Pith. sign in

REVIEW 4 major objections 7 minor 49 references

Multi-Robot Cooperative Herding through Backstepping Control Barrier Functions

T0 review · 4 major / 7 minor · reviewed 2026-08-06 · deepseek-v4-flash

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

Pith's one-line read A backstepping control barrier function controller coordinates multiple herder robots to drive evader robots into a goal region while preventing inter-evader collisions, using only repulsive interaction.

desk verdict A useful herding demo and a genuinely new application of backstepping CBFs, undermined by a Jacobian identity that only holds under a one-to-one repulsion assumption the implemented dynamics don't use. read the letter →

arxiv 2507.10249 v1 pith:YTQ2SRCB submitted 2025-07-14 cs.MA

Kang Li , Ming Li , Wenkang Ji , Zhiyong Sun , Shiyu Zhao This is my paper · ORCID
classification cs.MA
keywords herdingcooperativecontrolbarrierfunctionsbacksteppingmulti-robotsystemscollisionavoidanceunderactuatedquadraticprogramming
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

Backstepping control barrier functions can coordinate multiple herder robots to drive evader robots into a goal region while keeping evaders separated, even though the herders act on evaders only indirectly through repulsive forces. The paper reformulates the inverse-model herding dynamics into a control-affine form by treating each evader's velocity as a virtual control, then constructs hierarchical barrier functions for goal reaching and inter-evader collision avoidance, and finally filters a nominal stabilizing input through a quadratic program. Simulations with three herders and three evaders, together with physical experiments on omnidirectional robots, show the evaders reaching and staying in the goal region without collisions. If the claim holds, the result is a runtime safety filter for an underactuated, high-order herding system that avoids taking derivatives of the higher-order dynamics.

What carries the argument

The central mechanism is the backstepping control barrier function hierarchy. The evader velocity $v_{Ei}$ is declared a virtual input, so the herder dynamics split into two control-affine layers, $\dot{x}_{Ei}=v_{Ei}$ and $\dot v_{Ei}=(\partial f_{Ei}/\partial x_{Ei})v_{Ei}+(\partial f_{Ei}/\partial x_{Hk})u_{Hk}$; a goal-reaching barrier $h_1$ and an evader-separation barrier $h_2$ are then defined on this layered state, with virtual controllers $r_h(x_{Ei})$ and $r_a(x_{Eij})$ satisfying the inner conditions. The outer controller is a QP that keeps $\dot h_1$ and $\dot h_2$ nonnegative, which makes the set of safe states forward invariant without computing high-order derivatives of the original system.

What would settle it

Run the decentralized controller with $n=2$ herders and $m=3$ evaders (or with overlapping nearest-evader assignments), starting some herders beyond the distance at which $\partial f_{Ei}/\partial x_{Hk}$ is nearly singular, and check whether the QP in Eq. (27) becomes infeasible or any evader pair violates the $R_{\rm avoid}$ radius before the evaders reach the goal region; a single such failure refutes the claimed safety guarantee.

Watch

Extended reading notes

Core claim

The paper's central discovery is that underactuated herding with inverse-model evaders—where $\dot{x}_{Ei} = \kappa_H \sum_{k} (x_{Ei}-x_{Hk})/\|x_{Ei}-x_{Hk}\|^3$—can be put into control-affine form by taking the evader velocity $v_{Ei} = f_{Ei}(x_E,x_H)$ as an intermediate variable. With $\dot v_{Ei} = \frac{\partial f_{Ei}}{\partial x_{Ei}} v_{Ei} + \frac{\partial f_{Ei}}{\partial x_{Hk}} u_{Hk}$, the backstepping CBF framework applies: two barrier functions, $h_1$ for goal reaching and $h_2$ for evader separation, are stacked, virtual controllers are chosen to satisfy the inner CBF inequalities, and a quadratic-program safety filter minimizes deviation from a Sontag-formula nominal controller while enforcing the barrier conditions. The paper also gives centralized and decentralized decompositions of the collision-avoidance constraint and proves that, under a positivity condition on the $c_{ij}$ terms, the QP is feasible and the evader set stays safe.

Load-bearing premise

The design assumes each herder's command has full authority over its assigned evader's velocity through the repulsion Jacobian, but that coupling vanishes as $1/\|x_{Ei}-x_{Hk}\|^3$ with distance, and the one-to-one herder-to-evader assignment is only assumed; if these fail, the safety-filter QP can become infeasible and the barrier guarantees can break.

Editorial extensions

If this is right

  • All evaders can be driven into the designed goal region and kept there, with inter-evader distances above the safety radius, under both centralized and decentralized implementations.
  • The framework's CBF-QP structure accepts additional safety or task constraints, such as herder collision avoidance, by adding further barrier constraints.
  • Restricting collision constraints to a neighborhood set bounds the per-herder QP by $m-1$ constraints, so the decentralized controller scales with local evader density rather than total group size.
  • When barrier constraints stay inactive, the nominal Sontag-based controller drives the goal-reaching objective; when they activate, the safety filter preserves stability as much as the constraints allow.

Reading between the lines

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

  • The one-to-one assumption that each herder acts on only its nearest evader is not proven to hold dynamically; a testable extension would add explicit herder-to-evader assignment and check whether safety is preserved when two herders target the same evader.
  • Because the coupling Jacobian $\partial f_{Ei}/\partial x_{Hk}$ decays as $1/\|x_{Ei}-x_{Hk}\|^3$, herders far from an evader lose effective authority; a practical distance bound guaranteeing QP feasibility would make the safety claim tighter.
  • The paper's mention of multiple equilibria suggests the attractive goal-reaching force and the repulsive collision-avoidance force can balance in dense configurations; replacing the perturbation fix with a formal anti-equilibria construction would strengthen the guarantee.
  • The framework is presented for equal numbers of herders and evaders; extending to $n \neq m$ would require a coordination or task-allocation layer, which the paper leaves open.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

4 major / 7 minor

Summary. The paper proposes a cooperative herding controller based on backstepping control barrier functions (CBFs) for a team of herders driving a group of evaders to a goal region. The evaders follow an inverse dynamics model with repulsive interactions from all herders, and the herders are velocity-controlled. The authors reformulate the underactuated dynamics, construct separate backstepping CBFs for goal reaching and inter-evader collision avoidance, and combine them in a QP-based safety filter with centralized and decentralized variants. The claimed contribution is a systematic safety-guaranteed herding strategy, supported by simulations and hardware experiments with three herders and three evaders.

Significance. If the theoretical guarantees were valid, this would be a useful application of backstepping CBFs to an underactuated multi-robot herding problem, and the hardware demonstration with omnidirectional robots is a tangible strength. The paper is clearly written and the qualitative simulation and experimental results show plausible herding behavior. However, the central safety theorem currently rests on an incorrect Jacobian identity and on an unproved feasibility condition, and the goal-reaching objective is not established by any theorem. Because these issues are load-bearing for the paper's central claim of safe herding completion, the significance of the contribution is not yet established.

major comments (4)
  1. [Section 4.1, Eq. (11)] Equation (11) uses the identity ∂f_Ei/∂x_Hk = −∂f_Ei/∂x_Ei, which is not true for the inverse dynamics (2): ∂f_Ei/∂x_Ei is a sum over all herders, whereas ∂f_Ei/∂x_Hk contains only the k-th term. The time derivative of v_Ei should generally include a sum over all herder inputs. The identity is valid only under a one-to-one repulsion assignment, but that assumption is introduced only in the final paragraph of Section 4.3 and is not part of Eq. (2), which is the model used in the simulations and experiments. Consequently, the Lie derivatives in the backstepping conditions (16), (17), (23), and the QP constraints (27) are computed for a different system, so Theorem 1's collision-avoidance guarantee does not apply to the implemented dynamics.
  2. [Section 4.3, Theorem 1] The theorem's condition c_ij > 0 is not derived from the dynamics and is not shown to hold along closed-loop trajectories. The quantity c_ij(x_Eij, v_Eij) in (23) depends on the backstepping error and on h2, and nothing in the paper rules out c_ij becoming negative. The proof additionally states that 'd_ij = 0 is satisfied by assumption' without justification. Feasibility of QP (27) is therefore an unproven, state-dependent assumption, and the theorem does not establish the forward-invariance guarantee required by Problem 1.
  3. [Section 4.3, Eq. (27) and Remark 3] The QP (27) enforces only the collision-avoidance constraints (25); the goal-reaching condition (18) is absent from the optimization. Convergence to the goal region G is delegated to the nominal controller (26) and Remark 3, but no control-Lyapunov-function or asymptotic-stability argument is provided, and Remark 3 explicitly assumes the CBF constraints are inactive. Since the safety filter may modify u_Hk whenever collision constraints are active, the paper does not prove that the evaders reach or remain in G. Thus the 'herding completion' part of Problem 1 is not established.
  4. [Sections 4.3 and 5.1] The QP (27) is unconstrained in u_Hk, yet the simulations and experiments enforce a velocity saturation (vmax = 3 m/s in simulation, 0.3 m/s on hardware). If the QP returns a command above the limit, the saturated input is not the one for which the CBF condition was verified, so the invariance argument cannot be invoked. The paper should either include input bounds in the QP together with a feasibility analysis, or justify that the saturation is inactive in all reported scenarios.
minor comments (7)
  1. [Section 4.1, after Eq. (11)] The sentence 'I2×2 ∈ R2 is the identity matrix' should read 'I2×2 ∈ R^{2×2}'.
  2. [Algorithm 1, lines 2 and 5] Line 2 updates 'h1(x_Ei, v_Eij)' although h1 in (15) is defined over (x_Ei, v_Ei), and line 5 uses 'h2(x_Eij) < 0' where the backstepping CBF h2(x_Eij, v_Eij) from (22) appears to be intended.
  3. [Algorithm 1 and Section 5.1] The parameters k_h and k_a are initialized in Algorithm 1 and listed in Section 5.1, but they never appear in any controller equation; their role should be clarified or the parameters should be removed.
  4. [Figure 4b] The legend lists 'evader 3-2' in addition to 'evader 2-3'; one of these entries appears to be mislabeled.
  5. [Section 4.2.2 and Section 4.3] The paper claims both centralized and decentralized implementations, but only the decentralized QP (27) is actually formulated; the centralized version is represented only by the single inequality (24).
  6. [Section 3.2 and Remark 2] The opening paragraph of Section 3.2 says collision-free motion must be ensured both within and between the groups, while Remark 2 states that collisions between herders are out of scope; the problem statement should be made consistent.
  7. [Section 4.2.1] The text says 'the CBF condition defined in (3)', but the CBF condition is given by Eq. (9) in Definition 3, not Eq. (3).

Circularity Check

0 steps flagged · score 2.0 of 10

No significant circularity: the central derivation is self-contained and independent; only minor non-load-bearing self-citations appear.

full rationale

The paper's central derivation is not circular. The backstepping CBF construction follows the external framework of [37] (Taylor et al.), not the authors' own prior results. The candidate barrier functions h1 and h2 are defined from the goal region and inter-evader distances, and the QP constraints in Eq. (27) are the standard CBF derivative inequalities. No parameter is fitted to a target outcome. The simulation and experimental metrics are not used to define the controller. The self-citations [44] and [49] are minor and not load-bearing: [44] proposes a smooth universal formula that is mentioned but not actually used, since the paper verifies its simpler virtual control (14) directly and uses Sontag's formula [43] for the nominal controller; [49] is only a hardware platform reference for the experiments. Theorem 1 is conditional on the feasibility assumption cij > 0, which is an unproven completeness condition rather than a circular restatement. The later one-to-one repulsion assumption in Section 4.3 makes the Jacobian identity in Eq. (11) internally consistent, but it is not shown to hold for the summed dynamics (2); this is a correctness and validity gap, not circular reasoning.

Assumptions & free parameters 5 free parameters · 4 assumptions · 0 invented entities

The central derivation introduces no new physical entities. The intermediate variable v_Ei is a mathematical state for backstepping, not a new postulated quantity. The key free parameters are controller gains and the repulsion gain; task radii and speed limits are problem settings rather than tuning constants introduced ad hoc.

free parameters (5)
  • kappa_H (repulsive gain) = 10 in simulations
    Sets the strength of evader repulsion from herders in Eq. (2); chosen by hand, and all herding outcomes depend on it.
  • gamma_h (goal CBF gain) = 0.5
    Appears in the goal-reaching CBF condition (13) and virtual controller (14); controls convergence speed.
  • gamma_a (collision CBF gain) = 0.5
    Appears in the collision-avoidance CBF condition (20) and controller (21); controls how aggressively evaders separate.
  • mu (backstepping weight) = 1
    Weight in the backstepping CBFs (15) and (22); affects the relative penalty on virtual-input tracking error.
  • k_h, k_a = 1 (listed in Algorithm 1, otherwise undefined)
    Initialized in Algorithm 1 but never defined in the text or used in equations; suggests incomplete documentation of tuning parameters.
assumptions (4)
  • standard math CBF forward-invariance theory from [41]: a set C defined by h >= 0 is safe if Lf h + Lg h u + beta(h) >= 0.
    Used to derive the CBF inequalities (13), (16), (20), (22); standard control theory.
  • domain assumption Evaders obey the inverse dynamics model (2) with known gain kappa_H, and herders have full state knowledge (Assumption 1).
    The entire derivation and experiments rely on this exact model and perfect measurements; the authors acknowledge this as a limitation.
  • domain assumption Each herder influences only its nearest evader, giving a one-to-one repulsion structure.
    Stated in the paragraph after Remark 5; used to justify the decentralized controller and simplify the coupled dynamics, but no assignment algorithm or proof is given.
  • ad hoc to paper Theorem 1 requires c_ij > 0 (and the proof assumes d_ij = 0 when needed) for feasibility of the QP.
    The condition is assumed, not derived; the paper notes that in dense environments constraints may conflict, which would violate it.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Multi-Robot Cooperative Herding through Backstepping Control Barrier Functions." pith.science (2026). https://pith.science/paper/YTQ2SRCB

@misc{pith2026250710249,
  author       = {Pith},
  title        = {Pith review of: Multi-Robot Cooperative Herding through Backstepping Control Barrier Functions},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/YTQ2SRCB}},
  note         = {Machine review of arXiv:2507.10249}
}
read the original abstract

We propose a novel cooperative herding strategy through backstepping control barrier functions (CBFs), which coordinates multiple herders to herd a group of evaders safely towards a designated goal region. For the herding system with heterogeneous groups involving herders and evaders, the behavior of the evaders can only be influenced indirectly by the herders' motion, especially when the evaders follow an inverse dynamics model and respond solely to repulsive interactions from the herders. This indirect interaction mechanism inherently renders the overall system underactuated. To address this issue, we first construct separate CBFs for the dual objectives of goal reaching and collision avoidance, which ensure both herding completion and safety guarantees. Then, we reformulate the underactuated herding dynamics into a control-affine structure and employ a backstepping approach to recursively design control inputs for the hierarchical barrier functions, avoiding taking derivatives of the higher-order system. Finally, we present a cooperative herding strategy based on backstepping CBFs that allow herders to safely herd multiple evaders into the goal region. In addition, centralized and decentralized implementations of the proposed algorithm are developed, further enhancing its flexibility and applicability. Extensive simulations and real-world experiments validate the effectiveness and safety of the proposed strategy in multi-robot herding.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

49 extracted references · 47 canonical work pages

  1. [1]

    Current Biology 22(14), 561– 562 (2012)

    King, A.J., Wilson, A.M., Wilshin, S.D., Lowe, J., Haddadi, H., Hailes, S., Mor- ton, A.J.: Selfish-herd behaviour of sheep under threat. Current Biology 22(14), 561– 562 (2012)

    work page 2012

  2. [2]

    Men- don¸ ca, R., Hirata, S., Kubo, T., Ikeda, K., Yamamoto, S.: Herding mechanisms to main- tain the cohesion of a harem group: Two interaction phases during herding

    Ringhofer, M., Go, C.K., Inoue, S., S. Men- don¸ ca, R., Hirata, S., Kubo, T., Ikeda, K., Yamamoto, S.: Herding mechanisms to main- tain the cohesion of a harem group: Two interaction phases during herding. Journal of Ethology 38, 71–77 (2020)

    work page 2020

  3. [3]

    245–260 (2014)

    Coppinger, L., Coppinger, R.: Dogs for herd- ing and guarding livestock., pp. 245–260 (2014)

    work page 2014

  4. [4]

    Artificial Life and Robotics 27(2), 416–427 (2022)

    Kubo, M., Tashiro, M., Sato, H., Yamaguchi, A.: Herd guidance by multiple sheepdog agents with repulsive force. Artificial Life and Robotics 27(2), 416–427 (2022)

    work page 2022

  5. [5]

    In: Proceedings of the IEEE Conference on Decision and Control, pp

    Singh, R.K., Chakraborty, D.: Planar Herd- ing of Multiple Evaders by a Single Pursuer. In: Proceedings of the IEEE Conference on Decision and Control, pp. 7375–7380 (2024)

    work page 2024

  6. [6]

    IEEE/CAA Journal of Automatica Sinica 9(4), 732–734 (2022)

    Huang, J., Zhou, S., Tu, H., Yao, Y., Liu, Q.: Distributed optimization algorithm for multi-robot formation with virtual reference center. IEEE/CAA Journal of Automatica Sinica 9(4), 732–734 (2022)

    work page 2022

  7. [7]

    IEEE Transactions on Intelligent Transportation Systems 23(8), 11811–11822 (2021)

    Bai, C., Yan, P., Pan, W., Guo, J.: Learning- based multi-robot formation control with obstacle avoidance. IEEE Transactions on Intelligent Transportation Systems 23(8), 11811–11822 (2021)

    work page 2021

  8. [8]

    IEEE Transactions on Automatic Control 63(12), 4140–4155 (2018)

    Zhao, S.: Affine formation maneuver control of multiagent systems. IEEE Transactions on Automatic Control 63(12), 4140–4155 (2018)

    work page 2018

Show all 49 references
  1. [9]

    IEEE Transactions on Automatic Control 64(11), 4541–4554 (2019)

    Zhao, S., Li, Z., Ding, Z.: Bearing-only forma- tion tracking control of multiagent systems. IEEE Transactions on Automatic Control 64(11), 4541–4554 (2019)

    2019

  2. [10]

    Robotics and Autonomous Systems 161, 104314 (2023)

    Martin, J.G., Muros, F.J., Maestre, J.M., Camacho, E.F.: Multi-robot task allocation clustering based on game theory. Robotics and Autonomous Systems 161, 104314 (2023)

    2023

  3. [11]

    Nature Communications 14(1), 3476 (2023)

    Sun, G., Zhou, R., Ma, Z., Li, Y., Groß, R., Chen, Z., Zhao, S.: Mean-shift exploration in shape assembly of robot swarms. Nature Communications 14(1), 3476 (2023)

    2023

  4. [12]

    Journal of Guidance, Con- trol, and Dynamics 43(7), 1365–1373 (2020)

    Li, K., Wang, J., Lee, C.-H., Zhou, R., Zhao, S.: Distributed cooperative guidance for mul- tivehicle simultaneous arrival without numer- ical singularities. Journal of Guidance, Con- trol, and Dynamics 43(7), 1365–1373 (2020)

    2020

  5. [13]

    In: Proceedings of the Interna- tional Conference on Robotics and Automa- tion, pp

    Zhang, T., Liu, Z., Pu, Z., Yi, J.: Multi- target encirclement with collision avoidance via deep reinforcement learning using rela- tional graphs. In: Proceedings of the Interna- tional Conference on Robotics and Automa- tion, pp. 8794–8800 (2022)

    2022

  6. [14]

    IEEE Transactions on Robotics 34(4), 901–915 (2018)

    Paranjape, A.A., Chung, S.-J., Kim, K., Shim, D.H.: Robotic herding of a flock of birds using an unmanned aerial vehicle. IEEE Transactions on Robotics 34(4), 901–915 (2018)

    2018

  7. [15]

    IEEE Transactions on Robotics 40, 1706–1723 (2024)

    Zhang, S., Lei, X., Duan, M., Peng, X., Pan, J.: A distributed outmost push approach for multirobot herding. IEEE Transactions on Robotics 40, 1706–1723 (2024)

    2024

  8. [16]

    IEEE Transactions on Robotics 38(6), 3622–3635 (2022)

    Sebasti´ an, E., Montijano, E., Sag¨ u´ es, C.: Adaptive multirobot implicit control of het- erogeneous herds. IEEE Transactions on Robotics 38(6), 3622–3635 (2022)

    2022

  9. [17]

    Automatica 175, 112216 (2025)

    Zheng, C., Mi, Y., Guo, H., Chen, H., Lin, Z., Zhao, S.: Optimal spatial-temporal trian- gulation for bearing-only cooperative motion estimation. Automatica 175, 112216 (2025)

    2025

  10. [18]

    New Journal of Physics 26(1), 012001 (2024)

    Li, K., Li, L., Groß, R., Zhao, S.: A collec- tive perception model for neighbor selection 15 in groups based on visual attention mecha- nisms. New Journal of Physics 26(1), 012001 (2024)

    2024

  11. [19]

    Behavioral ecology and socio- biology 69, 755–764 (2015)

    Hemelrijk, C.K., Zuidam, L., Hildenbrandt, H.: What underlies waves of agitation in starling flocks. Behavioral ecology and socio- biology 69, 755–764 (2015)

    2015

  12. [20]

    Physical Review Research 6(3), 032012 (2024)

    Lama, A., Bernardo, M.: Shepherding and herdability in complex multiagent systems. Physical Review Research 6(3), 032012 (2024)

    2024

  13. [21]

    Animal Behaviour 77(1), 101– 107 (2009)

    Carere, C., Montanino, S., Moreschini, F., Zoratto, F., Chiarotti, F., Santucci, D., All- eva, E.: Aerial flocking patterns of wintering starlings, sturnus vulgaris, under different predation risk. Animal Behaviour 77(1), 101– 107 (2009)

    2009

  14. [22]

    arXiv:2407.15701 (2024)

    Hamandi, M., Khorrami, F., Tzes, A.: Robotic shepherding in cluttered and unknown environments using control barrier functions. arXiv:2407.15701 (2024)

    2024 arXiv

  15. [23]

    IEEE Transactions on Robotics 34(2), 517–525 (2017)

    Pierson, A., Schwager, M.: Controlling non- cooperative herds with robotic herders. IEEE Transactions on Robotics 34(2), 517–525 (2017)

    2017

  16. [24]

    In: Proceedings of the IEEE International Conference on Robotics and Automation, vol

    Lien, J.-M., Bayazit, O.B., Sowell, R.T., Rodriguez, S., Amato, N.M.: Shepherding behaviors. In: Proceedings of the IEEE International Conference on Robotics and Automation, vol. 4, pp. 4159–4164 (2004)

    2004

  17. [25]

    IEEE Control Systems Letters 2(1), 127–132 (2017)

    Licitra, R.A., Bell, Z.I., Doucette, E.A., Dixon, W.E.: Single agent indirect herding of multiple targets: A switched adaptive con- trol approach. IEEE Control Systems Letters 2(1), 127–132 (2017)

    2017

  18. [26]

    In: Proceedings of the IEEE Region 10 Confer- ence, pp

    Fujioka, K., Hayashi, S.: Effective shepherd- ing behaviours using multi-agent systems. In: Proceedings of the IEEE Region 10 Confer- ence, pp. 3179–3182 (2016)

    2016

  19. [27]

    In: Proceedings of the IEEE/RSJ Interna- tional Conference on Intelligent Robots and Systems, pp

    Vo, C., Harrison, J.F., Lien, J.-M.: Behavior- based motion planning for group control. In: Proceedings of the IEEE/RSJ Interna- tional Conference on Intelligent Robots and Systems, pp. 3768–3773 (2009)

    2009

  20. [28]

    Autonomous Robots 45, 613– 631 (2021)

    Song, H., Varava, A., Kravchenko, O., Kragic, D., Wang, M.Y., Pokorny, F.T., Hang, K.: Herding by caging: a formation- based motion planning framework for guiding mobile agents. Autonomous Robots 45, 613– 631 (2021)

    2021

  21. [29]

    IEEE Transactions on Robotics 40, 2729– 2748 (2024)

    Zhang, S., Lei, X., Peng, X., Pan, J.: Hetero- geneous targets trapping with swarm robots by using adaptive density-based interaction. IEEE Transactions on Robotics 40, 2729– 2748 (2024)

    2024

  22. [30]

    In: Proceedings of the IEEE International Conference on Robotics and Automation, pp

    Sebasti´ an, E., Montijano, E.: Multi-robot implicit control of herds. In: Proceedings of the IEEE International Conference on Robotics and Automation, pp. 1601–1607 (2021)

    2021

  23. [31]

    In: 2022 IEEE Conference on Decision and Control, pp

    Grover, J., Mohanty, N., Liu, C., Luo, W., Sycara, K.: Noncooperative herding with con- trol barrier functions: Theory and experi- ments. In: 2022 IEEE Conference on Decision and Control, pp. 80–86 (2022). IEEE

    2022

  24. [32]

    In: International Sym- posium on Distributed Autonomous Robotic Systems, pp

    Mohanty, N., Grover, J., Liu, C., Sycara, K.: Distributed multirobot control for non- cooperative herding. In: International Sym- posium on Distributed Autonomous Robotic Systems, pp. 317–332 (2022). Springer

    2022

  25. [33]

    In: 2019 IEEE 58th Conference on Decision and Control (CDC), pp

    Chipade, V.S., Panagou, D.: Herding an adversarial swarm in an obstacle environ- ment. In: 2019 IEEE 58th Conference on Decision and Control (CDC), pp. 3685–3690 (2019). IEEE

    2019

  26. [34]

    IEEE Robotics and Automation Letters 6(2), 4163–4168 (2021)

    Zhi, J., Lien, J.-M.: Learning to herd agents amongst obstacles: Training robust shepherd- ing behaviors using deep reinforcement learn- ing. IEEE Robotics and Automation Letters 6(2), 4163–4168 (2021)

    2021

  27. [35]

    In: Proceedings of the International Conference on Neural Information Process- ing, pp

    Nguyen, H.T., Nguyen, T.D., Garratt, M., Kasmarik, K., Anavatti, S., Barlow, M., Abbass, H.A.: A deep hierarchical reinforce- ment learner for aerial shepherding of ground swarms. In: Proceedings of the International Conference on Neural Information Process- ing, pp. 658–669 (2019) 16

    2019

  28. [36]

    Springer, Singapore (2025)

    Zhao, S.: Mathematical Foundations of Reinforcement Learning. Springer, Singapore (2025)

    2025

  29. [37]

    In: Proceedings of the IEEE Conference on Decision and Control, pp

    Taylor, A.J., Ong, P., Molnar, T.G., Ames, A.D.: Safe backstepping with control bar- rier functions. In: Proceedings of the IEEE Conference on Decision and Control, pp. 5775–5782 (2022)

    2022

  30. [38]

    In: 2016 IEEE 55th Conference on Decision and Control (CDC), pp

    Bell, Z., Parikh, A., Nezvadovitz, J., Dixon, W.E.: Adaptive control of a surface marine craft with parameter identification using inte- gral concurrent learning. In: 2016 IEEE 55th Conference on Decision and Control (CDC), pp. 389–394 (2016). IEEE

    2016

  31. [39]

    International Journal of Adap- tive Control and Signal Processing 33(12), 1775–1787 (2019)

    Parikh, A., Kamalapurkar, R., Dixon, W.E.: Integral concurrent learning: Adaptive con- trol with parameter convergence using finite excitation. International Journal of Adap- tive Control and Signal Processing 33(12), 1775–1787 (2019)

    2019

  32. [40]

    IEEE Transactions on Robotics 35(4), 847–860 (2019)

    Licitra, R.A., Bell, Z.I., Dixon, W.E.: Single- agent indirect herding of multiple targets with uncertain dynamics. IEEE Transactions on Robotics 35(4), 847–860 (2019)

    2019

  33. [41]

    IEEE Transactions on Automatic Control 62(8), 3861–3876 (2016)

    Ames, A.D., Xu, X., Grizzle, J.W., Tabuada, P.: Control barrier function based quadratic programs for safety critical systems. IEEE Transactions on Automatic Control 62(8), 3861–3876 (2016)

    2016

  34. [42]

    IF AC- PapersOnLine 48(27), 54–61 (2015)

    Xu, X., Tabuada, P., Grizzle, J.W., Ames, A.D.: Robustness of control barrier func- tions for safety critical control. IF AC- PapersOnLine 48(27), 54–61 (2015)

    2015

  35. [43]

    Systems & Control Letters 13(2), 117–123 (1989)

    Sontag, E.D.: A ’universal’ construction of artstein’s theorem on nonlinear stabilization. Systems & Control Letters 13(2), 117–123 (1989)

    1989

  36. [44]

    arXiv:2403.03030 (2024)

    Li, M., Sun, Z., Weiland, S.: Unifying con- troller design for stabilizing nonlinear sys- tems with norm-bounded control inputs. arXiv:2403.03030 (2024)

    2024 arXiv

  37. [45]

    IEEE Transactions on Robotics 33(3), 661–674 (2017)

    Wang, L., Ames, A.D., Egerstedt, M.: Safety barrier certificates for collisions-free multirobot systems. IEEE Transactions on Robotics 33(3), 661–674 (2017)

    2017

  38. [46]

    In: Proceedings of the European Control Conference, pp

    Ames, A.D., Coogan, S., Egerstedt, M., Notomista, G., Sreenath, K., Tabuada, P.: Control barrier functions: Theory and appli- cations. In: Proceedings of the European Control Conference, pp. 3420–3431 (2019). IEEE

    2019

  39. [47]

    IEEE Control Systems Let- ters 5(2), 731–736 (2020)

    Reis, M.F., Aguiar, A.P., Tabuada, P.: Con- trol barrier function-based quadratic pro- grams introduce undesirable asymptotically stable equilibria. IEEE Control Systems Let- ters 5(2), 731–736 (2020)

    2020

  40. [48]

    Auto- matica 159, 111359 (2024)

    Tan, X., Dimarogonas, D.V.: On the unde- sired equilibria induced by control barrier function based quadratic programs. Auto- matica 159, 111359 (2024)

    2024

  41. [49]

    In: Pro- ceedings of the 2024 International Conference on Control Automation, pp

    Ma, Z., Liang, J., Wang, H., Guo, S., Huo, P., Zhang, Y., Zhao, S.: Omnibot: A scalable vision-based robot swarm platform. In: Pro- ceedings of the 2024 International Conference on Control Automation, pp. 975–980 (2024) 17

    2024

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.