joint space control

What it is:

A control method where you directly command each joint to move to specific angles, rather than specifying where you want the hand to be in space. The robot controls motors individually at the joint level.

How it works:

You specify target angles for each joint:

1. Joint 1 → 45°
2. Joint 2 → 90°
3. Joint 3 → 120°
4. etc.

The control system (using PID controllers and encoders) drives each motor to reach its target angle, and the hand ends up wherever that configuration places it.

Contrast with Cartesian space control:

1. Joint space: "Move joint 1 to 45°, joint 2 to 90°" (you specify joint angles)
2. Cartesian space: "Move hand to position X=0.5, Y=0.3, Z=1.2" (you specify hand location)

Advantages:

1. Computationally simple — No inverse kinematics needed; direct motor commands
2. Fast execution — Less real-time calculation overhead
3. Predictable — Each joint moves independently and reliably
4. Direct motor control — Straightforward to implement with PID controllers
5. Robust — Works even with kinematic redundancy (multiple solutions)

Disadvantages:

1. Non-intuitive — You must mentally calculate joint angles to reach spatial targets
2. Limited adaptability — Hard to adjust motion if obstacles suddenly appear
3. Path unpredictability — The hand's path through space isn't guaranteed to be straight or smooth

Why traditional industrial robots use it:

Traditional robots perform repetitive, pre-programmed tasks in controlled factory environments:

1. Pick up part at location A, place at location B
2. Weld a seam along a known path
3. Stack boxes in a fixed pattern

Engineers pre-calculate the joint angles needed for each task, program them once, and the robot executes them millions of times. No real-time computation needed—just reliable joint-by-joint control.