Safety-Critical Whole-Body Control for Humanoid Robots via Input-to-State Safe Control Barrier Functions

Gyeongjae Park; Jaeheung Park; Kwanwoo Lee; Myeong-Ju Kim; Sanghyuk Park

arxiv: 2605.25546 · v1 · pith:Q2ZVEH3Xnew · submitted 2026-05-25 · 💻 cs.RO

Safety-Critical Whole-Body Control for Humanoid Robots via Input-to-State Safe Control Barrier Functions

Kwanwoo Lee , Sanghyuk Park , Gyeongjae Park , Myeong-Ju Kim , Jaeheung Park This is my paper

Pith reviewed 2026-06-29 21:56 UTC · model grok-4.3

classification 💻 cs.RO

keywords whole-body controlsafety-critical controlcontrol barrier functionshumanoid robotslocomotionteleoperationsafety filterdisturbance rejection

0 comments

The pith

Input-to-state safe control barrier functions allow kinematic safety guarantees to transfer to full dynamic humanoid control under disturbances.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper develops a three-layer control system that generates nominal motions at the kinematic level, applies a safety filter to adjust them, and then tracks the adjusted motions with a dynamic controller. This structure is meant to keep the robot inside joint limits, away from self-collisions and obstacles, and inside workspace boundaries even when the robot model is inaccurate or external forces act. A reader would care because humanoid robots need to operate safely near people, and earlier methods lose their safety properties once real disturbances appear. The central step is choosing the filter parameters conservatively so that proofs written on the simplified kinematic model still protect the full dynamic robot.

Core claim

The paper claims that inserting an ISSf-CBF safety filter between a kinematic whole-body controller and a dynamic whole-body controller, with parameters tuned conservatively, lets multiple kinematic safety constraints be enforced in real time on the full-order humanoid dynamics, producing larger safety margins under model mismatch and bounded disturbances during locomotion, teleoperation, and balancing tasks.

What carries the argument

The ISSf-CBF safety filter, which minimally modifies nominal joint references generated by the kinematic controller to satisfy safety constraints under bounded disturbances before the dynamic controller tracks them.

If this is right

Multiple safety constraints including joint limits, self-collision avoidance, obstacle avoidance, and workspace boundaries are enforced simultaneously in real time.
Safety margins increase under model uncertainties and external perturbations compared with prior approaches.
The same architecture supports locomotion, teleoperation, and single-leg balancing with hand control on physical humanoid hardware.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

The same separation of kinematic safety filtering from dynamic tracking could be tested on other high-degree-of-freedom platforms such as quadrupeds or manipulators.
Replacing the fixed conservative tuning with an online adaptation law might reduce performance loss while preserving the safety transfer property.
Directly embedding the safety filter inside the dynamic controller rather than at the kinematic level would remove the need to assume disturbance bounds are known in advance.

Load-bearing premise

That conservative tuning of the ISSf-CBF parameters makes the kinematic safety guarantees carry over to the full dynamic humanoid model even when unknown disturbances are present.

What would settle it

A hardware trial in which the robot violates a stated safety constraint, such as joint limit or self-collision, while the actual disturbances remain inside the bounded range used to design the filter would show the transfer of guarantees does not hold.

read the original abstract

Safety-critical control is essential for humanoid robots operating in complex human-centered environments, where physical safety constraints such as joint limits, self-collision avoidance, obstacle avoidance, and workspace boundaries must be satisfied during real-robot operation. However, existing approaches remain limited because kinematic safety guarantees can be degraded in the presence of unknown disturbances, such as model uncertainties, trajectory-tracking errors, and external perturbations. This paper presents a hierarchical safety-critical whole-body control framework for humanoid robots based on input-to-state safe control barrier functions (ISSf-CBFs). The proposed architecture integrates a kinematic-level whole-body controller (KinWBC), an ISSf-CBF safety filter, and a dynamic-level whole-body controller (DynWBC). KinWBC generates nominal joint-motion references from prioritized tasks; the ISSf-CBF filter minimally modifies these references to satisfy kinematic safety constraints under bounded disturbances; and DynWBC tracks the filtered references while enforcing full-body dynamic feasibility and contact stability. Safety constraints are imposed on a whole-body kinematic model, and the ISSf-CBF parameters are conservatively tuned so that the resulting kinematic safety guarantees can be transferred to full-order humanoid dynamics under unknown disturbances. Simulation and real-robot experiments demonstrate that the proposed framework improves safety margins under model mismatch and reliably enforces multiple safety constraints in real time during locomotion, teleoperation, and single-leg balancing with hand control. Project website: https://kwlee365.github.io/SafeWBC-Website/

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Desk Editor's Note private letter to a colleague

The paper layers an ISSf-CBF filter between kinematic and dynamic whole-body controllers for humanoid safety, but the transfer of guarantees under disturbances lacks explicit bounds.

read the letter

The main thing here is that the authors have a practical three-layer controller for humanoids that adds an ISSf-CBF safety filter between kinematic and dynamic whole-body control, and they show it working on hardware for a few tasks. The new part is the explicit handoff where they tune the CBF conservatively on the kinematic model so its safety properties survive when the dynamic controller tracks it under disturbances.

They do a decent job laying out the architecture and running experiments on locomotion, teleoperation, and balancing. The real-robot results apparently show tighter safety margins than without the filter, which is useful for anyone trying to deploy these systems around people.

The soft spot is right where the stress-test note says: the transfer from kinematic guarantees to full dynamics relies on "conservative tuning" without spelling out the disturbance set or giving a formal argument that the filtered references stay safe once the dynamic layer is in the loop. The abstract mentions model mismatch and external forces but doesn't quantify the bounds or show how the tuning ensures the ISSf-CBF properties hold after tracking errors. Experiments demonstrate improvement, yet without a stated limit on how large the disturbances can be before the guarantee breaks, the safety claim is more empirical than theoretical. That's not fatal for a robotics paper, but it does mean the central promise needs more support in the full text.

This is the kind of work that matters for the safe control community in robotics. Readers who care about CBF applications to humanoids will find the layering interesting and the hardware validation relevant. It deserves a serious referee because the idea is concrete, the experiments are on a real platform, and the gaps are fixable with more derivation or tighter bounds rather than being foundational problems.

I'd recommend sending it out for review with a note to the authors to strengthen the disturbance analysis section.

Referee Report

1 major / 0 minor

Summary. The paper proposes a hierarchical safety-critical whole-body control framework for humanoid robots that integrates a kinematic whole-body controller (KinWBC) to generate nominal joint references, an input-to-state safe control barrier function (ISSf-CBF) safety filter to minimally modify those references while enforcing kinematic safety constraints under bounded disturbances, and a dynamic whole-body controller (DynWBC) to track the filtered references while ensuring dynamic feasibility and contact stability. The central claim is that conservatively tuned ISSf-CBF parameters on a whole-body kinematic model allow kinematic safety guarantees to transfer to the full-order dynamics under unknown disturbances (model mismatch, tracking errors, external forces), with simulation and real-robot experiments on locomotion, teleoperation, and single-leg balancing demonstrating improved safety margins and reliable real-time enforcement of multiple constraints.

Significance. If the conservative tuning reliably transfers the ISSf-CBF kinematic guarantees to the dynamic level without explicit disturbance bounds or additional proof, the framework would offer a practical method for enforcing multiple safety constraints on humanoids in human-centered environments, extending standard CBF approaches with a clear separation between kinematic safety filtering and dynamic tracking.

major comments (1)

[Abstract] Abstract: the claim that 'the ISSf-CBF parameters are conservatively tuned so that the resulting kinematic safety guarantees can be transferred to full-order humanoid dynamics under unknown disturbances' is load-bearing for the safety guarantee, yet the provided description supplies neither an explicit disturbance set (e.g., bounds on model mismatch or external forces), Lipschitz constants, nor a formal argument (ISS or Lyapunov-style) showing that filtered kinematic references remain safe once tracked by DynWBC; experiments are stated to demonstrate improvement but without stated bounds that real-robot disturbances respect, the reliability rests on unverified extrapolation rather than the ISSf-CBF theory.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for their detailed and constructive feedback on our manuscript. We address the major comment point by point below.

read point-by-point responses

Referee: [Abstract] Abstract: the claim that 'the ISSf-CBF parameters are conservatively tuned so that the resulting kinematic safety guarantees can be transferred to full-order humanoid dynamics under unknown disturbances' is load-bearing for the safety guarantee, yet the provided description supplies neither an explicit disturbance set (e.g., bounds on model mismatch or external forces), Lipschitz constants, nor a formal argument (ISS or Lyapunov-style) showing that filtered kinematic references remain safe once tracked by DynWBC; experiments are stated to demonstrate improvement but without stated bounds that real-robot disturbances respect, the reliability rests on unverified extrapolation rather than the ISSf-CBF theory.

Authors: The ISSf-CBF formulation provides input-to-state safety guarantees with respect to bounded disturbances acting on the kinematic model, with the class-K function in the barrier condition explicitly designed to tolerate such disturbances. Conservative tuning of the CBF parameters is performed based on observed tracking errors between the kinematic references and the full-order dynamics under DynWBC, creating an additional safety margin. The manuscript does not supply explicit numerical bounds on disturbances, Lipschitz constants of the dynamics, or a complete Lyapunov-style proof establishing that the kinematic safety certificates transfer exactly to the closed-loop full-order system. Instead, the transfer is justified by the hierarchical separation (kinematic safety filter followed by stable dynamic tracking) together with empirical validation across simulation and hardware experiments. We agree that the current presentation leaves this transfer step insufficiently formalized and will revise the abstract and add a clarifying paragraph in Section IV or the discussion to state the modeling assumptions and the role of conservative tuning more explicitly. revision: partial

Circularity Check

0 steps flagged

No circularity: kinematic-to-dynamic safety transfer rests on stated conservative tuning assumption, not self-referential fitting or definition

full rationale

The paper presents a hierarchical architecture (KinWBC + ISSf-CBF filter + DynWBC) where safety constraints are imposed on the kinematic model and parameters are tuned conservatively to transfer guarantees under disturbances. This is an explicit engineering assumption rather than a derivation that reduces to its own inputs. No equations equate a prediction to a fitted quantity by construction, no self-citation chain bears the central claim, and no ansatz or uniqueness result is smuggled in. The approach extends standard CBF methods without renaming known results or forcing outcomes via self-definition. The derivation chain remains self-contained against external benchmarks.

Axiom & Free-Parameter Ledger

1 free parameters · 1 axioms · 0 invented entities

The central claim rests on one domain assumption about bounded disturbances and on the existence of tunable parameters that can be set conservatively; no new entities are introduced and no free parameters are numerically fitted in the abstract.

free parameters (1)

ISSf-CBF parameters
Parameters are conservatively tuned to ensure transfer of kinematic safety guarantees to the dynamic model.

axioms (1)

domain assumption Disturbances (model uncertainties, tracking errors, external perturbations) are bounded
This boundedness is required for the input-to-state safety property of the CBF filter to hold.

pith-pipeline@v0.9.1-grok · 5813 in / 1394 out tokens · 27893 ms · 2026-06-29T21:56:37.764155+00:00 · methodology

discussion (0)

Reference graph

Works this paper leans on

42 extracted references · 2 canonical work pages · 1 internal anchor

[1]

A Whole-Body Control Frame- work for Humanoids Operating in Human Environments,

L. Sentis and O. Khatib, “A Whole-Body Control Frame- work for Humanoids Operating in Human Environments,” Proc. 2006 IEEE Int. Conf. on Robotics and Automation (ICRA), pp. 2641–2648, 2006

2006
[2]

Real-Time Obstacle Avoidance for Manipu- lators and Mobile Robots,

O. Khatib, “Real-Time Obstacle Avoidance for Manipu- lators and Mobile Robots,”The International Journal of Robotics Research, vol. 5, no. 1, pp. 90–98, 1986

1986
[3]

Control Barrier Functions: Theory and Applications,

A. D. Ames, S. Coogan, M. Egerstedt, G. Notomista, K. Sreenath, and P. Tabuada, “Control Barrier Functions: Theory and Applications,”Proc. European Control Conf. (ECC), pp. 3420–3431, 2019

2019
[4]

Potential field methods and their inherent limitations for mobile robot navigation,

Y . Koren and J. Borenstein, “Potential field methods and their inherent limitations for mobile robot navigation,” Proc. IEEE Int. Conf. on Robotics and Automation (ICRA), pp. 1398–1404, 1991

1991
[5]

Comparative Analysis of Con- trol Barrier Functions and Artificial Potential Fields for Obstacle Avoidance,

A. Singletary, K. Klingebiel, J. Bourne, A. Browning, P. Tokumaru, and A. Ames, “Comparative Analysis of Con- trol Barrier Functions and Artificial Potential Fields for Obstacle Avoidance,”Proc. IEEE/RSJ Int. Conf. on Intelli- gent Robots and Systems (IROS), pp. 8129–8136, 2021

2021
[6]

Constraint-consistent task-oriented whole-body robot for- mulation: Task, posture, constraints, multiple contacts, and balance,

O. Khatib, M. Jorda, J. Park, L. Sentis, and S.-Y . Chung, “Constraint-consistent task-oriented whole-body robot for- mulation: Task, posture, constraints, multiple contacts, and balance,”The International Journal of Robotics Research, vol. 41, no. 13–14, pp. 1079–1098, 2022

2022
[7]

Con- trol Barrier Function Based Quadratic Programs for Safety Critical Systems,

A. D. Ames, X. Xu, J. W. Grizzle, and P. Tabuada, “Con- trol Barrier Function Based Quadratic Programs for Safety Critical Systems,”IEEE Trans. on Automatic Control, vol. 62, no. 8, pp. 3861–3876, 2017

2017
[8]

Real-Time Safety and Control of Robotic Manipula- tors with Torque Saturation in Operational Space,

M. A. Murtaza, S. Aguilera, V . Azimi, and S. Hutchin- son, “Real-Time Safety and Control of Robotic Manipula- tors with Torque Saturation in Operational Space,”Proc. IEEE/RSJ Int. Conf. on Intelligent Robots and Systems (IROS), pp. 702–708, 2021

2021
[9]

Control Barrier Functions for Singularity Avoidance in Passivity-Based Manipulator Control,

V . Kurtz, P. M. Wensing, and H. Lin, “Control Barrier Functions for Singularity Avoidance in Passivity-Based Manipulator Control,”Proc. IEEE Conf. on Decision and Control (CDC), pp. 6125–6130, 2021

2021
[10]

Ensuring Joint Con- straints of Torque-Controlled Robot Manipulators under Bounded Jerk,

D. Ko, J. Kim, and W. K. Chung, “Ensuring Joint Con- straints of Torque-Controlled Robot Manipulators under Bounded Jerk,”Proc. IEEE/RSJ Int. Conf. on Intelligent Robots and Systems (IROS), pp. 11954–11961, 2024

2024
[11]

Safe, task-consistent manip- ulation with operational space control barrier functions,

D. Morton and M. Pavone, “Safe, task-consistent manip- ulation with operational space control barrier functions,” Proc. IEEE/RSJ Int. Conf. on Intelligent Robots and Sys- tems (IROS), pp. 187–194, 2025

2025
[12]

Multi-layered safety for legged robots via control barrier functions and model predictive control,

R. Grandia, A. J. Taylor, A. D. Ames, and M. Hutter, “Multi-layered safety for legged robots via control barrier functions and model predictive control,”Proc. IEEE Int. Conf. on Robotics and Automation (ICRA), pp. 8352–8358, 2021

2021
[13]

Safety-Critical Coordina- tion for Cooperative Legged Locomotion via Control Bar- rier Functions,

J. Kim, J. Lee, and A. D. Ames, “Safety-Critical Coordina- tion for Cooperative Legged Locomotion via Control Bar- rier Functions,”Proc. IEEE/RSJ Int. Conf. on Intelligent Robots and Systems (IROS), pp. 2368–2375, 2023

2023
[14]

Hu- manoid Self-Collision Avoidance Using Whole-Body Con- trol with Control Barrier Functions,

C. Khazoom, D. Gonzalez-Diaz, Y . Ding, and S. Kim, “Hu- manoid Self-Collision Avoidance Using Whole-Body Con- trol with Control Barrier Functions,”Proc. IEEE-RAS Int. Conf. on Humanoid Robots, pp. 558–565, 2022. Manuscript Template for the International Journal of Control, Automation, and Systems: ICROS & KIEE13

2022
[15]

Safe Whole-Body Task Space Control for Humanoid Robots,

V . C. Paredes and A. Hereid, “Safe Whole-Body Task Space Control for Humanoid Robots,”Proc. American Control Conf. (ACC), pp. 949–956, 2024

2024
[16]

Safety-Critical Manipulation for Collision-Free Food Preparation,

A. Singletary, W. Guffey, T. G. Molnar, R. Sinnet, and A. D. Ames, “Safety-Critical Manipulation for Collision-Free Food Preparation,”IEEE Robotics and Automation Letters, vol. 7, no. 4, pp. 10954–10961, 2022

2022
[17]

Safety- Critical Kinematic Control of Robotic Systems,

A. Singletary, S. Kolathaya, and A. D. Ames, “Safety- Critical Kinematic Control of Robotic Systems,”IEEE Control Systems Letters, vol. 6, pp. 139–144, 2022

2022
[18]

Online Control Barrier Function Construction for Safety- Critical Motion Control of Manipulators,

X. Ding, H. Wang, Y . Ren, Y . Zheng, C. Chen, and H. He, “Online Control Barrier Function Construction for Safety- Critical Motion Control of Manipulators,”IEEE Trans. on Systems, Man, and Cybernetics: Systems, vol. 54, no. 8, pp. 4761–4771, 2024

2024
[19]

Exponential Control Bar- rier Functions for enforcing high relative-degree safety- critical constraints,

Q. Nguyen and K. Sreenath, “Exponential Control Bar- rier Functions for enforcing high relative-degree safety- critical constraints,”Proc. American Control Conf. (ACC), pp. 322–328, 2016

2016
[20]

Robust- ness of control barrier functions for safety-critical control,

X. Xu, P. Tabuada, J. W. Grizzle, and A. D. Ames, “Robust- ness of control barrier functions for safety-critical control,” IF AC-PapersOnLine, vol. 48, no. 27, pp. 54–61, 2015

2015
[21]

Robust Safety-Critical Con- trol for Dynamic Robotics,

Q. Nguyen and K. Sreenath, “Robust Safety-Critical Con- trol for Dynamic Robotics,”IEEE Trans. on Automatic Control, vol. 67, no. 3, pp. 1073–1088, 2022

2022
[22]

Input-to-State Safety With Control Barrier Functions,

S. Kolathaya and A. D. Ames, “Input-to-State Safety With Control Barrier Functions,”IEEE Control Systems Letters, vol. 3, no. 1, pp. 108–113, 2019

2019
[23]

Safety- critical control for autonomous systems: Control barrier functions via reduced-order models,

M. H. Cohen, T. G. Molnar, and A. D. Ames, “Safety- critical control for autonomous systems: Control barrier functions via reduced-order models,”Annual Reviews in Control, vol. 57, p. 100947, 2024

2024
[24]

Model-Free Safety-Critical Con- trol for Robotic Systems,

T. G. Molnar, R. K. Cosner, A. W. Singletary, W. Ubel- lacker, and A. D. Ames, “Model-Free Safety-Critical Con- trol for Robotic Systems,”IEEE Robotics and Automation Letters, vol. 7, no. 2, pp. 944–951, 2022

2022
[25]

Safety-Critical Control with Bounded Inputs via Reduced Order Models,

T. G. Molnar and A. D. Ames, “Safety-Critical Control with Bounded Inputs via Reduced Order Models,”Proc. American Control Conf. (ACC), pp. 1414–1421, 2023

2023
[26]

Safety-critical controller syn- thesis with reduced-order models,

M. H. Cohen, N. Csomay-Shanklin, W. D. Compton, T. G. Molnar, and A. D. Ames, “Safety-critical controller syn- thesis with reduced-order models,”Proc. American Con- trol Conf. (ACC), pp. 5216–5221, 2025

2025
[27]

Control barrier functions and input-to-state safety with application to automated vehicles,

A. Alan, A. J. Taylor, C. R. He, A. D. Ames, and G. Orosz, “Control barrier functions and input-to-state safety with application to automated vehicles,”IEEE Transactions on Control Systems Technology, vol. 31, no. 6, pp. 2744–2759, 2023

2023
[28]

Dynamic locomotion for passive-ankle biped robots and humanoids using whole-body locomotion control,

D. Kim, S. J. Jorgensen, J. Lee, J. Ahn, J. Luo, and L. Sen- tis, “Dynamic locomotion for passive-ankle biped robots and humanoids using whole-body locomotion control,” The International Journal of Robotics Research, vol. 39, no. 8, pp. 936–956, 2020

2020
[29]

Dynamic locomotion in the MIT Cheetah 3 through con- vex model-predictive control,

J. Di Carlo, P. M. Wensing, B. Katz, G. Bledt, and S. Kim, “Dynamic locomotion in the MIT Cheetah 3 through con- vex model-predictive control,”Proc. IEEE/RSJ Int. Conf. on Intelligent Robots and Systems (IROS), pp. 1–9, 2018

2018
[30]

Biped walking stabilization based on linear inverted pendulum tracking,

S. Kajita, M. Morisawa, K. Miura, S. Nakaoka, K. Harada, K. Kaneko, F. Kanehiro, and K. Yokoi, “Biped walking stabilization based on linear inverted pendulum tracking,” Proc. IEEE/RSJ Int. Conf. on Intelligent Robots and Sys- tems (IROS), pp. 4489–4496, 2010

2010
[31]

Gauss’ principle and the dynamics of redundant and constrained manipulators,

H. Bruyninckx and O. Khatib, “Gauss’ principle and the dynamics of redundant and constrained manipulators,” Proc. IEEE Int. Conf. Robot. Autom. (ICRA), vol. 3, pp. 2563–2568, 2000

2000
[32]

Stability of sur- face contacts for humanoid robots: Closed-form formulae of the contact wrench cone for rectangular support areas,

S. Caron, Q.-C. Pham, and Y . Nakamura, “Stability of sur- face contacts for humanoid robots: Closed-form formulae of the contact wrench cone for rectangular support areas,” Proc. IEEE Int. Conf. on Robotics and Automation (ICRA), pp. 5107–5112, 2015

2015
[33]

Detection and tracking of 2D geometric ob- stacles from LRF data,

M. Przybyła, “Detection and tracking of 2D geometric ob- stacles from LRF data,”Proc. 11th Int. Workshop on Robot Motion and Control (RoMoCo), pp. 135–141, 2017

2017
[34]

Design of the Humanoid Robot TOCABI,

M. Schwartz, J. Sim, J. Ahn, S. Hwang, Y . Lee, and J. Park, “Design of the Humanoid Robot TOCABI,”Proc. IEEE- RAS Int. Conf. on Humanoid Robots, pp. 322–329, 2022

2022
[35]

Passivity-based whole- body balancing for torque-controlled humanoid robots in multi-contact scenarios,

B. Henze, M. A. Roa, and C. Ott, “Passivity-based whole- body balancing for torque-controlled humanoid robots in multi-contact scenarios,”The International Journal of Robotics Research, vol. 35, no. 12, pp. 1522–1543, 2016

2016
[36]

MuJoCo: A physics engine for model-based control,

E. Todorov, T. Erez, and Y . Tassa, “MuJoCo: A physics engine for model-based control,”Proc. IEEE/RSJ Int. Conf. on Intelligent Robots and Systems (IROS), pp. 5026–5033, 2012

2012
[37]

qpOASES: A parametric active-set algorithm for quadratic programming,

H. J. Ferreau, C. Kirches, A. Potschka, H. G. Bock, and M. Diehl, “qpOASES: A parametric active-set algorithm for quadratic programming,”Mathematical Programming Computation, vol. 6, no. 4, pp. 327–363, 2014

2014
[38]

RBDL: an efficient rigid-body dynamics library using recursive algorithms,

M. L. Felis, “RBDL: an efficient rigid-body dynamics library using recursive algorithms,”Autonomous Robots, vol. 41, no. 2, pp. 495–511, 2017

2017
[39]

High-Speed and Impact Resilient Teleop- eration of Humanoid Robots,

S. Bertrand, L. Penco, D. Anderson, D. Calvert, V . Roy, S. McCrory, K. Mohammed, S. Sanchez, W. Griffith, S. Morfey, A. Maslyczyk, A. Mohan, C. Castello, B. Ma, K. Suryavanshi, P. Dills, J. Pratt, V . Ragusila, B. Shrewsbury, and R. Griffin, “High-Speed and Impact Resilient Teleop- eration of Humanoid Robots,”Proc. IEEE-RAS Int. Conf. on Humanoid Robots,...

2024
[40]

Zero-moment point— thirty five years of its life,

M. Vukobratovi ´c and B. Borovac, “Zero-moment point— thirty five years of its life,”International Journal of Hu- manoid Robotics, vol. 1, no. 1, pp. 157–173, 2004

2004
[41]

Hub: Learning Extreme Humanoid Balance,

T. Zhang, B. Zheng, R. Nai, Y . Hu, Y .-J. Wang, G. Chen, F. Lin, J. Li, C. Hong, K. Sreenath,et al., “Hub: Learning Extreme Humanoid Balance,”arXiv preprint arXiv:2505.07294, 2025

work page arXiv 2025
[42]

CBF-RL: Safety Filtering Reinforcement Learning in Training with Control Barrier Functions

Y . Yang, B. Werner, M. de Sa, and A. D. Ames, “CBF-RL: Safety Filtering Reinforcement Learning in Training with Control Barrier Functions,”arXiv preprint arXiv:2510.14959, 2025

work page internal anchor Pith review Pith/arXiv arXiv 2025

[1] [1]

A Whole-Body Control Frame- work for Humanoids Operating in Human Environments,

L. Sentis and O. Khatib, “A Whole-Body Control Frame- work for Humanoids Operating in Human Environments,” Proc. 2006 IEEE Int. Conf. on Robotics and Automation (ICRA), pp. 2641–2648, 2006

2006

[2] [2]

Real-Time Obstacle Avoidance for Manipu- lators and Mobile Robots,

O. Khatib, “Real-Time Obstacle Avoidance for Manipu- lators and Mobile Robots,”The International Journal of Robotics Research, vol. 5, no. 1, pp. 90–98, 1986

1986

[3] [3]

Control Barrier Functions: Theory and Applications,

A. D. Ames, S. Coogan, M. Egerstedt, G. Notomista, K. Sreenath, and P. Tabuada, “Control Barrier Functions: Theory and Applications,”Proc. European Control Conf. (ECC), pp. 3420–3431, 2019

2019

[4] [4]

Potential field methods and their inherent limitations for mobile robot navigation,

Y . Koren and J. Borenstein, “Potential field methods and their inherent limitations for mobile robot navigation,” Proc. IEEE Int. Conf. on Robotics and Automation (ICRA), pp. 1398–1404, 1991

1991

[5] [5]

Comparative Analysis of Con- trol Barrier Functions and Artificial Potential Fields for Obstacle Avoidance,

A. Singletary, K. Klingebiel, J. Bourne, A. Browning, P. Tokumaru, and A. Ames, “Comparative Analysis of Con- trol Barrier Functions and Artificial Potential Fields for Obstacle Avoidance,”Proc. IEEE/RSJ Int. Conf. on Intelli- gent Robots and Systems (IROS), pp. 8129–8136, 2021

2021

[6] [6]

Constraint-consistent task-oriented whole-body robot for- mulation: Task, posture, constraints, multiple contacts, and balance,

O. Khatib, M. Jorda, J. Park, L. Sentis, and S.-Y . Chung, “Constraint-consistent task-oriented whole-body robot for- mulation: Task, posture, constraints, multiple contacts, and balance,”The International Journal of Robotics Research, vol. 41, no. 13–14, pp. 1079–1098, 2022

2022

[7] [7]

Con- trol Barrier Function Based Quadratic Programs for Safety Critical Systems,

A. D. Ames, X. Xu, J. W. Grizzle, and P. Tabuada, “Con- trol Barrier Function Based Quadratic Programs for Safety Critical Systems,”IEEE Trans. on Automatic Control, vol. 62, no. 8, pp. 3861–3876, 2017

2017

[8] [8]

Real-Time Safety and Control of Robotic Manipula- tors with Torque Saturation in Operational Space,

M. A. Murtaza, S. Aguilera, V . Azimi, and S. Hutchin- son, “Real-Time Safety and Control of Robotic Manipula- tors with Torque Saturation in Operational Space,”Proc. IEEE/RSJ Int. Conf. on Intelligent Robots and Systems (IROS), pp. 702–708, 2021

2021

[9] [9]

Control Barrier Functions for Singularity Avoidance in Passivity-Based Manipulator Control,

V . Kurtz, P. M. Wensing, and H. Lin, “Control Barrier Functions for Singularity Avoidance in Passivity-Based Manipulator Control,”Proc. IEEE Conf. on Decision and Control (CDC), pp. 6125–6130, 2021

2021

[10] [10]

Ensuring Joint Con- straints of Torque-Controlled Robot Manipulators under Bounded Jerk,

D. Ko, J. Kim, and W. K. Chung, “Ensuring Joint Con- straints of Torque-Controlled Robot Manipulators under Bounded Jerk,”Proc. IEEE/RSJ Int. Conf. on Intelligent Robots and Systems (IROS), pp. 11954–11961, 2024

2024

[11] [11]

Safe, task-consistent manip- ulation with operational space control barrier functions,

D. Morton and M. Pavone, “Safe, task-consistent manip- ulation with operational space control barrier functions,” Proc. IEEE/RSJ Int. Conf. on Intelligent Robots and Sys- tems (IROS), pp. 187–194, 2025

2025

[12] [12]

Multi-layered safety for legged robots via control barrier functions and model predictive control,

R. Grandia, A. J. Taylor, A. D. Ames, and M. Hutter, “Multi-layered safety for legged robots via control barrier functions and model predictive control,”Proc. IEEE Int. Conf. on Robotics and Automation (ICRA), pp. 8352–8358, 2021

2021

[13] [13]

Safety-Critical Coordina- tion for Cooperative Legged Locomotion via Control Bar- rier Functions,

J. Kim, J. Lee, and A. D. Ames, “Safety-Critical Coordina- tion for Cooperative Legged Locomotion via Control Bar- rier Functions,”Proc. IEEE/RSJ Int. Conf. on Intelligent Robots and Systems (IROS), pp. 2368–2375, 2023

2023

[14] [14]

Hu- manoid Self-Collision Avoidance Using Whole-Body Con- trol with Control Barrier Functions,

C. Khazoom, D. Gonzalez-Diaz, Y . Ding, and S. Kim, “Hu- manoid Self-Collision Avoidance Using Whole-Body Con- trol with Control Barrier Functions,”Proc. IEEE-RAS Int. Conf. on Humanoid Robots, pp. 558–565, 2022. Manuscript Template for the International Journal of Control, Automation, and Systems: ICROS & KIEE13

2022

[15] [15]

Safe Whole-Body Task Space Control for Humanoid Robots,

V . C. Paredes and A. Hereid, “Safe Whole-Body Task Space Control for Humanoid Robots,”Proc. American Control Conf. (ACC), pp. 949–956, 2024

2024

[16] [16]

Safety-Critical Manipulation for Collision-Free Food Preparation,

A. Singletary, W. Guffey, T. G. Molnar, R. Sinnet, and A. D. Ames, “Safety-Critical Manipulation for Collision-Free Food Preparation,”IEEE Robotics and Automation Letters, vol. 7, no. 4, pp. 10954–10961, 2022

2022

[17] [17]

Safety- Critical Kinematic Control of Robotic Systems,

A. Singletary, S. Kolathaya, and A. D. Ames, “Safety- Critical Kinematic Control of Robotic Systems,”IEEE Control Systems Letters, vol. 6, pp. 139–144, 2022

2022

[18] [18]

Online Control Barrier Function Construction for Safety- Critical Motion Control of Manipulators,

X. Ding, H. Wang, Y . Ren, Y . Zheng, C. Chen, and H. He, “Online Control Barrier Function Construction for Safety- Critical Motion Control of Manipulators,”IEEE Trans. on Systems, Man, and Cybernetics: Systems, vol. 54, no. 8, pp. 4761–4771, 2024

2024

[19] [19]

Exponential Control Bar- rier Functions for enforcing high relative-degree safety- critical constraints,

Q. Nguyen and K. Sreenath, “Exponential Control Bar- rier Functions for enforcing high relative-degree safety- critical constraints,”Proc. American Control Conf. (ACC), pp. 322–328, 2016

2016

[20] [20]

Robust- ness of control barrier functions for safety-critical control,

X. Xu, P. Tabuada, J. W. Grizzle, and A. D. Ames, “Robust- ness of control barrier functions for safety-critical control,” IF AC-PapersOnLine, vol. 48, no. 27, pp. 54–61, 2015

2015

[21] [21]

Robust Safety-Critical Con- trol for Dynamic Robotics,

Q. Nguyen and K. Sreenath, “Robust Safety-Critical Con- trol for Dynamic Robotics,”IEEE Trans. on Automatic Control, vol. 67, no. 3, pp. 1073–1088, 2022

2022

[22] [22]

Input-to-State Safety With Control Barrier Functions,

S. Kolathaya and A. D. Ames, “Input-to-State Safety With Control Barrier Functions,”IEEE Control Systems Letters, vol. 3, no. 1, pp. 108–113, 2019

2019

[23] [23]

Safety- critical control for autonomous systems: Control barrier functions via reduced-order models,

M. H. Cohen, T. G. Molnar, and A. D. Ames, “Safety- critical control for autonomous systems: Control barrier functions via reduced-order models,”Annual Reviews in Control, vol. 57, p. 100947, 2024

2024

[24] [24]

Model-Free Safety-Critical Con- trol for Robotic Systems,

T. G. Molnar, R. K. Cosner, A. W. Singletary, W. Ubel- lacker, and A. D. Ames, “Model-Free Safety-Critical Con- trol for Robotic Systems,”IEEE Robotics and Automation Letters, vol. 7, no. 2, pp. 944–951, 2022

2022

[25] [25]

Safety-Critical Control with Bounded Inputs via Reduced Order Models,

T. G. Molnar and A. D. Ames, “Safety-Critical Control with Bounded Inputs via Reduced Order Models,”Proc. American Control Conf. (ACC), pp. 1414–1421, 2023

2023

[26] [26]

Safety-critical controller syn- thesis with reduced-order models,

M. H. Cohen, N. Csomay-Shanklin, W. D. Compton, T. G. Molnar, and A. D. Ames, “Safety-critical controller syn- thesis with reduced-order models,”Proc. American Con- trol Conf. (ACC), pp. 5216–5221, 2025

2025

[27] [27]

Control barrier functions and input-to-state safety with application to automated vehicles,

A. Alan, A. J. Taylor, C. R. He, A. D. Ames, and G. Orosz, “Control barrier functions and input-to-state safety with application to automated vehicles,”IEEE Transactions on Control Systems Technology, vol. 31, no. 6, pp. 2744–2759, 2023

2023

[28] [28]

Dynamic locomotion for passive-ankle biped robots and humanoids using whole-body locomotion control,

D. Kim, S. J. Jorgensen, J. Lee, J. Ahn, J. Luo, and L. Sen- tis, “Dynamic locomotion for passive-ankle biped robots and humanoids using whole-body locomotion control,” The International Journal of Robotics Research, vol. 39, no. 8, pp. 936–956, 2020

2020

[29] [29]

Dynamic locomotion in the MIT Cheetah 3 through con- vex model-predictive control,

J. Di Carlo, P. M. Wensing, B. Katz, G. Bledt, and S. Kim, “Dynamic locomotion in the MIT Cheetah 3 through con- vex model-predictive control,”Proc. IEEE/RSJ Int. Conf. on Intelligent Robots and Systems (IROS), pp. 1–9, 2018

2018

[30] [30]

Biped walking stabilization based on linear inverted pendulum tracking,

S. Kajita, M. Morisawa, K. Miura, S. Nakaoka, K. Harada, K. Kaneko, F. Kanehiro, and K. Yokoi, “Biped walking stabilization based on linear inverted pendulum tracking,” Proc. IEEE/RSJ Int. Conf. on Intelligent Robots and Sys- tems (IROS), pp. 4489–4496, 2010

2010

[31] [31]

Gauss’ principle and the dynamics of redundant and constrained manipulators,

H. Bruyninckx and O. Khatib, “Gauss’ principle and the dynamics of redundant and constrained manipulators,” Proc. IEEE Int. Conf. Robot. Autom. (ICRA), vol. 3, pp. 2563–2568, 2000

2000

[32] [32]

Stability of sur- face contacts for humanoid robots: Closed-form formulae of the contact wrench cone for rectangular support areas,

S. Caron, Q.-C. Pham, and Y . Nakamura, “Stability of sur- face contacts for humanoid robots: Closed-form formulae of the contact wrench cone for rectangular support areas,” Proc. IEEE Int. Conf. on Robotics and Automation (ICRA), pp. 5107–5112, 2015

2015

[33] [33]

Detection and tracking of 2D geometric ob- stacles from LRF data,

M. Przybyła, “Detection and tracking of 2D geometric ob- stacles from LRF data,”Proc. 11th Int. Workshop on Robot Motion and Control (RoMoCo), pp. 135–141, 2017

2017

[34] [34]

Design of the Humanoid Robot TOCABI,

M. Schwartz, J. Sim, J. Ahn, S. Hwang, Y . Lee, and J. Park, “Design of the Humanoid Robot TOCABI,”Proc. IEEE- RAS Int. Conf. on Humanoid Robots, pp. 322–329, 2022

2022

[35] [35]

Passivity-based whole- body balancing for torque-controlled humanoid robots in multi-contact scenarios,

B. Henze, M. A. Roa, and C. Ott, “Passivity-based whole- body balancing for torque-controlled humanoid robots in multi-contact scenarios,”The International Journal of Robotics Research, vol. 35, no. 12, pp. 1522–1543, 2016

2016

[36] [36]

MuJoCo: A physics engine for model-based control,

E. Todorov, T. Erez, and Y . Tassa, “MuJoCo: A physics engine for model-based control,”Proc. IEEE/RSJ Int. Conf. on Intelligent Robots and Systems (IROS), pp. 5026–5033, 2012

2012

[37] [37]

qpOASES: A parametric active-set algorithm for quadratic programming,

H. J. Ferreau, C. Kirches, A. Potschka, H. G. Bock, and M. Diehl, “qpOASES: A parametric active-set algorithm for quadratic programming,”Mathematical Programming Computation, vol. 6, no. 4, pp. 327–363, 2014

2014

[38] [38]

RBDL: an efficient rigid-body dynamics library using recursive algorithms,

M. L. Felis, “RBDL: an efficient rigid-body dynamics library using recursive algorithms,”Autonomous Robots, vol. 41, no. 2, pp. 495–511, 2017

2017

[39] [39]

High-Speed and Impact Resilient Teleop- eration of Humanoid Robots,

S. Bertrand, L. Penco, D. Anderson, D. Calvert, V . Roy, S. McCrory, K. Mohammed, S. Sanchez, W. Griffith, S. Morfey, A. Maslyczyk, A. Mohan, C. Castello, B. Ma, K. Suryavanshi, P. Dills, J. Pratt, V . Ragusila, B. Shrewsbury, and R. Griffin, “High-Speed and Impact Resilient Teleop- eration of Humanoid Robots,”Proc. IEEE-RAS Int. Conf. on Humanoid Robots,...

2024

[40] [40]

Zero-moment point— thirty five years of its life,

M. Vukobratovi ´c and B. Borovac, “Zero-moment point— thirty five years of its life,”International Journal of Hu- manoid Robotics, vol. 1, no. 1, pp. 157–173, 2004

2004

[41] [41]

Hub: Learning Extreme Humanoid Balance,

T. Zhang, B. Zheng, R. Nai, Y . Hu, Y .-J. Wang, G. Chen, F. Lin, J. Li, C. Hong, K. Sreenath,et al., “Hub: Learning Extreme Humanoid Balance,”arXiv preprint arXiv:2505.07294, 2025

work page arXiv 2025

[42] [42]

CBF-RL: Safety Filtering Reinforcement Learning in Training with Control Barrier Functions

Y . Yang, B. Werner, M. de Sa, and A. D. Ames, “CBF-RL: Safety Filtering Reinforcement Learning in Training with Control Barrier Functions,”arXiv preprint arXiv:2510.14959, 2025

work page internal anchor Pith review Pith/arXiv arXiv 2025