pith. sign in

arxiv: 2605.16097 · v1 · pith:BTREPN7Znew · submitted 2026-05-15 · 💻 cs.MA

Multi-Agent Cooperative Transportation: Optimal and Efficient Task Allocation and Path Finding

Ning Zhou , Nikolai W.F. Bode , Edmund R. Hunt This is my paper

Pith reviewed 2026-05-19 18:25 UTC · model grok-4.3

classification 💻 cs.MA
keywords multi-agent path findingtask allocationcooperative transportationconflict-based searchmulti-robot systemsteam formationcombinatorial search
0
0 comments X

The pith

The CT-TCBS algorithm with incremental expansion solves cooperative multi-robot transport by pruning the dominant search space for team assignments while integrating collision-free paths.

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

The paper formalizes the CT-TAPF problem so that groups of robots can jointly move large objects that exceed single-agent capacity, while still producing collision-free routes. It introduces an optimal solver that grows team formations step by step rather than enumerating every possible grouping in advance. This matters for warehouse and logistics settings where rigid single-robot task models leave many practical jobs impossible. The work also supplies faster but non-optimal alternatives that pick the next task from a global view of difficulty. If the pruning works as claimed, it removes the main computational barrier that previously blocked integrated team formation and path planning.

Core claim

The authors claim that the Cooperative Transportation Task Conflict-Based Search (CT-TCBS) solver, equipped with an Incremental Expansion strategy, optimally resolves the integrated problem of team formation, task assignment, and collision-free path finding for cooperative transportation. The strategy expands candidate teams only when lower-cost conflicts force it, thereby avoiding the full combinatorial explosion of possible groupings. Empirical tests show this pruning succeeds where a naive enumeration of all team combinations fails within time limits. The same evaluation identifies a task-conflict expansion dilemma in which stronger conflict resolvers can slow the overall search. Sub- and

What carries the argument

The Incremental Expansion strategy inside CT-TCBS, which defers full enumeration of team formations until conflict resolution requires it, thereby pruning the task-allocation search space while preserving optimality guarantees.

If this is right

  • The incremental expansion strategy succeeds in pruning the dominant task-allocation component of the search.
  • Sophisticated conflict resolvers that help standalone large-agent path finding can increase overall runtime inside the integrated CT-TAPF setting.
  • Task-centric sub-optimal solvers that select the next assignment by global difficulty metrics achieve better quality-runtime trade-offs than agent-centric baselines.
  • The CT-TAPF formalization extends existing MAPF and TAPF frameworks to problems that require simultaneous multi-agent effort on a single physical object.

Where Pith is reading between the lines

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

  • The same incremental-expansion pattern could be tested on other multi-agent problems where team composition must be decided alongside routing.
  • Real-world deployment would require checking whether the reported runtime gains survive sensor noise and imperfect communication between robots.
  • Warehouse planners could use the global-difficulty task selectors as a fast heuristic when optimal solutions are not required.
  • Extending the model to include heterogeneous robot capabilities or time-varying obstacles would be a direct next step consistent with the current integration of allocation and planning.

Load-bearing premise

That the chosen conflict-resolution heuristics allow the incremental expansion to discard large portions of the team-formation space without losing any optimal solutions.

What would settle it

A benchmark instance in which the incremental-expansion version of CT-TCBS returns a higher-cost solution than exhaustive enumeration within the same time bound.

Figures

Figures reproduced from arXiv: 2605.16097 by Edmund R. Hunt, Nikolai W.F. Bode, Ning Zhou.

Figure 1
Figure 1. Figure 1: A CT-TAPF problem instance with four agents and [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: The incremental search tree of CT-TCBS for an [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: An illustration of the Combinatorial explosion [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Success rates on collision-rich set with different [PITH_FULL_IMAGE:figures/full_fig_p007_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Solution quality (optimality gap, %) of WT vs. BT on [PITH_FULL_IMAGE:figures/full_fig_p008_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: Success rates of suboptimal task selectors (BT vs. [PITH_FULL_IMAGE:figures/full_fig_p008_6.png] view at source ↗
read the original abstract

Multi-robot systems are integral to modern logistics, but their capabilities are often limited to tasks executable by individual agents. This paper addresses a critical gap in existing frameworks like Multi-Agent Path Finding (MAPF) and Task Allocation and Path Finding (TAPF), which lack true cooperation for transporting large items that require multiple agents. To this end, we formalise the Cooperative Transportation Task Allocation and Path Finding (CT-TAPF) problem, which integrates team formation, task assignment, and collision-free pathfinding. We present an optimal solver, Cooperative Transportation Task Conflict-Based Search (CT-TCBS), which features a novel Incremental Expansion strategy to tackle the combinatorial explosion inherent in team formation. Recognising the computational cost of optimality, we also develop a family of sub-optimal solvers that employ a global, task-centric perspective, selecting the next task to assign based on a global difficulty metric (Best Task or Worst Task). Our comprehensive empirical evaluation demonstrates three key findings: (1) the incremental expansion strategy significantly outperforms the naive combinatorial approach by successfully pruning the dominant task-allocation search space; (2) we identify a task-conflict expansion dilemma, where sophisticated conflict resolvers effective for large-agent pathfinding subproblems can be detrimental in the integrated CT-TAPF setting; and (3) our proposed sub-optimal solvers establish a new, more efficient frontier on the solution quality-runtime spectrum compared to "nn-" agent-centric baselines. This work provides a foundational framework and a set of effective algorithms for a new, practical class of cooperative multi-agent problems.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

2 major / 2 minor

Summary. The manuscript formalizes the Cooperative Transportation Task Allocation and Path Finding (CT-TAPF) problem, which extends MAPF and TAPF to handle tasks requiring multi-agent teams for transporting large items. It proposes an optimal solver CT-TCBS that uses a novel Incremental Expansion strategy to mitigate the combinatorial explosion in team formation and task allocation, while also introducing sub-optimal solvers that select tasks using global difficulty metrics (Best Task or Worst Task). The paper reports three empirical findings: the incremental expansion prunes the task-allocation search space more effectively than naive combinatorial search, a task-conflict expansion dilemma exists where sophisticated conflict resolvers can harm performance in the integrated setting, and the sub-optimal solvers achieve a new efficiency frontier relative to agent-centric baselines.

Significance. If the optimality of CT-TCBS and the pruning-without-loss claim hold under the integrated conflict resolution, this work provides a foundational algorithmic framework for a practically relevant class of cooperative multi-agent transportation problems in logistics. The identification of the task-conflict expansion dilemma and the global task-centric sub-optimal solvers could usefully inform future designs that interleave team formation with pathfinding.

major comments (2)
  1. [Abstract and §3] Abstract and §3 (CT-TCBS description): the claim that CT-TCBS is optimal and that incremental expansion prunes the dominant task-allocation search space without loss of optimality lacks an explicit admissibility or completeness argument. The abstract itself highlights the task-conflict expansion dilemma, in which resolver choice determines surviving branches; without a proof that the expansion order remains complete and admissible for arbitrary team sizes and conflict patterns, the pruning claim is not yet load-bearing.
  2. [Empirical evaluation] Empirical evaluation section: the three key findings rest on comparisons whose experimental setup, instance generation, metrics (e.g., success rate, runtime, solution cost), baselines, number of trials, and statistical tests are not described in sufficient detail to allow verification of the pruning effectiveness or the new efficiency frontier.
minor comments (2)
  1. [Algorithm description] Notation for team-formation expansions and conflict-based search interleaving could be clarified with a small pseudocode snippet or diagram to make the incremental strategy easier to follow.
  2. [Abstract] The abstract would benefit from one sentence explicitly stating the formal definition of CT-TAPF (e.g., input/output and objective) before describing the solvers.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive feedback and the recommendation for major revision. We will revise the manuscript to strengthen the theoretical claims and improve the experimental description.

read point-by-point responses

  1. Referee: [Abstract and §3] Abstract and §3 (CT-TCBS description): the claim that CT-TCBS is optimal and that incremental expansion prunes the dominant task-allocation search space without loss of optimality lacks an explicit admissibility or completeness argument. The abstract itself highlights the task-conflict expansion dilemma, in which resolver choice determines surviving branches; without a proof that the expansion order remains complete and admissible for arbitrary team sizes and conflict patterns, the pruning claim is not yet load-bearing.

    Authors: We agree that an explicit admissibility and completeness argument would strengthen the presentation. The manuscript currently establishes optimality by construction: CT-TCBS extends standard CBS to a joint space of team formations and task allocations, with Incremental Expansion enumerating team cardinalities in increasing order while preserving all feasible combinations for subsequent conflict resolution. The task-conflict expansion dilemma is reported as an empirical observation on practical performance trade-offs rather than a theoretical barrier to completeness. To address the concern directly, we will insert a formal subsection in §3 that proves the search remains complete and admissible for arbitrary team sizes and conflict patterns, independent of the low-level resolver. revision: yes

  2. Referee: [Empirical evaluation] Empirical evaluation section: the three key findings rest on comparisons whose experimental setup, instance generation, metrics (e.g., success rate, runtime, solution cost), baselines, number of trials, and statistical tests are not described in sufficient detail to allow verification of the pruning effectiveness or the new efficiency frontier.

    Authors: We concur that greater detail is required for reproducibility. In the revised manuscript we will augment the experimental section with: (i) explicit instance-generation parameters (map sizes, agent counts, task distributions, and team-size requirements for large items); (ii) precise metric definitions (success rate under a 300 s timeout, wall-clock runtime, and sum-of-costs solution quality); (iii) complete baseline specifications including the exact “nn-” agent-centric variants; (iv) the number of independent trials per configuration (50); and (v) the statistical procedures employed (means with standard deviations and Wilcoxon signed-rank tests). These additions will allow independent verification of the reported pruning gains and efficiency frontier. revision: yes

Circularity Check

0 steps flagged

No circularity detected in CT-TAPF formalization or CT-TCBS derivation

full rationale

The paper introduces a novel problem definition for Cooperative Transportation Task Allocation and Path Finding (CT-TAPF) and constructs the CT-TCBS solver around an incremental expansion strategy for handling team formation combinatorics. All central claims, including outperformance over naive combinatorial search and identification of the task-conflict expansion dilemma, rest on explicit algorithmic descriptions and empirical comparisons to external baselines rather than any self-referential definitions, fitted parameters renamed as predictions, or load-bearing self-citations. No equations or steps reduce by construction to the paper's own inputs; the derivation chain remains self-contained against standard MAPF/TAPF frameworks.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

The work rests on standard assumptions from multi-agent path finding and domain assumptions about cooperative physical tasks; no free parameters or invented entities are introduced.

axioms (2)
  • standard math Collision-free paths exist and can be found via conflict-based search extensions
    Inherited from prior MAPF literature and used as foundation for CT-TCBS.
  • domain assumption Agents are capable of forming teams to jointly transport items
    Core modeling choice that defines the CT-TAPF problem.

pith-pipeline@v0.9.0 · 5810 in / 1180 out tokens · 43506 ms · 2026-05-19T18:25:21.173511+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

31 extracted references · 31 canonical work pages

  1. [1]

    Olukunle Amoo, Enoch Sodiya, Uchenna Umoga, and Akoh Atadoga. 2024. AI- driven Warehouse Automation: A Comprehensive Review of Systems.GSC Advanced Research and Reviews18 (Feb. 2024), 272–282. https://doi.org/10.30574/ gscarr.2024.18.2.0063

    work page 2024

  2. [2]

    Stuckey, Liron Cohen, Jiaoyang Li, and Sven Koenig

    Eli Boyarski, Ariel Felner, Daniel Harabor, Peter J. Stuckey, Liron Cohen, Jiaoyang Li, and Sven Koenig. 2020. Iterative-Deepening Conflict-Based Search. InProceed- ings of the Twenty-Ninth International Joint Conference on Artificial Intelligence. Association for the Advancement of Artificial Intelligence (AAAI), 4084–4090. https://doi.org/10.24963/ijcai...

    work page doi:10.24963/ijcai.2020/565 2020

  3. [3]

    Eli Boyarski, Ariel Felner, Roni Stern, Guni Sharon, Oded Betzalel, David Tolpin, and Eyal Shimony. 2015. ICBS: The Improved Conflict-Based Search Algorithm for Multi-Agent Pathfinding.Proceedings of the International Symposium on Combinatorial Search6, 1 (2015), 223–225. https://doi.org/10.1609/socs.v6i1.18343

    work page doi:10.1609/socs.v6i1.18343 2015

  4. [4]

    Jorge Cortés and Magnus Egerstedt. 2017. Coordinated Control of Multi-Robot Systems: A Survey.SICE Journal of Control, Measurement, and System Integration (Nov. 2017). https://doi.org/10.9746/jcmsi.10.495

    work page doi:10.9746/jcmsi.10.495 2017

  5. [5]

    Ariel Felner, Jiaoyang Li, Eli Boyarski, Hang Ma, Liron Cohen, Thirunarayanapu- ram Krishnamachari Satish Kumar, and Sven Koenig. 2018. Adding Heuristics to Conflict-Based Search for Multi-Agent Path Finding.Proceedings of the Interna- tional Conference on Automated Planning and Scheduling28 (June 2018), 83–87. https://doi.org/10.1609/icaps.v28i1.13883

    work page doi:10.1609/icaps.v28i1.13883 2018

  6. [6]

    Holte, and Jonathan Schaeffer

    Meir Goldenberg, Ariel Felner, Roni Stern, Guni Sharon, Nathan Sturtevant, Robert C. Holte, and Jonathan Schaeffer. 2014. Enhanced Partial Expansion A*.Journal of Artificial Intelligence Research50 (May 2014), 141–187. https: //doi.org/10.1613/jair.4171

    work page doi:10.1613/jair.4171 2014

  7. [7]

    Nir Greshler, Ofir Gordon, Oren Salzman, and Nahum Shimkin. 2021. Cooperative Multi-Agent Path Finding: Beyond Path Planning and Collision Avoidance. In 2021 International Symposium on Multi-Robot and Multi-Agent Systems (MRS). 20–28. https://doi.org/10.1109/MRS50823.2021.9620590

    work page doi:10.1109/mrs50823.2021.9620590 2021

  8. [8]

    Christian Henkel, Jannik Abbenseth, and Marc Toussaint. 2019. An Optimal Algorithm to Solve the Combined Task Allocation and Path Finding Problem. In 2019 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS). 4140–4146. https://doi.org/10.1109/IROS40897.2019.8968096

    work page doi:10.1109/iros40897.2019.8968096 2019

  9. [9]

    Durham, and Nora Ayanian

    Wolfgang Hoenig, Sven Kiesel, Andrew Tinka, James W. Durham, and Nora Ayanian. 2018. Conflict-Based Search with Optimal Task Assignment.Proceedings of the International Joint Conference on Autonomous Agents and Multiagent Systems (Jan. 2018)

    work page 2018

  10. [10]

    He Jiang, Yulun Zhang, Rishi Veerapaneni, and Jiaoyang Li. 2024. Scaling Lifelong Multi-Agent Path Finding to More Realistic Settings: Research Challenges and Opportunities.Proceedings of the International Symposium on Combinatorial Search17 (June 2024), 234–242. https://doi.org/10.1609/socs.v17i1.31565

    work page doi:10.1609/socs.v17i1.31565 2024

  11. [11]

    Stuckey, and Sven Koenig

    Jiaoyang Li, Zhe Chen, Daniel Harabor, Peter J. Stuckey, and Sven Koenig. 2022. MAPF-LNS2: Fast Repairing for Multi-Agent Path Finding via Large Neighbor- hood Search.Proceedings of the AAAI Conference on Artificial Intelligence36, 9 (June 2022), 10256–10265. https://doi.org/10.1609/aaai.v36i9.21266

    work page doi:10.1609/aaai.v36i9.21266 2022

  12. [12]

    Stuckey, Hang Ma, and Sven Koenig

    Jiaoyang Li, Graeme Gange, Daniel Harabor, Peter J. Stuckey, Hang Ma, and Sven Koenig. 2020. New Techniques for Pairwise Symmetry Breaking in Multi-Agent Path Finding.Proceedings of the International Conference on Automated Planning and Scheduling30 (June 2020), 193–201. https://doi.org/10.1609/icaps.v30i1.6661

    work page doi:10.1609/icaps.v30i1.6661 2020

  13. [13]

    Stuckey, Ariel Felner, Hang Ma, and Sven Koenig

    Jiaoyang Li, Daniel Harabor, Peter J. Stuckey, Ariel Felner, Hang Ma, and Sven Koenig. 2019. Disjoint Splitting for Multi-Agent Path Finding with Conflict-Based Search.Proceedings of the International Conference on Automated Planning and Scheduling29 (July 2019), 279–283. https://doi.org/10.1609/icaps.v29i1.3487

    work page doi:10.1609/icaps.v29i1.3487 2019

  14. [14]

    Jiaoyang Li, Pavel Surynek, Ariel Felner, Hang Ma, Thirunarayanapuram Krish- namachari Satish Kumar, and Sven Koenig. 2019. Multi-Agent Path Finding for Large Agents.Proceedings of the AAAI Conference on Artificial Intelligence33, 01 (July 2019), 7627–7634. https://doi.org/10.1609/aaai.v33i01.33017627

    work page doi:10.1609/aaai.v33i01.33017627 2019

  15. [15]

    Hang Ma, Jiaoyang Li, Thirunarayanapuram Krishnamachari Satish Kumar, and Sven Koenig. 2017. Lifelong Multi-Agent Path Finding for Online Pickup and Delivery Tasks. InProceedings of the 16th Conference on Autonomous Agents and MultiAgent Systems (AAMAS ’17). International Foundation for Autonomous Agents and Multiagent Systems, Richland, SC, 837–845

    work page 2017

  16. [16]

    Hang Ma, Craig Tovey, Guni Sharon, Thirunarayanapuram Krishnamachari Kumar, and Sven Koenig. 2016. Multi-Agent Path Finding with Payload Transfers and the Package-Exchange Robot-Routing Problem.Proceedings of the AAAI Conference on Artificial Intelligence30, 1 (March 2016). https://doi.org/10.1609/ aaai.v30i1.10409

    work page 2016

  17. [17]

    Naoki Mizumoto, Katsuhide Fujita, Yoshihiro Ueda, and Takayoshi Mori. 2025. Lifelong MAPF and Task Assignment Considering Workers in Warehouses. InPro- ceedings of the 6th International Workshop on Multi-Agent Path Finding (WoMAPF). Workshop at the 39th AAAI Conference on Artificial Intelligence (AAAI)

    work page 2025

  18. [18]

    Keisuke Okumura. 2023. LaCAM: Search-Based Algorithm for Quick Multi-Agent Pathfinding.Proceedings of the AAAI Conference on Artificial Intelligence37, 10 (June 2023), 11655–11662. https://doi.org/10.1609/aaai.v37i10.26377

    work page doi:10.1609/aaai.v37i10.26377 2023

  19. [19]

    David Portugal and Rui Rocha. 2011. A Survey on Multi-robot Patrolling Algo- rithms. InTechnological Innovation for Sustainability, Luis M. Camarinha-Matos (Ed.). Springer, Berlin, Heidelberg, 139–146. https://doi.org/10.1007/978-3-642- 19170-1_15

    work page doi:10.1007/978-3-642- 2011

  20. [20]

    Sartaj Sahni and Teofilo Gonzalez. 1976. P-Complete Approximation Problems. J. ACM23, 3 (July 1976), 555–565. https://doi.org/10.1145/321958.321975

    work page doi:10.1145/321958.321975 1976

  21. [21]

    Artificial Intelligence219, 40– 18 I

    Guni Sharon, Roni Stern, Ariel Felner, and Nathan R. Sturtevant. 2015. Conflict- Based Search for Optimal Multi-Agent Pathfinding.Artificial Intelligence219 (2015), 40–66. https://doi.org/10.1016/j.artint.2014.11.006

    work page doi:10.1016/j.artint.2014.11.006 2015

  22. [22]

    Guni Sharon, Roni Stern, Meir Goldenberg, and Ariel Felner. 2013. The Increasing Cost Tree Search for Optimal Multi-Agent Pathfinding.Artificial Intelligence195 (Feb. 2013), 470–495. https://doi.org/10.1016/j.artint.2012.11.006

    work page doi:10.1016/j.artint.2012.11.006 2013

  23. [23]

    David Silver. 2005. Cooperative Pathfinding.Proceedings of the AAAI Conference on Artificial Intelligence and Interactive Digital Entertainment1, 1 (2005), 117–122. https://doi.org/10.1609/aiide.v1i1.18726

    work page doi:10.1609/aiide.v1i1.18726 2005

  24. [24]

    Trevor Standley. 2010. Finding Optimal Solutions to Cooperative Pathfinding Problems.Proceedings of the AAAI Conference on Artificial Intelligence24, 1 (July 2010), 173–178. https://doi.org/10.1609/aaai.v24i1.7564

    work page doi:10.1609/aaai.v24i1.7564 2010

  25. [25]

    Roni Stern, Nathan Sturtevant, Ariel Felner, Sven Koenig, Hang Ma, Thayne Walker, Jiaoyang Li, Dor Atzmon, Liron Cohen, Thirunarayanapuram Krishna- machari Kumar, Roman Barták, and Eli Boyarski. 2019. Multi-Agent Pathfinding: Definitions, Variants, and Benchmarks.Proceedings of the International Sympo- sium on Combinatorial Search10, 1 (2019), 151–158. ht...

    work page doi:10.1609/socs 2019

  26. [26]

    Sturtevant

    Nathan R. Sturtevant. 2012. Benchmarks for Grid-Based Pathfinding.IEEE Transactions on Computational Intelligence and AI in Games4, 2 (June 2012), 144–148. https://doi.org/10.1109/TCIAIG.2012.2197681

    work page doi:10.1109/tciaig.2012.2197681 2012

  27. [27]

    Pavel Surynek, Ariel Felner, Roni Stern, and Eli Boyarski. 2016. Efficient SAT Approach to Multi-Agent Path Finding under the Sum of Costs Objective. In Proceedings of the Twenty-second European Conference on Artificial Intelligence (ECAI’16). IOS Press, NLD, 810–818. https://doi.org/10.3233/978-1-61499-672-9- 810

    work page doi:10.3233/978-1-61499-672-9- 2016

  28. [28]

    Yimin Tang, Zhongqiang Ren, Jiaoyang Li, and Katia Sycara. 2023. Solving Multi- Agent Target Assignment and Path Finding with a Single Constraint Tree. In2023 International Symposium on Multi-Robot and Multi-Agent Systems (MRS). 8–14. https://doi.org/10.1109/MRS60187.2023.10416794

    work page doi:10.1109/mrs60187.2023.10416794 2023

  29. [29]

    Glenn Wagner and Howie Choset. 2011. M*: A Complete Multirobot Path Plan- ning Algorithm with Performance Bounds. InInternational Conference on In- telligent Robots and Systems (IROS) (IEEE International Conference on Intelligent Robots and Systems). 3260–3267. https://doi.org/10.1109/IROS.2011.6048671

    work page doi:10.1109/iros.2011.6048671 2011

  30. [30]

    Jingjin Yu and Steven LaValle. 2013. Structure and Intractability of Optimal Multi- Robot Path Planning on Graphs.Proceedings of the AAAI Conference on Artificial Intelligence27, 1 (June 2013), 1443–1449. https://doi.org/10.1609/aaai.v27i1.8541

    work page doi:10.1609/aaai.v27i1.8541 2013

  31. [31]

    Jingjin Yu and Steven M. LaValle. 2013. Planning Optimal Paths for Multiple Robots on Graphs. In2013 IEEE International Conference on Robotics and Automa- tion. 3612–3617. https://doi.org/10.1109/ICRA.2013.6631084

    work page doi:10.1109/icra.2013.6631084 2013