints (e.g. if the joints are revolute joints, then these are the angles in radians). And then using forward kinematics we can find the actual position of the end-effector (the mallet).

Is my understanding correct?

[boris-il-forte]

Yes, you are correct.
Please refer to the documentation here for the description of environment state variables:
https://air-hockey-challenges-docs.readthedocs.io/en/latest/environments.html