Section 46.3: Whole-body and operational-space control

"Whole-body control is where kinematics stops making promises it cannot keep under contact."

A Contact Dynamics Seminar

A floating-base humanoid is often modeled as $M(q)\ddot q + h(q, \dot q) = S^\top \tau + J_c(q)^\top \lambda$, with configuration $q$, mass matrix $M$, bias term $h$, selection matrix $S$, joint torques $\tau$, and contact wrench multipliers $\lambda$. This is the minimum equation needed to keep manipulation, balance, and contact in one mathematical object.

Operational-space control adds task variables $x = \phi(q)$ with Jacobian $J = \partial \phi / \partial q$. The task-space inertia $\Lambda = (J M^{-1} J^\top)^{-1}$ exposes how hard it is to accelerate a hand, torso, or center of mass in a given configuration. The insight is practical: reaching with the hand changes the effective control problem for the whole body.

Theory

Whole-body control usually combines equality constraints, such as rigid contacts or desired accelerations, with inequality constraints, such as torque limits, friction cones, and joint bounds. The most common implementation is a hierarchy or quadratic program that trades exact satisfaction of high-priority constraints against soft lower-priority objectives.

This is where model-based libraries remain crucial even in learned systems. A learned policy can propose targets, contact schedules, or residuals, but the execution stack still benefits from explicit multibody dynamics and force feasibility checks.

The strongest evaluation artifact is a synchronized trace with task errors, solver status, contact wrench estimates, torque saturation, and recovery actions.

Worked Example

A small whole-body ledger is enough to show whether a reach task is balance-limited or simply poorly tuned.

trial = {
    "hand_error_cm": 1.8,
    "com_error_cm": 0.9,
    "max_torque_ratio": 0.87,
    "contact_slip_cm": 0.2,
    "qp_status": "solved",
}
print(trial)
print({"feasible": trial["qp_status"] == "solved" and trial["max_torque_ratio"] < 1.0})

Expected output interpretation. The hand target is being tracked within a small error while torque and slip remain inside the envelope. This is the kind of evidence that distinguishes a successful whole-body trial from a lucky reach.

Practical Recipe

1. Write the exact task stack and task priority policy before tuning weights.
2. Instrument contact consistency and torque saturation from the first day.
3. Test reaching while perturbing support, payload, and friction.
4. Compare one model-based baseline to one learned or residual-augmented variant on the same task panel.
5. Archive the solver traces, not only the end-effector plots.

This section is an ideal place to emphasize why multibody modeling has not disappeared in the age of foundation models. Whole-body control is still where feasibility becomes numerically explicit.

It is also where students can see the difference between 'the robot reached the target' and 'the robot reached the target with a feasible dynamic budget.' That distinction matters throughout humanoid research.

If whole-body control fails, first isolate whether the task is dynamically infeasible, incorrectly prioritized, or simply under-instrumented. Those are three very different debugging paths.