Projection-Retraction MPPI: Exact Constraint-Manifold Control for Manipulators

Leesai Park; Minhyeong Kang; Sanghyun Kim; Seulchan Lee

REVIEW 2 major objections 6 minor 19 references

Projection-Retraction MPPI: Exact Constraint-Manifold Control for Manipulators

T0 review · 2 major / 6 minor · reviewed 2026-08-11 · deepseek-v4-flash

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

Pith's one-line read PR-MPPI enforces hard equality and inequality constraints inside MPPI rollouts, returning commands that satisfy the closed-chain equality to numerical tolerance independent of task weighting.

desk verdict A genuinely new nested projection-retraction MPPI for dual-arm equality plus inequality constraints, with honest reporting of its structural limit; conditional accept with a real soft spot at infeasible clearance directions. read the letter →

arxiv 2608.07573 v1 pith:SPWOUKQC submitted 2026-08-04 cs.RO

Seulchan Lee , Leesai Park , Minhyeong Kang , Sanghyun Kim This is my paper · ORCID

classification cs.RO

keywords modelpredictivepathintegralMPPIconstraintmanifoldcontrolbarrierfunctionsnull-spaceprojectionretractiondual-armmanipulationclosedkinematicchain

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

The paper seeks to show that MPPI, a sampling-based model-predictive controller, can enforce hard constraints by putting them in the sampled dynamics rather than the cost. PR-MPPI projects every sampled velocity into the tangent space of the equality constraint manifold and, within that subspace, into the half-spaces of active inequality constraints, then retracts the returned command onto the manifold. If the construction works, manipulation controllers can keep a two-arm closed chain exact and respect joint limits and obstacle clearances even when an aggressive task objective would overwhelm soft penalties. This matters because MPPI's gradient-free parallel sampling is attractive for nonconvex manipulation costs, but its standard form has no mechanism for constraints that must hold throughout the motion.

What carries the argument

The load-bearing object is the active constraint projector $\Pi_{\text{act}}$ of Eq. (5), which solves $\min_{u} \tfrac{1}{2}\|u-\tilde{u}\|^2$ subject to $J_c(q)u=0$ and $J_h(q)u\geq -\Gamma \bar{h}(q)$. It is implemented as a null-space projection $N(q)=I-J_c^\dagger J_c$ followed by an active-set correction that adds violated control-barrier-function half-spaces one at a time, each correction term carrying $N$ as a left factor so the equality stays satisfied to first order. The second mechanism is the retraction map $R_{\mathcal{M}}$, a Gauss-Newton iteration $q\leftarrow q-J_c^\dagger(q)c(q)$ run until $\|c(q)\|<\varepsilon_{\text{tol}}$, which removes the $O(\Delta t^2)$ drift of a finite step. These two pieces, projection inside the rollouts and retraction of the executed command, carry the exactness and margin claims.

What would settle it

Run the controller from a state such as the one reported in Section III-D where $N(q)\nabla \bar{h}(q)$ is nearly zero and the closed-chain equality forbids the retreating motion; if no feasible projection exists, the controller has no command that satisfies both constraints, contradicting the claim that constraints are enforced to tolerance in every state.

Watch

Extended reading notes

Core claim

The central claim is that a single velocity-space projection, applied at every rollout step and again during execution, lets MPPI maintain both an equality constraint $c(q)=0$ and inequality margins $h(q)\geq 0$ without turning them into cost penalties. Because each inequality is corrected within the null space of the equality Jacobian, an active margin can never push the state off the equality tangent subspace. The finite-step drift left by this first-order projection is then removed by a Gauss-Newton retraction onto $\mathcal{M}$, so whenever the retraction terminates the returned command satisfies $\|c(q^+)\| < \varepsilon_{\text{tol}}$ independently of the task cost, the MPPI temperature and covariance, and the inequality parameters (Remark 1). Lemma 1 bounds the residual inequality error as $O(\Delta t^2)$, small enough to be absorbed by a fixed safety margin. The paper supports this with a 14-DoF dual-arm simulation study and a humanoid hardware demonstration of reactive obstacle avoidance.

Load-bearing premise

The exactness guarantee assumes that the projection in Eq. (5) is feasible at every state the controller visits; Section III-D documents a hardware configuration where the projected obstacle-margin gradient nearly vanishes, so no admissible velocity can increase clearance without breaking the grasp, and the paper specifies no fallback.

Editorial extensions

If this is right

The returned joint command satisfies every channel of the closed-chain equality to the commanded tolerance, independent of how strongly the task cost pulls toward a constraint boundary.
Inequality margins are kept to an $O(\Delta t^2)$ finite-step layer, so a conservatively chosen safety buffer $h_{\text{safe}}$ is enough to keep measured states outside true limits.
Task costs no longer need soft constraint penalties; in the joint-limit stress test Full PR-MPPI runs without a limit-violation penalty and still never crosses the bound.
Projecting inside the rollouts, not only at execution, makes the optimizer average over feasible samples, which the randomized obstacle task links to a higher completion rate.
The same pipeline runs online at 30 Hz on a humanoid, reactively reshaping both arms around a moving obstacle while the tray remains level.

Reading between the lines

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

Inference: because the projection is purely kinematic, the same nested construction should transfer to sampled torques by using a dynamically consistent null-space projection, moving the exactness guarantee to the actuation layer; the paper leaves this extension to future work.
Inference: near configurations where the projected obstacle-margin gradient nearly vanishes, relaxing the grasp equality with a priority ordering would restore feasibility exactly where the current guarantee is vacuous, at the price of a small commanded grasp slip.
Inference: the $O(\Delta t^2)$ margin-layer bound suggests a quantitative tuning rule, namely shrinking the safety margin as $\Delta t$ decreases, which the paper does not state explicitly.

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

Two things you should know. First, this paper's actual contribution is real: a single velocity-space projector that nests CBF half-space corrections inside the equality tangent space, so inequality enforcement cannot break the closed-chain equality, plus a Gauss-Newton retraction that makes the returned command satisfy the equality to a declared tolerance. That combination is absent from the closest prior work (DQ-MPPI, MC-MPPI, Shield-MPPI), and the paper argues that gap clearly. Second, the paper's central guarantee is honest but conditional, and the authors say so themselves: Remark 1 states the equality exactness is a stopping-rule design property, and Lemma 1's O(Δt²) margin bound explicitly assumes the projection QP (5) is feasible at every cycle with a usable clearance direction.

What it does well: the active-set projection in (10)-(11) is sound and solved only over violated or banded margins, which is why compute stays at ~19 ms versus ~239 ms for DQ-MPPI on the same hardware. The ablation design is genuinely informative—execution-only inequality projection still enforces the constraints but degrades search, isolating the value of projecting inside rollouts. The simulator results are clean, and the hardware demo is real: the Unitree H1-2 arms avoid a moving obstacle reactively. The paper also reports the structural failure mode in §III-D plainly: when N∇h nearly vanishes, no tangent-space input can restore clearance, and no fallback is defined. That is a load-bearing limitation, not a hidden one.

Soft spots, in proportion. The feasibility assumption in Lemma 1 is not decorative. The paper's own hardware section documents a configuration where the feasible set contains no retreat direction, so the CBF condition cannot provide forward invariance there. The O(Δt²) bound is not made quantitative: constants κ_i are never computed, and the safety margin is chosen conservatively rather than derived from the bound. The active-set convergence proof is also missing; termination is asserted, not shown. None of this sinks the method—the equality guarantee is a design property and the inequality layer is a practical safety filter—but the claim "inequality enforcement with exact equality maintenance" should be read as "inequality enforcement when a clearance direction exists in the tangent space." The authors acknowledge this in the conclusion, though not in the abstract.

The citation pattern looks fair, and the self-citation to MC-MPPI is not a problem because that baseline is clearly separated in the experiments. The paper is worth a serious referee: it is clearly written, the algorithmic combination is new, and the failure mode is honestly documented. I would send it to review with a request for a declared infeasibility fallback or a stated relaxation strategy, and ideally a quantitative bound on the invariant-layer width. For a robotics audience this is a useful, citable contribution.

Referee Report

2 major / 6 minor

Summary. The paper proposes Projection-Retraction MPPI (PR-MPPI), a sampling-based MPC that enforces equality and inequality constraints inside the rollout dynamics rather than through soft penalties. At each rollout step, a sampled velocity is projected onto the intersection of the equality-constraint tangent space and the CBF half-spaces that are active within that tangent space, so that inequality corrections never break the equality. After the MPPI update, the averaged command sequence is filtered once more on its own predicted states, and the first commanded configuration is retracted onto the constraint manifold to a numerical tolerance. The paper contributes a command-level equality property (Remark 1), a finite-step inequality error bound (Lemma 1, O(Δt²)), and experiments on a 14-DoF dual-arm Franka system in simulation and on a Unitree H1-2 humanoid in hardware, including ablations and a randomized obstacle-avoidance study.

Significance. If the claims hold, PR-MPPI is a valuable addition to sampling-based MPC for manipulation: it offers a principled, penalty-free way to embed hard equality constraints (closed kinematic chains) together with CBF inequality constraints inside the same projection, with a clean composition that keeps inequality handling from destroying equality satisfaction. The experiments are well designed, include meaningful ablations, and the hardware demonstration addresses a realistic reactive-avoidance scenario. The paper is also honest in explicitly flagging a structural limitation in Section III-D. The main weaknesses are that the inequality guarantee is conditional on an assumption that the paper's own hardware experiment violates, and that the constants appearing in Lemma 1 are not quantified, so the advertised 'safety margin' argument remains qualitative.

major comments (2)

[Section III-D and Lemma 1] Lemma 1's finite-step margin bound (Eq. 16) is derived under the assumption that the projection QP (5) is feasible at every control cycle. Section III-D explicitly documents a configuration where the projected obstacle gradient N∇h nearly vanishes, so no admissible tangent-space input can increase clearance, and the paper states that this is a structural failure rather than a rate or latency artifact. In such a configuration, once the margin becomes negative the QP (5) is infeasible, and the algorithm has no declared fallback; the O(Δt²) bound therefore does not cover the regime where the inequality constraint is most needed. The authors acknowledge the limitation but do not amend Lemma 1 or the controller with a relaxation strategy. The manuscript should either define a fallback (e.g., selective equality relaxation or a minimum-norm infeasibility handling) and extend the bound to that case, or explicitly restrict the safety claim to states where (5) remains feasible.
[Lemma 1 and Section III-B] The constants κ_i in Lemma 1 are existential; the paper does not provide a way to compute or bound them in terms of known quantities (second-derivative bounds, Lipschitz constants of J_c and J_h, the tolerance ε_tol, and the input bound ū). Consequently the statement that the O(Δt²) error is 'small enough to be absorbed by the safety margin' (Section III-B) is not quantitative, and the choice of h_safe = 0.05 rad or 0.02 m is heuristic. As written, the lemma is a qualitative characterization rather than a safety certificate, and the forward-invariant-layer interpretation should be flagged as such, or the paper should supply a constructive bound on κ_i together with a margin-selection rule.

minor comments (6)

[Remark 1 / Algorithm 1] The retraction guarantee holds only 'whenever the iteration terminates.' For a real-time controller, the paper should specify a maximum iteration count and define the behavior if the tolerance is not reached within that budget, otherwise the 30 Hz hardware claim is not fully supported.
[Section II-B, Eq. (10)-(11)] The active-set iteration is stated to terminate and yield the minimizer of (5), but no termination argument is given. Since constraints are only added and never removed, a short proof of convergence (or a citation to a standard active-set QP convergence result) would make the projection step fully rigorous.
[Table II and Figure 2] The difference between 'Max. bound viol.' and 'Max. margin pen.' is not defined precisely; please clarify whether margin penetration is measured against the shifted guard boundary or the true bound, and make the same distinction in the text of Section III-B.
[Section III-B] The sentence 'The 0.05 rad safety shift is selected conservatively' is not supported by a computation. Since Lemma 1 does not provide κ_i, this choice is empirical; consider stating explicitly that the margin is a tuned parameter and reporting sensitivity to it.
[Section II-A / II-B] The feasibility of the QP (5) is not discussed in the implementation details. The paper should state how infeasibility is detected in the active-set iteration (e.g., maximum iterations or a residual threshold) and what the controller does in that case.
[Section III-C] The comparison against MC-MPPI and DQ-MPPI is useful, but a comparison against a CBF-QP filter or Shield-MPPI would more directly isolate the benefit of the proposed projection-plus-retraction design for inequality enforcement.

Circularity Check

1 steps flagged · score 4.0 of 10

Retraction exactness is the loop's own stopping condition (Remark 1); the advertised equality guarantee is therefore definitional, while Lemma 1's margin bound is conditional on an unverified feasibility assumption the paper's own §III-D documents failing.

self definitional [Section II-C, Eq. (15), Remark 1]
"Remark1 (Retraction exactness). The stopping rule of (15) makes the commanded tolerance a design property rather than a theorem: whenever the iteration terminates, the retracted command satisfies ∥c(q+)∥ < εtol, independently of the task cost, the MPPI temperature and covariance, and the inequality-handling parameters."

The central 'exact command' guarantee is literally the exit test of the retraction loop: (15) iterates q←q−J†_c(q)c(q) 'until ∥c(q)∥<εtol'. Hence 'the returned command satisfies the equality to numerical tolerance' is true by construction whenever the loop stops; it does not follow from the projection, the MPPI weighting, or any external property. The paper explicitly says it 'is a design property rather than a theorem', so the abstract's claim that retraction returns the command 'back onto the constraint to numerical tolerance and independent of task weighting' reduces to the algorithm's own termination predicate. This is a self-definitional statement rather than a derived prediction.

full rationale

No data-fitting or renamed-known-result circularity was found. The QP projection (5), the nested active-set solver (10)-(11), and execution filtering (13) are independent algorithmic constructions; the equality preservation J_c N=0 is a computed identity, and comparisons against DQ-MPPI, MC-MPPI, and ablations are external benchmarks rather than circular validation. The self-citations [8] and [18] describe baselines or related work and are not load-bearing supports of the paper's derivation. The one genuine circular element is the 'exactness' headline: Remark 1 concedes that the command-level equality guarantee is the stopping rule of (15), so presenting it as a result is definitional. Because the authors flag this explicitly and the rest of the framework has independent content, the score is moderate rather than severe. Separately, Lemma 1 is a legitimate conditional bound, but its premise 'If (5) is feasible at q_t at every cycle' is unverified: Section III-D reports N∇h ≈ 0 configurations where no admissible tangent input improves clearance, so the O(Δt²) inequality layer does not cover the regime where it matters; this is a safety and completeness limitation, not an additional circularity.

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

The central claims rest on feasibility of the projection at every state and on termination of the active-set and retraction iterations. The paper states the feasibility condition explicitly in Lemma 1 and observes its failure in hardware, but offers no fallback. The inequality margin bound uses regularity constants that are never computed, so the safety margin is a hand-chosen parameter. No new physical entities are introduced.

free parameters (5)

CBF gain gamma_i = 5 for all inequalities
Chosen identically across experiments; appears in Eq. (4) and in Lemma 1's condition gamma_i * delta_t <= 1. The bound's constants kappa_i depend on this gain indirectly.
Safety margin h_safe = 0.05 rad (joint-limit), 0.02 m (obstacle)
Selected conservatively to absorb O(delta_t^2) drift and tracking error; not derived from the computed kappa_i of Lemma 1, so the paper does not provide a constructive formula for this margin.
Exploration covariance sigma = 0.03 rad/s
Standard MPPI exploration tuning parameter, chosen by the authors and shared across experiments. It does not affect the equality guarantee but shapes the sampled rollouts.
Retraction tolerance epsilon_tol = <1e-9 (reported)
A design parameter that controls the stopping rule of Eq. (15); Remark 1 makes the equality guarantee conditional on this tolerance being reached.
Activation band h_band = not reported in the text
Parameter in Eq. (9) that includes near-boundary margins into the active set preemptively. The paper does not give its numerical value or a rule for choosing it.

assumptions (5)

domain assumption The constrained projection (5) is feasible at every state visited by the controller.
Lemma 1 explicitly assumes feasibility at every cycle. Section III-D documents a structural failure when the projected margin gradient nearly vanishes, so this assumption can be violated in practice.
domain assumption The Gauss-Newton retraction iteration (15) terminates at the prescribed tolerance.
Remark 1 makes the equality guarantee conditional on termination. No convergence radius, regularity condition, or fallback is provided for the retraction map.
ad hoc to paper The active-set iteration (10)-(11) terminates and yields the minimizer of (5).
The paper asserts KKT optimality at termination but gives no proof that the residual-checking loop converges in a finite number of steps or does not cycle.
domain assumption Regularity conditions of Lemma 1: bounded second derivatives of c and h_i, constant row rank of J_c, bounded pseudoinverse, bounded u and grad h_i, and gamma_i * delta_t <= 1.
These stated assumptions are needed for the O(delta_t^2) inequality bound. They are reasonable for the experiments but not verified pointwise over the whole operating set.
domain assumption The CBF condition (4) plus a safety buffer is an acceptable model of constraint safety.
The paper models inequality safety through first-order CBF half-spaces and a shifted guard boundary. This is standard but still a modeling choice, and the margin h_safe is not computed from the bound.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Projection-Retraction MPPI: Exact Constraint-Manifold Control for Manipulators." pith.science (2026). https://pith.science/paper/SPWOUKQC

@misc{pith2026260807573,
  author       = {Pith},
  title        = {Pith review of: Projection-Retraction MPPI: Exact Constraint-Manifold Control for Manipulators},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/SPWOUKQC}},
  note         = {Machine review of arXiv:2608.07573}
}

read the original abstract

Model Predictive Path Integral (MPPI) control is widely used in manipulation for its gradient-free, parallel handling of non-convex costs. Manipulation tasks, however, often impose constraints that hold throughout the motion: a closed kinematic chain that two grasping arms keep exactly, or joint limits and obstacle clearances that are never crossed. MPPI handles such constraints only through the cost, as soft penalties that hold approximately and fail under a strong task cost. To address this, we propose Projection-Retraction MPPI (PR-MPPI), which enforces the constraints inside the sampled dynamics. At every rollout step, the sampled velocity is projected to satisfy both constraint types: the equality restricts it to a subspace, and each inequality to a half-space within that subspace, so inequality handling never breaks the equality. This projection, however, satisfies the constraints only to first order, and a finite step leaves a small drift off the equality. Therefore, we retract the returned command back onto the constraint to numerical tolerance and independent of task weighting. We validate PR-MPPI on 14-DoF dual-arm systems. In simulation, the returned commands satisfy the closed-chain equality to numerical tolerance through a joint-limit stress test and randomized obstacle avoidance. On real hardware, the arms of a Unitree H1-2 humanoid reactively avoid a moving obstacle. Code and experiment videos are available at https://rcilab.github.io/prmppi.

Figures

Figures reproduced from arXiv: 2608.07573 by the authors.

**Figure 1.** Overview of PR-MPPI. Sampled control sequences are rolled out through the constraint-aware dynamics, where the active constraint projector [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗

**Figure 2.** Joint-limit stress test. (a) Dual-arm tray-lowering motion: the goal joint configuration in green and the trajectory executed by Full PR-MPPI in blue. (b) Measured fourth-joint trajectories of both arms for the four controllers. The dashed line is the artificial lower bound q4,min = −2.35 rad and the dotted line the shifted guard boundary at −2.30 rad; the shaded band between them is the 0.05 rad CBF margin. B. Expe… view at source ↗

**Figure 3.** Randomized obstacle-avoidance task at the central lateral offset. The [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗

Figures from the paper (1 more)

**Figure 4.** Figure 4: Hardware demonstration on the Unitree H1-2. The two arms hold a tray carrying a payload under the closed-chain grasp while a hand-held obstacle [PITH_FULL_IMAGE:figures/full_fig_p007_4.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

19 extracted references · 14 canonical work pages

[1]

Aggressive driving with model predictive path integral control,

G. Williams, P. Drews, B. Goldfain, J. M. Rehg, and E. A. Theodorou, “Aggressive driving with model predictive path integral control,” in 2016 IEEE international conference on robotics and automation (ICRA). IEEE, 2016, pp. 1433–1440

2016
[2]

Model predictive path integral control: From theory to parallel computation,

G. Williams, A. Aldrich, and E. A. Theodorou, “Model predictive path integral control: From theory to parallel computation,”Journal of Guidance, Control, and Dynamics, vol. 40, no. 2, pp. 344–357, 2017

2017
[3]

Storm: An integrated framework for fast joint- space model-predictive control for reactive manipulation,

M. Bhardwaj, B. Sundaralingam, A. Mousavian, N. D. Ratliff, D. Fox, F. Ramos, and B. Boots, “Storm: An integrated framework for fast joint- space model-predictive control for reactive manipulation,” inConference on Robot Learning. PMLR, 2022, pp. 750–759

2022
[4]

Chance-constrained information- theoretic stochastic model predictive control with safety shielding,

J. Yin, P. Tsiotras, and K. Berntorp, “Chance-constrained information- theoretic stochastic model predictive control with safety shielding,” in Proc. IEEE Conf. Decision and Control (CDC), 2024, pp. 653–658

work page 2024
[5]

Shield model predictive path integral: A computationally efficient robust MPC method using control barrier functions,

J. Yin, C. Dawson, C. Fan, and P. Tsiotras, “Shield model predictive path integral: A computationally efficient robust MPC method using control barrier functions,”IEEE Robot. Autom. Lett., vol. 8, no. 11, pp. 7106–7113, 2023

work page 2023
[6]

PRIEST: Projection Guided Sampling-Based Optimization For Autonomous Navigation

F. Rastgar, H. Masnavi, B. Sharma, A. Aabloo, J. Swevers, and A. K. Singh, “PRIEST: Projection guided sampling-based optimization for autonomous navigation,”arXiv preprint arXiv:2309.08235, 2023

work page Pith review arXiv 2023
[7]

Online motion generation via tangential sampling-based MPC around nonconvex obstacles,

G. Zhao, N. Jin, J. Wu, and Z. Xiong, “Online motion generation via tangential sampling-based MPC around nonconvex obstacles,”IEEE Robot. Autom. Lett., vol. 10, no. 6, pp. 5537–5544, 2025

work page 2025
[8]

CSC-MPPI: A novel constrained MPPI framework with DBSCAN for reliable obstacle avoidance,

L. Park, K. Jang, and S. Kim, “CSC-MPPI: A novel constrained MPPI framework with DBSCAN for reliable obstacle avoidance,”arXiv preprint arXiv:2506.16386, 2025

arXiv 2025

Show all 19 references

[9]

π-MPPI: A projection-based model predictive path integral scheme for smooth optimal control of fixed-wing aerial vehicles,

E. M. Andrejev, A. Manoharan, K.-E. Unt, and A. K. Singh, “π-MPPI: A projection-based model predictive path integral scheme for smooth optimal control of fixed-wing aerial vehicles,”IEEE Robot. Autom. Lett., vol. 10, no. 6, 2025

2025
[10]

Inverse kinematic solutions with singularity robustness for robot manipulator control,

Y . Nakamura and H. Hanafusa, “Inverse kinematic solutions with singularity robustness for robot manipulator control,”J. Dyn. Syst. Meas. Control, vol. 108, no. 3, pp. 163–171, 1986

1986
[11]

A general framework for managing multiple tasks in highly redundant robotic systems,

B. Siciliano and J.-J. E. Slotine, “A general framework for managing multiple tasks in highly redundant robotic systems,” inProc. Int. Conf. Adv. Robot. (ICAR), 1991, pp. 1211–1216

1991
[12]

A weighted least-norm solution based scheme for avoiding joint limits for redundant joint manipulators,

T. F. Chan and R. V . Dubey, “A weighted least-norm solution based scheme for avoiding joint limits for redundant joint manipulators,”IEEE Trans. Robot. Autom., vol. 11, no. 2, pp. 286–292, 1995

1995
[13]

Constrained closed loop inverse kinematics,

B. Dariush, Y . Zhu, A. Arumbakkam, and K. Fujimura, “Constrained closed loop inverse kinematics,” inProc. IEEE Int. Conf. Robot. Autom. (ICRA), 2010, pp. 2499–2506

2010
[14]

Manipula- tion planning on constraint manifolds,

D. Berenson, S. S. Srinivasa, D. Ferguson, and J. J. Kuffner, “Manipula- tion planning on constraint manifolds,” inProc. IEEE Int. Conf. Robot. Autom. (ICRA), 2009, pp. 625–632

2009
[15]

Path planning under kinematic constraints by rapidly exploring manifolds,

L. Jaillet and J. M. Porta, “Path planning under kinematic constraints by rapidly exploring manifolds,”IEEE Trans. Robot., vol. 29, no. 1, pp. 105–117, 2013

2013
[16]

Exploring implicit spaces for constrained sampling-based planning,

Z. Kingston, M. Moll, and L. E. Kavraki, “Exploring implicit spaces for constrained sampling-based planning,”Int. J. Robot. Res., vol. 38, no. 10–11, pp. 1151–1178, 2019

2019
[17]

Real-time dual-arm cooperative manipulation under multiple constraints: A two-stage sampling MPC approach,

T. Zhu, J. Mao, J. Yang, and S. Li, “Real-time dual-arm cooperative manipulation under multiple constraints: A two-stage sampling MPC approach,”IEEE Transactions on Robotics, 2026

2026
[18]

Manifold-constrained model predictive path integral control,

S. Lee and S. Kim, “Manifold-constrained model predictive path integral control,”arXiv preprint arXiv:2605.24813, 2026

2026 arXiv
[19]

Mujoco: A physics engine for model- based control,

E. Todorov, T. Erez, and Y . Tassa, “Mujoco: A physics engine for model- based control,” in2012 IEEE/RSJ international conference on intelligent robots and systems. IEEE, 2012, pp. 5026–5033

2012

Pith tools

Reviewed August 11, 2026 · model on record in the stance chip above.

[1] [1]

Aggressive driving with model predictive path integral control,

G. Williams, P. Drews, B. Goldfain, J. M. Rehg, and E. A. Theodorou, “Aggressive driving with model predictive path integral control,” in 2016 IEEE international conference on robotics and automation (ICRA). IEEE, 2016, pp. 1433–1440

2016

[2] [2]

Model predictive path integral control: From theory to parallel computation,

G. Williams, A. Aldrich, and E. A. Theodorou, “Model predictive path integral control: From theory to parallel computation,”Journal of Guidance, Control, and Dynamics, vol. 40, no. 2, pp. 344–357, 2017

2017

[3] [3]

Storm: An integrated framework for fast joint- space model-predictive control for reactive manipulation,

M. Bhardwaj, B. Sundaralingam, A. Mousavian, N. D. Ratliff, D. Fox, F. Ramos, and B. Boots, “Storm: An integrated framework for fast joint- space model-predictive control for reactive manipulation,” inConference on Robot Learning. PMLR, 2022, pp. 750–759

2022

[4] [4]

Chance-constrained information- theoretic stochastic model predictive control with safety shielding,

J. Yin, P. Tsiotras, and K. Berntorp, “Chance-constrained information- theoretic stochastic model predictive control with safety shielding,” in Proc. IEEE Conf. Decision and Control (CDC), 2024, pp. 653–658

work page 2024

[5] [5]

Shield model predictive path integral: A computationally efficient robust MPC method using control barrier functions,

J. Yin, C. Dawson, C. Fan, and P. Tsiotras, “Shield model predictive path integral: A computationally efficient robust MPC method using control barrier functions,”IEEE Robot. Autom. Lett., vol. 8, no. 11, pp. 7106–7113, 2023

work page 2023

[6] [6]

PRIEST: Projection Guided Sampling-Based Optimization For Autonomous Navigation

F. Rastgar, H. Masnavi, B. Sharma, A. Aabloo, J. Swevers, and A. K. Singh, “PRIEST: Projection guided sampling-based optimization for autonomous navigation,”arXiv preprint arXiv:2309.08235, 2023

work page Pith review arXiv 2023

[7] [7]

Online motion generation via tangential sampling-based MPC around nonconvex obstacles,

G. Zhao, N. Jin, J. Wu, and Z. Xiong, “Online motion generation via tangential sampling-based MPC around nonconvex obstacles,”IEEE Robot. Autom. Lett., vol. 10, no. 6, pp. 5537–5544, 2025

work page 2025

[8] [8]

CSC-MPPI: A novel constrained MPPI framework with DBSCAN for reliable obstacle avoidance,

L. Park, K. Jang, and S. Kim, “CSC-MPPI: A novel constrained MPPI framework with DBSCAN for reliable obstacle avoidance,”arXiv preprint arXiv:2506.16386, 2025

arXiv 2025