docs(ik): demonstrate inverse kinematics with a locked joint

docs/source/IK/numerical.rst

@@ -261,3 +261,32 @@ In the following example, we create a :py:class:`~roboticstoolbox.models.URDF.Pa
     >>> ets = panda.ets()  # Get the ETS
     >>> Tep = ets.fkine([0, -0.3, 0, -2.2, 0, 2, 0.7854])  # Make a goal pose
     >>> ets.ikine_LM(Tep)  # Solve the IK problem
+
+.. rubric:: IK with a locked joint
+
+An arm with seven joints can still reach a six-dimensional end-effector pose
+while one joint is held at a specified angle.  :meth:`~roboticstoolbox.ETS.fixjoint`
+replaces that joint by a constant transform and returns a new ETS; the original
+robot and its seven-joint ETS are unchanged.  The initial guess for IK has
+only six coordinates, with the locked joint omitted.
+
+This example starts from a reachable Panda pose, locks joint 2 (counted from
+zero), then solves the resulting six-joint IK problem.
+
+.. runblock:: pycon
+
+    >>> import numpy as np
+    >>> import roboticstoolbox as rtb
+    >>> panda = rtb.models.ETS.Panda()
+    >>> ets = panda.ets()
+    >>> q_goal = np.array([0.2, -0.4, 0.3, -1.5, 0.25, 1.6, -0.2])
+    >>> target = ets.fkine(q_goal)
+    >>> reduced = ets.fixjoint(2, q_goal[2])
+    >>> reduced.n
+    6
+    >>> seed = np.delete(q_goal, 2) + 0.02
+    >>> sol = reduced.ikine_LM(target, q0=seed)
+    >>> sol.success
+    True
+    >>> np.allclose(reduced.fkine(sol.q).A, target.A, atol=1e-5)
+    True