Staff Robotics Control Engineer - Whole-Body Control (Wheeled Platform)
Humanoid · London
Here at Humanoid, we believe in a future where robots amplify human potential. That’s why we’ve set out on a mission to build the world’s most capable, commercially-scalable, and safe humanoid robots. We’re bringing that mission to life with HMND‑01 - our rapidly developed humanoid platform being deployed in real industrial environments - and we’re growing the team to take it even further.
About the Role
We are looking for a Staff Robotics Control Engineer to join our Control Team in London and lead the evolution of whole-body control for our Wheeled Alpha platform - a humanoid whose whole body spans a wheeled base, torso and two arms, controlled as one coordinated system.
We need genuine depth in classical whole-body control. You should know hierarchical and weighted QP formulations from the inside - not as a library you have called, but as something you have derived, implemented, tuned, profiled and debugged on real hardware under demo pressure. You should have opinions about strict hierarchies versus soft-weighted stacks, regularisation and nullspace behaviour, what to do when a task set becomes infeasible, and how to keep a redundant system clear of singularities without it feeling sluggish. Alongside that depth, you will set technical direction and mentor a strong team of control engineers.
What You'll Do
Whole-Body Control Architecture & Algorithms:
Lead the evolution of the Wheeled Alpha whole-body controller: task formulation, prioritisation, weighting, constraint handling and solver selection.
Coordinate wheeled base, torso and dual-arm motion within a single whole-body formulation, with base motion that tracks commanded targets accurately and stays well-behaved under aggressive commands and long-duration operation.
Own the constraint layer - self-collision avoidance, joint and velocity limit guarding, and singularity handling - plus feasibility handling and recovery when the solver fails.
Drive the evolution toward predictive formulations: evaluate and develop MPC-based whole-body control where anticipating dynamics and constraints outperforms instantaneous QP resolution.
Dynamics, Compliance & Payload:
Own the dynamic models the controller runs on - inertial, gravity and friction modelling, system identification, and online payload estimation - so the robot knows what it is carrying and controls accordingly.
Advance compliance and force behaviour - impedance, admittance and force control that stays compliant at contact - and extend payload capability without sacrificing either.
Own stability-aware behaviour under changing payload and dynamic motion
Safety Integration:
Make whole-body control a first-class citizen of the safety architecture: the controller must accept and honour requests from dedicated safety compute - velocity capping, controlled deceleration, constrained operating envelopes - provably and without degrading motion quality.
Co-own the controls-to-safety interface with the safety team, and design control behaviour so that safety interventions are graceful rather than abrupt.
Technical Leadership:
Define the performance metrics for whole-body control and insist on deterministic, reproducible evaluation of every change, in simulation and on hardware.
Mentor control engineers on optimisation-based control, numerical robustness and hardware debugging; act as reviewer of record for the most consequential changes.
Work with hardware teams so control constraints and mechanical design decisions stay coherent.
What We're Looking For
Deep, demonstrable expertise in optimisation-based whole-body control: hierarchical and weighted QP formulations, task prioritisation, constraint handling, and the numerical realities of solving them inside a real-time loop.
Strong command of the classical WBC toolkit - operational space and task-space control, inverse dynamics and IK formulations, redundancy resolution, nullspace projection, impedance and admittance control, contact and force control - and working knowledge of MPC formulations for whole-body or mobile-manipulation control.
Deployed whole-body controllers on real high-DoF hardware (humanoids, mobile manipulators, dual-arm or legged platforms), including coordinating base and manipulator motion.
Practical fluency with QP/NLP solvers and their failure modes: conditioning, infeasibility, warm starting, solve-time budgets, and what to do at the limit.
Strong dynamic modelling foundations: rigid body kinematics/dynamics (Jacobians, dynamics formulations, URDF-based modelling), inertial and friction system identification, and payload/external-load estimation.
Expert modern C++ for real-time control with Python for analysis and prototyping.
Track record of taking control algorithms to production reliability rather than to a paper or a one-off demo, plus demonstrated technical leadership and mentoring.
What We Offer
Competitive equity: stock options with meaningful upside as we scale.
30+ paid days off, including 23 days of annual leave, all UK bank holidays, and additional company closure days (including Christmas–New Year shutdown).
Private healthcare, including virtual and in-person care.
Pension scheme with 8% total contribution (5% employee, 3% employer) on full earnings.
Free daily breakfast, catered lunch, and snacks in-office.
Work at the frontier - collaborate daily with world-class engineers, researchers, and product experts building the next generation of AI and humanoid robotics.
Real ownership - direct access to founding leadership, meaningful input on product direction, and the ability to drive key initiatives from day one.