Control Barrier Functions With Real-Time Gaussian Process Modeling

Jesse B. Hoagg; Ricardo Gutierrez

REVIEW 3 major objections 4 minor 30 references

Control Barrier Functions With Real-Time Gaussian Process Modeling

T0 review · 3 major / 4 minor · reviewed 2026-08-15 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read A fixed-memory Gaussian-process model can be updated in real time while retaining a deterministic safety guarantee for control systems with unknown dynamics.

desk verdict Useful fixed-memory streaming GP-CBF scheme, but main theorems are unproved, the data-update rule has a labeling inconsistency, and examples never verify the RKHS bound the safety guarantee rests on. read the letter →

arxiv 2505.06765 v1 pith:EA2KOEPT submitted 2025-05-10 eess.SY cs.SY

Ricardo Gutierrez , Jesse B. Hoagg This is my paper · ORCID

classification eess.SYcs.SY

keywords controlbarrierfunctionsGaussianprocessregressionreal-timelearningrecursiveupdatenonparametricuncertaintyforwardinvariancestateconstraintskernelmethods

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

This paper tries to establish that state constraints can be guaranteed for a nonlinear control system with an unknown, nonparametric disturbance by maintaining a Gaussian-process model of that disturbance in real time with fixed memory. New measurements enter the model at every sampling step and selected old measurements are dropped, so the model never grows; a recursive update computes the needed GP quantities in $O(p^2)$ time instead of the batch $O(p^3)$. The GP mean and a deterministic error bound are fed into a control-barrier-function constraint, and the resulting closed-form minimum-intervention control keeps the state in the safe set for all time on the maximal interval of existence. The authors demonstrate the method on a nonlinear pendulum and a nonholonomic ground robot, showing that using the adaptive estimate in both the safety filter and the desired control improves tracking while preserving safety.

What carries the argument

The load-bearing mechanism is the pair formed by a fixed-memory data-update rule and the recursive formulas (18)--(30) that propagate $\Omega(X_k)^{-1}$, $\Omega(X_k)^{-1}Y_k$, and $P(X_k)\mathbf{1}_p$ without matrix inversion. At each sampling time the newest measurement is added, one least-weight local point is reclassified and one most-correlated point is discarded, keeping exactly $p$ points; this keeps memory fixed and makes each update $O(p^2)$. On top of that, the control-barrier-function constraint $\psi(x,\hat\mu,\hat\varphi,\hat u,\hat\delta)\ge 0$, with $\hat\varphi \ge |w-\hat\mu|$, is a guaranteed lower bound for the ideal CBF constraint that would use the true $w$, so enforcing the computable constraint enforces safety; Proposition 3 makes this comparison explicit, and the closed-form minimizer $u^*,\delta^*$ resolves the constrained quadratic program at each instant.

What would settle it

Run the pendulum example's control law with the same $b=100$ but replace the simulated $w_2$ by a function whose RKHS norm under the kernel $q(x_i,x_j)=100\exp(-0.5\|x_i-x_j\|_2^2)$ is larger than $100$; if the state leaves $\bar{C}$ or $|(\mu-w)_2|$ exceeds $(\varphi)_2$, the premise of Proposition 1 fails in exactly the regime the theorem requires to be checked. Alternatively, compute the RKHS norm of the paper's own $w_2$ under that kernel and compare it with $b=100$.

Watch

Extended reading notes

Core claim

The central claim is that a control-barrier-function safety filter can be built on a Gaussian-process model that is updated online with a sliding window of exactly $p$ data points, and that the update can be done recursively in $O(p^2)$ time while preserving a deterministic safety guarantee. Specifically, with $\Sigma_k = \Omega(X_k)^{-1}$, $\vartheta_k = \Omega(X_k)^{-1}Y_k$, and $\varsigma_k = P(X_k)\mathbf{1}_p$ maintained by the recursive equations, the predictive mean, standard deviation, and error-bound factor are available at each step; Proposition 1 bounds $|(\mu - w)_i|$ by $(B_k \sigma_k)_i$ under the assumption that each component of $w$ lies in the kernel's RKHS with known norm bound $b$. Control $u^*$ from (37)--(41) minimizes the quadratic cost subject to the CBF constraint $\psi \ge 0$, and Theorem 2 states that for every $x_0 \in \bar{C}$ the closed-loop state remains in $\bar{C}$ on the maximal interval of existence, with the guarantee holding for all $t\ge 0$ whenever that interval is infinite. The pendulum and robot simulations illustrate that the adaptive estimate and bound reduce conservatism compared with using fixed initial estimates.

Load-bearing premise

The guarantee collapses if the chosen bound $b$ is not a true upper bound on the RKHS norms of the unknown disturbance components; the examples simply set $b = 100$ without verifying it, so a real disturbance with larger norm could break the error bound (10) and the safety argument that leans on it.

Editorial extensions

If this is right

For any initial state in the safe set, the closed-loop system stays in the safe set on the maximal interval of existence despite unknown nonparametric dynamics, provided the RKHS norm bound holds.
The model memory and per-update cost stay constant as data stream in: each update is $O(p^2)$ with exactly $p$ stored points, so the method can run indefinitely without growing storage.
The controller is the closed-form minimizer of the quadratic cost subject to safety, so it intervenes only when needed; in the pendulum and robot examples the adaptive estimate lets the desired control track well without constant filter activation.
Because the error bound is deterministic rather than probabilistic, the safety guarantee does not rely on confidence levels or chance constraints.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

The same recursive inverse-Gram-matrix update could be reused for other online kernel methods, such as online regression or active learning, though the paper does not explore those applications.
The data-selection rule splits data into local and global sets with fixed sizes $p_l$ and $p_g$; a natural testable extension is to adapt $p_l$ online based on how fast the state moves, since the paper sets these counts by hand.
If the true RKHS norm is unknown, one could try to estimate $b$ online from observed data, for example by inflating it with the empirical quantity $\vartheta_k^T Y_k$; the paper assumes $b$ is given, so bounding $b$ from data remains an open step.
The safety theorem covers the maximal interval of existence, and the paper states the all-$t\ge0$ guarantee only when that interval is infinite; turning this condition into a global existence result, for instance via boundedness of $u^*$, would be the next step toward deployment.

Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

Dear colleague,

Short version: this paper has a genuinely useful idea—fixed-memory streaming GP with recursive O(p^2) updates—but the manuscript as written doesn't fully support the safety theorem it claims. The core mechanism is plausible; the proof gaps and the unverified RKHS bound are fixable, but they need to be fixed before I'd rely on the result.

What's new: prior GP-CBF work uses static datasets or streaming with growing complexity. Here the authors keep p data points, add the newest point, remove one selected point, and update the inverse covariance, the coefficient vector, and the column-sum vector in O(p^2) via (18)–(30). That's a real algorithmic contribution, and the selection rule—drop the least-weighted local point and the most-correlated point from the nonlocal pool—is sensible. The two simulation studies support the qualitative claim that the adaptive estimate/error bound reduces conservatism relative to using the initial GP.

Soft spots, in order of concern. First, the central results are statements without proofs. Proposition 2 (the recursive inverse update), Theorem 1 (the QP minimizer), and Theorem 2 (the safety guarantee) are all "omitted for space." For a theorem-driven paper, that's a lot to take on faith, especially since the recursive update is the main contribution and a small algebraic mistake there would invalidate everything. Second, the data-update rule appears internally inconsistent. The c-vector is defined with 0=local, 1=nonlocal, and the newest data gets c=1 in (11), yet the text says the newest data is added to improve the local model. Also l_k in (15) is allowed to be j_k, which would shrink the local pool. This may be a typo, but as written the fixed p_l/p_g partition is not preserved. Third, the safety guarantee depends on Assumption 3 (known RKHS norm bound b) and the examples set b=100 by hand without any check. The paper does not report the finite-sample lower bound Y^T Ω(X)^{-1}Y, which would at least be consistent with b. The plots show |µ−w| ≤ φ along the simulated trajectories, but that is not a certificate for all reachable states, and the controller is itself built from φ.

I don't think these are fatal to the approach. The recursive update is likely correct—the formulas look like standard block-matrix manipulations—and the safety argument follows the usual CBF logic if Proposition 1 and Assumption 3 hold. But a paper whose main theorems are all unproved and whose examples don't verify a load-bearing bound should not be taken as fully established.

Who's this for: researchers working on safe learning-based control, especially GP-CBF with streaming data. It deserves a serious referee, not a desk reject, but the referee should insist on proofs or clear references for the main results and at least a discussion of how b could be verified or estimated in practice.

Best, [Your name]

Referee Report

3 major / 4 minor

Summary. The manuscript proposes a real-time Gaussian process (GP) modeling framework for use in control barrier function (CBF) based safety filters. New measurements are added to the GP dataset and selected old measurements are removed, keeping the dataset size fixed; a recursive update is claimed to compute the required GP quantities in O(p^2) time instead of O(p^3). The GP predictive mean is used in the control law and the GP error bound is used to construct a sufficient CBF constraint. The authors state a safety theorem (Theorem 2) and illustrate the approach in two simulations: an inverted pendulum and a nonholonomic mobile robot.

Significance. If the recursive update and the safety theorem are correct, the paper addresses a practical bottleneck in GP-CBF control: maintaining a fixed-size streaming GP model with bounded computational cost. The idea of combining a deterministic GP error bound with a CBF constraint is a natural and potentially useful extension of existing static-dataset GP-CBF methods. The paper also gives explicit examples with performance comparisons across three cases, which are informative. However, the central claims are currently supported mainly by statements without proofs, and the examples do not verify the key RKHS norm bound used by the safety theorem. The contribution is therefore promising but not yet established at the level required for publication.

major comments (3)

[§VII–§VIII, Assumption 3, Prop. 1, Thm. 2] The safety guarantee in Theorem 2 rests on Proposition 1, which is valid only if Assumption 3 holds with the chosen bound b. In both examples, b=100 is selected by hand without verifying that the true RKHS norms of w2 (pendulum) or w4 and w5 (robot) are bounded by 100. The simulation plots showing |μ−w|≤φ along the closed-loop trajectory are not sufficient, because they cover only the realized trajectory and because the controller is itself designed using φ. The authors should verify Assumption 3 for the specific kernels and functions used, for example by computing or conservatively bounding the RKHS norm or by checking the finite-sample lower bound Y^TΩ(X)^{-1}Y against b^2. Without such verification, the examples do not demonstrate the hypotheses of Theorem 2.
[§V-A, Eqs. (11)–(15)] The update rule for the data partition appears internally inconsistent. The text states that c_k[i]=0 marks data used for the local model and that the newest data are added to improve the local model, yet Eq. (11) appends the new data with c_{k+1}=1, which is the nonlocal label. In addition, if l_k in Eq. (15) is chosen as a nonlocal index, then removing that point and appending a point with c=1 changes the total number of local points, so the fixed p_l/p_g partition described in the text is not preserved. Please either correct the labeling or revise the algorithm so that the local/nonlocal counts are fixed and the description matches the equations.
[§V-B and §VI, Prop. 2, Prop. 3, Thm. 1, Thm. 2] The main new technical results are stated without proof. Proposition 2, which is the basis for the claimed O(p^2) recursive update and for the correctness of the updated GP quantities, is asserted with the proof omitted. Theorem 2, the main safety certificate, is also stated without proof, and its existence/uniqueness part requires regularity of u* that is not verified. Proposition 3 and Theorem 1 are delegated to prior work with only a sentence indicating similarity. Since the recursive update is not a standard CBF result and no reference is provided for Proposition 2, the manuscript does not currently substantiate its central correctness and complexity claims. Please include complete proofs, or precise statements and complete proofs of the specific referenced results as used here.

minor comments (4)

[§III] The word 'Lipshchitz' should be 'Lipschitz'.
[§VI, Thm. 2] Theorem 2 assumes that 'h′0 is locally Lipschitz', but h0 is not defined in the manuscript; it is presumably ψ0 and should be stated as such.
[Eqs. (14)–(15) and (29)–(30)] The argmin and argmax operations need a tie-breaking rule, since multiple indices may attain the extremal value.
[Prop. 3(b)] The statement 'For all t≥0, ψ(x,μ(t,x),φ(t,x),...)≤ψ*(...)' should quantify x explicitly, e.g., 'for all x∈R^n and all t≥0'.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the safety guarantee is conditional on stated assumptions and external GP error-bound results; self-citations appear only for omitted proofs of standard/elementary lemmas.

full rationale

The claimed derivation chain is not circular. Assumption 3 supplies a norm bound b, and Proposition 1 (cited to [17]) converts it into the deterministic error bound (10) with B defined in (9). This is an externally sourced conditional theorem, not a fitted prediction; choosing b=100 in the examples without computing the RKHS norm is a validation gap, not circularity. The recursive update (18)-(30) is asserted in Proposition 2 as algebraic identities Sigma_k = Omega(X_k)^{-1}, theta_k = Omega(X_k)^{-1} Y_k, and s_k = P(X_k) 1_p; although the proof is omitted, this is a direct block-inversion/rank-one update identity and is self-contained algebra against definitions (5)-(6). Section VI's Proposition 3 is an elementary comparison: if (phi_hat)_i >= |(mu_hat - w)_i|, then mu_hat - |dpsi/dx| phi_hat <= dpsi/dx * w componentwise, so psi <= psi*; the citation to [10, Prop. 4] is a proof-style reference, and the statement does not assume the paper's safety conclusion. Theorem 1 is the standard KKT characterization of a strictly convex quadratic program with one linear constraint, cited to [26] for proof similarity; Theorem 2 assembles the bound, Proposition 3, and the standard CBF forward-invariance argument [6]. None of these steps reduces the theorem to its inputs: phi is not fitted to the safety violation it certifies, b is a prior assumption rather than a calibrated parameter, and the omitted proofs are for standard lemmas rather than the paper's novel claim. The explicit 'proof omitted for space' statements are noted, but absence of an included proof is not circularity under the stated rules.

Assumptions & free parameters 5 free parameters · 5 assumptions · 0 invented entities

The theoretical guarantee rests on three domain assumptions specific to the system (relative degree, boundary control authority, RKHS norm bound) and on external standard results (GP error bound and CBF invariance). No new physical entities are introduced. The numerical examples additionally depend on hand-chosen tuning parameters (b, rho, kernel, p, p_l, T_s), which are not fitted to data but affect the achieved performance and the validity of the assumed bounds.

free parameters (5)

RKHS norm bound b = 100 in both examples
Assumption 3 requires a known b such that ||(w)[i]||_q <= b for all i. The examples set b=100 by hand without verifying the true norm; the safety guarantee depends on b being a valid upper bound.
Noise bound rho = 1 (pendulum), 0.5 (robot)
The measurement model y_j = w(x_j) + nu assumes ||nu||_infty <= rho. Chosen by hand; affects the regularization term rho^2 I_p in Omega and the error bound.
Kernel amplitude and length scale = 100 exp(-0.5||x-x'||^2) and 100 exp(-0.1||x-x'||^2)
The GP kernel hyperparameters are chosen by hand for each example. They determine the RKHS and the quality of the GP estimate, but are not fitted from data.
Data set size p and local fraction p_l = p=100, p_l=50
The fixed memory size and the number of local points are user-chosen. They set the computational complexity O(p^2) and trade off local accuracy versus global retention.
Sampling time T_s = 1 ms
Controls how often new data is incorporated and how often the control is updated. Chosen for simulation, not dictated by the theory.

assumptions (5)

domain assumption Assumption 1: There exists d such that for all i in {0,...,d-2} and all x, L_g L_f^i psi0(x)=0 and L_w L_f^i psi0(x)=0.
This relative-degree condition is required for the higher-order CBF construction in Section III. If the unknown dynamics w enters with relative degree lower than d, equation (34) is not a valid CBF condition.
domain assumption Assumption 2: For all x in bd C_{d-1}, L_g L_f^{d-1} psi0(x) != 0.
Ensures the safe set boundary is controllable through u, which is needed for the closed-form QP solution and for forward invariance.
domain assumption Assumption 3: (w)[i] is in the RKHS H_q and ||(w)[i]||_q <= b for all i.
The deterministic error bound in Proposition 1 comes from this assumption. The safety guarantee is void if b is not a valid norm bound for the true unknown function.
standard math Proposition 1 (external): |(mu(x,X,Y)-w(x))[i]| <= (B(X,Y)sigma(x,X))[i] for all x,i.
Borrowed from [17], this bound is the bridge between the GP estimate and the sufficient CBF constraint. The paper does not prove it, only cites it.
standard math Standard CBF forward invariance: if psi*(x(t),u,delta) >= 0 for all t, then x(t) stays in the safe set C bar.
Used without proof in Theorem 2 as the standard CBF analysis from [6]. This is the mechanism by which the local constraint psi >= 0 implies global safety.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Control Barrier Functions With Real-Time Gaussian Process Modeling." pith.science (2026). https://pith.science/paper/EA2KOEPT

@misc{pith2026250506765,
  author       = {Pith},
  title        = {Pith review of: Control Barrier Functions With Real-Time Gaussian Process Modeling},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/EA2KOEPT}},
  note         = {Machine review of arXiv:2505.06765}
}

read the original abstract

We present an approach for satisfying state constraints in systems with nonparametric uncertainty by estimating this uncertainty with a real-time-update Gaussian process (GP) model. Notably, new data is incorporated into the model in real time as it is obtained and select old data is removed from the model. This update process helps improve the model estimate while keeping the model size (memory required) and computational complexity fixed. We present a recursive formulation for the model update, which reduces time complexity of the update from O(p3) to O(p2), where p is the number of data used. The GP model includes a computable upper bound on the model error. Together, the model and upper bound are used to construct a control-barrier-function (CBF) constraint that guarantees state constraints are satisfied.

Figures

Figures reproduced from arXiv: 2505.06765 by the authors.

**Figure 1.** γ, γ˙ and u for Cases 1, 2 and 3, Note that γd, γ˙d and ud are shown with dashed lines. 0 0.2 0.4 0.6 A 0 Case 1 Case 2 Case 3 0 20 40 t (s) 0 40 80 120 A1 0 20 40 t (s) 0 20 40 t (s) [PITH_FULL_IMAGE:figures/full_fig_p005_1.png] view at source ↗

**Figure 2.** ψ0 and ψ1 for Cases 1, 2 and 3 [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗

**Figure 3.** ψ for Cases 1, 2 and 3. 0 20 40 t (s) -6 -3 0 3 6 7[2] (r a d = s 2) #101 Case 1 0 20 40 t (s) Case 2 0 20 40 t (s) Case 3 [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗

Figures from the paper (7 more)

**Figure 4.** Figure 4: µ[2] for Cases 1, 2 and 3. Note that w[2] is shown using dashed line. 0 20 40 t (s) 10-6 10-2 102 (')[2] Case 1 0 20 40 t (s) Case 2 0 20 40 t (s) Case 3 [PITH_FULL_IMAGE:figures/full_fig_p006_4.png]

**Figure 5.** Figure 5: (φ)[2] for Cases 1, 2 and 3. Note that |(µ − w)[2]| is shown with dashed lines. VIII. NONHOLONOMIC GROUND ROBOT Consider the nonholonomic differential drive mobile robot modeled by (1), where f(x) =       v cos γ − ldω sin γ v sin γ + ldω cos γ ω 0 0       …

**Figure 6.** Figure 6: Phase-Portrait of 3 closed-loop trajectories for Cases [PITH_FULL_IMAGE:figures/full_fig_p007_6.png]

**Figure 7.** Figure 7: qx, qy, γ, v and ω for Cases 1, 2 and 3. Note that qdx, qdy and ωd are shown with dashed lines. [7] Q. Nguyen and K. Sreenath, “Exponential control barrier functions for enforcing high relative-degree safety-critical constraints,” in Proc. Amer. Contr. Conf., pp. 322–3…

**Figure 11.** Figure 11: (φ) [4] and (φ) [5] for Cases 1, 2 and 3. Note that |(µ − w)[4]| and |(µ − w)[5]| are shown with dashed lines. IEEE Contr. Syst. Letts., vol. 5, no. 3, pp. 1031–1036, 2020. [15] M. H. Cohen, C. Belta, and R. Tron, “Robust control barrier functions for nonlinear contro…

**Figure 9.** Figure 9: min ϕj,0, min ϕj,1, ψ0 and ψ for Cases 1, 2 and 3. -4 -2 0 2 4 7[4] (m = s 2) Case 1 Case 2 Case 3 0 20 40 60 t (s) -2 0 2 7[5] (r a d = s 2) 0 20 40 60 t (s) 0 20 40 60 t (s) [PITH_FULL_IMAGE:figures/full_fig_p008_9.png]

**Figure 10.** Figure 10: µ[4] and µ[5] for Cases 1, 2 and 3. Note that w[4] and w[5] are shown with dashed lines. 2018. [12] M. H. Cohen and C. Belta, “High order robust adaptive control barrier functions and exponentially stabilizing adaptive control lyapunov functions,” in Proc. Amer. Contr…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

30 extracted references · 14 canonical work pages

[1]

Borrelli, A

F. Borrelli, A. Bemporad, and M. Morari, Predictive control for linear and hybrid systems . Cambridge University Press, 2017

2017
[2]

Model predictive control based on linear programming˜ the explicit solution,

A. Bemporad, F. Borrelli, M. Morari, et al. , “Model predictive control based on linear programming˜ the explicit solution,” IEEE Trans. Autom. Contr ., vol. 47, no. 12, pp. 1974–1985, 2002

1974
[3]

An algorithm for multi-parametric quadratic programming and explicit mpc solutions,

P. Tøndel, T. A. Johansen, and A. Bemporad, “An algorithm for multi-parametric quadratic programming and explicit mpc solutions,” Automatica, vol. 39, no. 3, pp. 489–497, 2003

2003
[4]

A framework for worst-case and stochastic safety verification using barrier certificates,

S. Prajna, A. Jadbabaie, and G. J. Pappas, “A framework for worst-case and stochastic safety verification using barrier certificates,” IEEE Trans. Autom. Contr ., vol. 52, no. 8, pp. 1415–1428, 2007

2007
[5]

Constructive safety using control barrier functions,

P. Wieland and F. Allg ¨ower, “Constructive safety using control barrier functions,” IF AC Proc., vol. 40, no. 12, pp. 462–467, 2007

work page 2007
[6]

Control barrier function based quadratic programs for safety critical systems,

A. D. Ames, X. Xu, J. W. Grizzle, and P. Tabuada, “Control barrier function based quadratic programs for safety critical systems,” IEEE Trans. Autom. Contr ., vol. 62, no. 8, pp. 3861–3876, 2016. -1 0 1 2 3 4qx (m) -1 0 1 2 3qy (m) ObstacleCase1 (qx(0);qy(0))Case2 qd Case3 Fig. 6: Phase-Portrait of 3 closed-loop trajectories for Cases 1, 2 and 3. The nota...

work page 2016
[7]

7: qx, qy, γ, v and ω for Cases 1, 2 and 3

(rad) 0 1v (m=s) 0 20 40 60t (s) -2 0 2 4! (rad=s) 0 20 40 60t (s) 0 20 40 60t (s) Fig. 7: qx, qy, γ, v and ω for Cases 1, 2 and 3. Note that qdx, qdy and ωd are shown with dashed lines

work page
[8]

Exponential control barrier functions for enforcing high relative-degree safety-critical constraints,

Q. Nguyen and K. Sreenath, “Exponential control barrier functions for enforcing high relative-degree safety-critical constraints,” in Proc. Amer . Contr . Conf., pp. 322–328, IEEE, 2016

work page 2016

Show all 30 references

[9]

High-order control barrier functions,

W. Xiao and C. Belta, “High-order control barrier functions,” IEEE Trans. Autom. Contr ., vol. 67, no. 7, pp. 3655–3662, 2021

2021
[10]

High-order barrier functions: Robustness, safety, and performance-critical control,

X. Tan, W. S. Cortez, and D. V . Dimarogonas, “High-order barrier functions: Robustness, safety, and performance-critical control,” IEEE Trans. Autom. Contr ., vol. 67, no. 6, pp. 3021–3028, 2021

2021
[11]

Adaptive control barrier functions with vanishing conservativeness under persistency of excitation,

R. Gutierrez and J. B. Hoagg, “Adaptive control barrier functions with vanishing conservativeness under persistency of excitation,” arXiv preprint arXiv:2411.12899, 2024

2024 arXiv
[12]

Robust control barrier functions for constrained sta- bilization of nonlinear systems,

M. Jankovic, “Robust control barrier functions for constrained sta- bilization of nonlinear systems,” Automatica, vol. 96, pp. 359–367, -40 0 40ur (V) Case 1 Case 2 Case 3 0 20 40 60t (s) -40 0 40ul (V) 0 20 40 60t (s) 0 20 40 60t (s) Fig. 8: ur and ul for Cases 1, 2 and 3. No...

2018
[13]

High order robust adaptive control barrier functions and exponentially stabilizing adaptive control lyapunov functions,

M. H. Cohen and C. Belta, “High order robust adaptive control barrier functions and exponentially stabilizing adaptive control lyapunov functions,” in Proc. Amer . Contr . Conf., pp. 2233–2238, IEEE, 2022

2022
[14]

Adaptive safety with control barrier functions,

A. J. Taylor and A. D. Ames, “Adaptive safety with control barrier functions,” in Proc. Amer . Contr . Conf., pp. 1399–1405, IEEE, 2020

2020
[15]

Robust adaptive control barrier functions: An adaptive and data-driven approach to safety,

B. T. Lopez, J.-J. E. Slotine, and J. P. How, “Robust adaptive control barrier functions: An adaptive and data-driven approach to safety,” 10-6 10-2 102 (')[4] Case 1 Case 2 Case 3 0 20 40 60t (s) 10-6 10-2 102 (')[5] 0 20 40 60t (s) 0 20 40 60t (s) Fig. 11: (φ)[4] and (φ)[5] ...

2020
[16]

Robust control barrier functions for nonlinear control systems with uncertainty: A duality-based approach,

M. H. Cohen, C. Belta, and R. Tron, “Robust control barrier functions for nonlinear control systems with uncertainty: A duality-based approach,” in Proc. Conf. Dec. Contr ., pp. 174–179, IEEE, 2022

2022
[17]

Nonparametric adaptive control and prediction: Theory and randomized algorithms,

N. M. Boffi, S. Tu, and J.-J. E. Slotine, “Nonparametric adaptive control and prediction: Theory and randomized algorithms,” Journal of Machine Learning Research , vol. 23, no. 281, pp. 1–46, 2022

2022
[18]

Learning-based symbolic abstractions for nonlinear control systems,

K. Hashimoto, A. Saoud, M. Kishida, T. Ushio, and D. V . Dimarogonas, “Learning-based symbolic abstractions for nonlinear control systems,” Automatica, vol. 146, p. 110646, 2022

2022
[19]

Error bounds for gaussian process regression under bounded support noise with applications to safety certification,

R. Reed, L. Laurenti, and M. Lahijanian, “Error bounds for gaussian process regression under bounded support noise with applications to safety certification,” arXiv preprint arXiv:2408.09033 , 2024

2024 arXiv
[20]

Control barrier functions for unknown nonlinear systems using gaussian processes,

P. Jagtap, G. J. Pappas, and M. Zamani, “Control barrier functions for unknown nonlinear systems using gaussian processes,” in Proc. Conf. Dec. Contr ., pp. 3699–3704, IEEE, 2020

2020
[21]

Safe multi-agent interaction through robust control barrier functions with learned uncertainties,

R. Cheng, M. J. Khojasteh, A. D. Ames, and J. W. Burdick, “Safe multi-agent interaction through robust control barrier functions with learned uncertainties,” in Proc. Conf. Dec. Contr ., pp. 777–783, IEEE, 2020

2020
[22]

Gaussian control barrier functions: Safe learning and control,

M. Khan and A. Chatterjee, “Gaussian control barrier functions: Safe learning and control,” in Proc. Conf. Dec. Contr ., pp. 3316–3322, IEEE, 2020

2020
[23]

Safety uncertainty in control barrier functions using gaussian processes,

M. Khan, T. Ibuki, and A. Chatterjee, “Safety uncertainty in control barrier functions using gaussian processes,” in Proc. IEEE Int. Conf. Rob. Auto., pp. 6003–6009, IEEE, 2021

2021
[24]

Gaussian process-based real-time learning for safety critical applications,

A. Lederer, A. J. O. Conejo, K. A. Maier, W. Xiao, J. Umlauft, and S. Hirche, “Gaussian process-based real-time learning for safety critical applications,” in Int. Conf. Mach. Learn. , pp. 6055–6064, PMLR, 2021

2021
[25]

C. K. Williams and C. E. Rasmussen, Gaussian processes for machine learning, vol. 2. MIT press Cambridge, MA, 2006

2006
[26]

Deterministic error bounds for kernel-based learning techniques under bounded noise,

E. T. Maddalena, P. Scharnhorst, and C. N. Jones, “Deterministic error bounds for kernel-based learning techniques under bounded noise,” Automatica, vol. 134, p. 109896, 2021

2021
[27]

Time-varying soft-maximum barrier functions for safety in unmapped and dynamic environments,

A. Safari and J. B. Hoagg, “Time-varying soft-maximum barrier functions for safety in unmapped and dynamic environments,” arXiv preprint arXiv:2409.01458, 2024

2024 arXiv
[28]

Non-holonomic differential drive mobile robot control & design: Critical dynamics and coupling constraints,

I. Anvari, “Non-holonomic differential drive mobile robot control & design: Critical dynamics and coupling constraints,” tech. rep., Arizona State University, 2013

2013
[29]

Soft-minimum and soft-maximum barrier functions for safety with actuation constraints,

P. Rabiee and J. B. Hoagg, “Soft-minimum and soft-maximum barrier functions for safety with actuation constraints,” Automatica, vol. 171, p. 111921, 2025

2025
[30]

A closed-form control for safety under input constraints using a composition of control barrier functions,

P. Rabiee and J. B. Hoagg, “A closed-form control for safety under input constraints using a composition of control barrier functions,” arXiv preprint arXiv:2406.16874 , 2024

2024 arXiv

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.

[1] [1]

Borrelli, A

F. Borrelli, A. Bemporad, and M. Morari, Predictive control for linear and hybrid systems . Cambridge University Press, 2017

2017

[2] [2]

Model predictive control based on linear programming˜ the explicit solution,

A. Bemporad, F. Borrelli, M. Morari, et al. , “Model predictive control based on linear programming˜ the explicit solution,” IEEE Trans. Autom. Contr ., vol. 47, no. 12, pp. 1974–1985, 2002

1974

[3] [3]

An algorithm for multi-parametric quadratic programming and explicit mpc solutions,

P. Tøndel, T. A. Johansen, and A. Bemporad, “An algorithm for multi-parametric quadratic programming and explicit mpc solutions,” Automatica, vol. 39, no. 3, pp. 489–497, 2003

2003

[4] [4]

A framework for worst-case and stochastic safety verification using barrier certificates,

S. Prajna, A. Jadbabaie, and G. J. Pappas, “A framework for worst-case and stochastic safety verification using barrier certificates,” IEEE Trans. Autom. Contr ., vol. 52, no. 8, pp. 1415–1428, 2007

2007

[5] [5]

Constructive safety using control barrier functions,

P. Wieland and F. Allg ¨ower, “Constructive safety using control barrier functions,” IF AC Proc., vol. 40, no. 12, pp. 462–467, 2007

work page 2007

[6] [6]

Control barrier function based quadratic programs for safety critical systems,

A. D. Ames, X. Xu, J. W. Grizzle, and P. Tabuada, “Control barrier function based quadratic programs for safety critical systems,” IEEE Trans. Autom. Contr ., vol. 62, no. 8, pp. 3861–3876, 2016. -1 0 1 2 3 4qx (m) -1 0 1 2 3qy (m) ObstacleCase1 (qx(0);qy(0))Case2 qd Case3 Fig. 6: Phase-Portrait of 3 closed-loop trajectories for Cases 1, 2 and 3. The nota...

work page 2016

[7] [7]

7: qx, qy, γ, v and ω for Cases 1, 2 and 3

(rad) 0 1v (m=s) 0 20 40 60t (s) -2 0 2 4! (rad=s) 0 20 40 60t (s) 0 20 40 60t (s) Fig. 7: qx, qy, γ, v and ω for Cases 1, 2 and 3. Note that qdx, qdy and ωd are shown with dashed lines

work page

[8] [8]

Exponential control barrier functions for enforcing high relative-degree safety-critical constraints,

Q. Nguyen and K. Sreenath, “Exponential control barrier functions for enforcing high relative-degree safety-critical constraints,” in Proc. Amer . Contr . Conf., pp. 322–328, IEEE, 2016

work page 2016