Kinematics
Pose
A pose is the complete spatial state of a rigid body: its position and its orientation together, six numbers for a body in free space.
What it is
A pose fully specifies where something is and which way it's facing. For a rigid body in 3D that takes six numbers: three for position (X, Y, Z) and three for orientation (some triple like roll, pitch, yaw). Drop the orientation and you no longer have a pose, you have a point.
In robot terms, a pose is what you command the tool center point to reach. Telling the arm to put the TCP at X=600, Y=0, Z=400 isn't enough on its own, because the wrist can point the tool a hundred different ways at that same location. The pose pins down both.
How it actually works
The position half is straightforward: three linear coordinates in whatever frame you're working in. The orientation half describes how the tool is rotated about that point, and controllers express it as an RPY triple (roll, pitch, yaw) or an equivalent representation like a quaternion or rotation matrix. Together that's a rigid-body pose with six degrees of freedom, which is exactly why general-purpose arms carry six axes: one joint per DOF you need to hit an arbitrary pose in the workspace.
Two poses can share the same position and differ only in orientation, and they can also share both position and orientation while the arm reaches them through different joint solutions. That second point is why a pose alone doesn't determine the arm's shape. A robot like the FANUC CRX-10iA can present the same TCP pose in more than one configuration (elbow up versus elbow down, wrist flipped), so the pose says where the tool is and how it's aimed, and the configuration says how the arm folded to get there.
How it differs
- Position · Position is only the three location coordinates. A pose is position plus orientation. Ask for a position and the tool can face any direction; ask for a pose and the facing is pinned too.
- Orientation · Orientation is only the rotational half, how the tool is aimed. A pose bundles orientation with position so both are specified in one target.
- Configuration · A pose says where the TCP is and how it's oriented. Configuration says which arm posture (elbow up/down, wrist flip) achieves that pose. The same pose can be reached by several configurations.
Where you meet it in the field
- FANUC CRX-10iA · A six-axis collaborative arm you command to a full pose, position plus orientation, at the tool center point.
Common questions
- Is a pose the same as a position?
- No. A position is just the three location coordinates. A pose adds the three orientation values, so it also fixes which way the tool points. If a target only lists X, Y, Z, it's a position, not a full pose.
- Why six numbers for a pose?
- A rigid body in free space has six degrees of freedom: three to move it anywhere (X, Y, Z) and three to rotate it into any orientation. Six numbers cover all of them, which is why an arm needs six axes to reach an arbitrary pose.
- If I command a pose, does the robot always take the same arm shape?
- Not necessarily. A single pose can often be reached through more than one joint solution (elbow up versus down, wrist flipped). The pose fixes the tool's location and orientation; the configuration determines the posture the arm uses to get there.