Pith. sign in

REVIEW 5 major objections 7 minor 34 references

Investigating the Impact of Communication-Induced Action Space on Exploration of Unknown Environments with Decentralized Multi-Agent Reinforcement Learning

T0 review · 5 major / 7 minor · reviewed 2026-08-10 · deepseek-v4-flash

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

Pith's one-line read Giving each exploring robot a discrete 'communicate' action, trained with a reward that balances sharing and exploring, lets a decentralized team map an unknown environment faster and with less overlap.

desk verdict The Gymnasium step improvements are real but modest, the Gazebo table contradicts the headline claim, and the lack of stats/code makes the paper a conditional at best. read the letter →

arxiv 2412.20075 v1 pith:TA2VHL6P submitted 2024-12-28 cs.RO

Gabriele Calzolari (1) , Vidya Sumathy (1) , Christoforos Kanellakis (1) , George Nikolakopoulos (1) ((1) Lule{aa} University of Technology) This is my paper · ORCID
classification cs.RO
keywords multi-agentreinforcementlearningdecentralizedexplorationcommunication-inducedactionspaceHAPPOrewardshapingoccupancygridmappingmulti-robotunknownenvironments
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 a team of homogeneous mobile robots exploring an unknown environment maps it more efficiently when each robot's action space includes a discrete 'communicate' action, letting it choose when to share its locally built map with teammates instead of always moving. The agents are trained with the heterogeneous-agent proximal policy optimization algorithm, using reward functions that balance exploration progress against the value of shared information. Across 100 randomized test environments in Gymnasium, the best communication-aware reward (Study Case 4) reached 90% map coverage in a mean of 454 steps versus 557 for the no-communication baseline, and succeeded in 86% of environments versus 60% for the baseline, while also lowering overlap between agent maps. The paper argues this matters because real-world multi-robot exploration is constrained by communication limits, so letting agents decide when to transmit can reduce redundant coverage and speed up mapping.

What carries the argument

The central mechanism is the communication-induced action space and the reward functions that make communication worth choosing. Each agent keeps three maps: an agent-specific map of its own discoveries, a collaborative map that merges information received from teammates, and a transmission map of agents currently in communication range. When an agent performs the 'communicate' action, its collaborative map is merged with those of all agents in the same communication network, per the identity $M_{i_k,\mathrm{co}} = \bigcup_{i_j \in C} M_{i_j,\mathrm{co}}$, so shared information propagates through chains of agents. The reward functions combine an exploration term proportional to newly discovered cells, a communication term proportional to the knowledge gained from the merged maps, and penalties for staying still or collisions; Study Case 4's history-dependent coefficient $p_k$ conditions the communication reward on how much the agent has discovered since its last communication, which is what makes the learned policy share data at informative moments.

What would settle it

Run the same training and evaluation with a communication channel that drops a fraction of messages or charges an energy cost per transmission; if the Study Case 4 policy no longer beats the no-communication baseline in mean steps and robustness, the central claim would be falsified for realistic settings.

Watch

Extended reading notes

Core claim

The paper's central claim is that adding a communication action to the action space, together with a reward term that rewards information gain from communication, improves decentralized multi-agent exploration. In the proposed formulation each agent's action set is {up, up-right, right, down-right, down, down-left, left, up-left, stay, communicate}, and selecting 'communicate' shares the agent's collaborative map with every agent in the same direct or indirect communication network, after which all participating maps are merged. The authors compare four reward functions; Study Case 4, which weights the communication reward by the average number of cells the agent has discovered since its last communication with each network member, achieves the best results: lower mean steps to 90% coverage (454 vs 557), higher robustness (86% vs 60%), and reduced overlap as measured by the Jaccard index. The conclusion states that agents trained with communication-based rewards reduce both overlap in explored areas and the steps needed to explore the environment.

Load-bearing premise

The load-bearing premise is that a 'communicate' action succeeds reliably whenever agents are within range, with no latency, bandwidth, packet loss, or energy cost; under realistic communication imperfections the learned balance between moving and sharing could change.

Editorial extensions

If this is right

  • A decentralized multi-robot team can learn an implicit division of labor — when to move and when to share — without a central coordinator, purely from the reward signal.
  • Rewarding communication in proportion to newly shared information reduces duplicated coverage between agents, as measured by lower Jaccard overlap.
  • The best-tuned communication reward (Study Case 4) improves robustness across unseen environments, from 60% to 86% success.
  • The policy transfers from the Gymnasium grid world to a Gazebo simulation with TurtleBot3 Burger robots and laser-SLAM occupancy grids, though performance depends on the starting configuration.

Reading between the lines

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

  • The simulations assume an ideal communication channel with no latency, bandwidth, or energy cost, so the reported gains likely represent an upper bound; a natural extension would run the same reward scheme under packet-loss or power-budget models.
  • The history-dependent coefficient in Study Case 4 suggests a general principle: communication rewards should be gated by what the transmitting agent has learned since the last exchange, which could transfer to other cooperative MARL tasks beyond mapping.
  • A sensitivity sweep over the communication range $r_c$ would reveal how dependent the benefit is on connectivity; the paper fixes $r_c = 5$ and does not test sparser or denser networks.
  • Because the policy broadcasts entire collaborative maps, a message-content variant that shares only frontier or high-uncertainty cells might achieve similar gains with far less data, aligning with the paper's real-world communication constraints.
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.

Referee Report

5 major / 7 minor

Summary. This paper proposes adding a discrete 'communicate' action to the action space of decentralized multi-agent reinforcement learning (D-MARL) for exploring unknown environments. Agents are trained with HAPPO and a shared critic, using three map-based observations (agent-specific, collaborative, transmission). Four reward functions are compared as study cases, with communication-aware rewards in Cases 2-4. Training is done in Gymnasium/PettingZoo, and evaluation is performed both in Gymnasium (100 random environments) and in a Gazebo simulation with four TurtleBot3 Burgers. The paper claims that communication-induced action spaces reduce exploration overlap and the number of steps needed to explore, with Study Case 4 performing best.

Significance. If the result holds, the contribution is a conceptually simple but potentially useful modification: making communication an explicit, learnable action with reward shaping that balances exploration and information sharing. The paper also provides a ROS2/Gazebo framework that bridges RL policies to simulated robots, which is a practical resource. The strengths are the explicit reward decomposition, the HAPPO algorithm pseudocode, and the full Gazebo validation setup. However, the central claim is currently not established: the Gymnasium results lack statistical support, and the Gazebo results in Table 5 contradict the paper's own validation sentence. No code, random seeds, or statistical significance tests are provided, and the robustness metric is not precisely defined. The significance of the work therefore depends on substantial additional evidence.

major comments (5)
  1. [§5.3.3, Table 5] The sentence "in all the setups the policies that were trained by rewarding inter-agent communication adequately, outperformed the first study case as shown by the map coverage ratio" is contradicted by Table 5. In Setup 1 only Study Case 4 (91.4%) exceeds the Case 1 value (76.8%), while Case 2 (37.7%) and Case 3 (53.2%) are lower; in Setup 2 only Case 4 (52.2%) exceeds Case 1 (33.7%); in Setup 3 only Case 3 (69.8%) exceeds Case 1 (64.6%); in Setup 4 only Case 2 (75.8%) exceeds Case 1 (63.3%). Across the twelve communication-policy runs, exactly six have higher map coverage than the baseline. This internal inconsistency invalidates the validation claim in its current form.
  2. [§5.2, Table 3] The reported differences are not supported by any statistical significance test or confidence interval, and no random seeds are reported. For example, the number of steps for Case 1 is 557.00 ± 169.07 and for Case 4 is 454.33 ± 184.85, and the Jaccard index for Case 1 is 0.28 ± 0.07 versus 0.25 ± 0.05 for Cases 2-4; the overlapping standard deviations do not establish a robust reduction. The headline robustness values (0.86 versus 0.60) are single ratios with no uncertainty quantification, so the claim that communication improves robustness is not statistically supported.
  3. [§5.1] The robustness metric is defined as the "ratio of the successfully explored arena over the total number of testing environments," but the criterion for a "successfully explored" environment is not specified in the metric definition. The earlier text mentions a termination condition of p = 90% of free grids explored, which is not necessarily the same as a success threshold for the robustness calculation. Without an explicit success criterion, the 86% versus 60% comparison in Table 3 is uninterpretable.
  4. [§6] The conclusion states that "the exploration task performed by agents trained with reward functions implementing inter-agent communication reduces both the overlap in the explored areas and the time steps needed to explore the environment," but this is not supported by the Gazebo data in Table 5. Several Jaccard coefficients for Study Case 4 are larger than the corresponding Case 1 values (e.g., Setup 1: Ji1,i2 = 0.295 vs 0.095, Ji1,i3 = 0.391 vs 0.262, Ji1,i4 = 0.294 vs 0.071), and in half of the communication-policy runs the map coverage is lower than the baseline. The conclusion overstates the evidence and should be revised or backed by additional experiments.
  5. [§2.4] The communication model is idealized: a 'communicate' action succeeds whenever agents are within range, with no latency, bandwidth limits, packet loss, or energy cost, even though the introduction (Section 1) motivates the work precisely by citing real-world data transmission constraints such as signal latency and bandwidth. The paper does not test how the learned policies behave under constrained or imperfect communication, so the claimed real-world applicability is not demonstrated. Either the simulation should model such constraints or the claims should be tempered accordingly.
minor comments (7)
  1. [Eq. (14)] The summation condition "j>k" is undefined because the outer sum uses the index ik; the notation should be clarified, for instance by summing over unordered pairs of agent indices.
  2. [Table 2] The table headings "na ne ns nb" are not explained in the caption; they presumably denote number of agents, number of episodes, steps per episode, and batch size, but these should be stated explicitly.
  3. [Figure 6] The training reward curves are presented as single traces without confidence intervals or multiple seeds, making it difficult to assess convergence variability across training runs.
  4. [Data Availability] The Data Availability statement says "This manuscript does not report data generation or analysis," which is inconsistent with the simulation results reported in Tables 3 and 5; this statement should be corrected.
  5. [Table 1] The heading contains a typo: "T able 1" should be "Table 1."
  6. [§5.2] The claim that "study cases 2 and 4 succeeded in lowering the median of the number of steps by more than 100 steps" should report the actual median values, since Table 3 only lists means and standard deviations.
  7. [§5.1] The notation "p = 90%" is potentially confusing because p is also used as a coefficient in Eqs. (10)-(12); please use distinct notation for the exploration threshold.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the claims are empirical comparisons of hand-designed reward functions, not predictions derived from fitted inputs or from self-citations.

full rationale

The paper's central claim is that adding a discrete communicate action and communication-aware rewards improves multi-robot exploration efficiency. This is an empirical RL study: the four reward functions are hand-specified in Section 3.5, the policies are trained in Gymnasium (Section 4), and the resulting behaviors are evaluated on separate test environments (Table 3 and Table 5). No parameter is fitted to the evaluation data and then reported as an independent prediction; the observation that communication-rewarded policies produce higher Dshared is a manipulation check, not a circular derivation. HAPPO is taken from an external published algorithm [27], and there are no load-bearing self-citations. The main weaknesses are evidential, not circular: the Gazebo results in Table 5 do not consistently support the text's claim that communication-trained policies outperformed Study Case 1 in all setups (e.g., Setup 1 map coverage is 91.4% for SC4 but 37.7% for SC2 and 53.2% for SC3), and the novelty claim that this is 'the first time such an approach has been proposed' is unverified. Those are correctness or evidence concerns and do not amount to the paper's derivation reducing to its own inputs.

Assumptions & free parameters 8 free parameters · 5 assumptions · 0 invented entities

The central empirical claim rests on hand-tuned reward weights and an idealized communication channel. No new physical entities are introduced. The key free parameters are the reward coefficients and the communication range, none of which are varied in sensitivity analyses.

free parameters (8)
  • r*_rep (stay-still penalty) = -1
    Hand-chosen penalty for staying still, used in all reward functions; no sensitivity analysis.
  • r*_bou (boundary penalty) = -1
    Hand-chosen penalty for approaching map boundaries; used in Study Cases 1-3.
  • r*_col (collision penalty) = -10
    Hand-chosen penalty for collisions or moving out of bounds; dominates the reward.
  • r*_ndi (low-exploration penalty) = 0.2
    Penalty applied when exploration reward is below r*_exp=0.3; hand-chosen.
  • r*_exp (exploration reward threshold) = 0.3
    Threshold for triggering the low-exploration penalty; hand-chosen.
  • p_k offset in Study Case 3 = +0.6
    Constant added in Eq (10) to weight communication reward; hand-chosen.
  • p_k offset in Study Case 4 = +0.8
    Constant added in Eq (12); hand-chosen.
  • Communication range r_c = 5 (grid cells)
    Chosen in Table 2 for the Gymnasium environment; affects when agents can communicate and thus the exploration behavior.
assumptions (5)
  • domain assumption The environment transition function T is deterministic (Section 2.1).
    The paper states T is deterministic; in reality, robot motion and sensing have noise, especially in Gazebo, where the model is not exact.
  • domain assumption Agents within communication range can merge maps perfectly with no errors or delays (Section 2.4).
    Map merging is assumed lossless and instantaneous; real communication has bandwidth limits, latency, and potential packet loss, which the paper motivates but does not model.
  • domain assumption Exploration is considered successful at 90% coverage of free cells (Section 5.1).
    The termination criterion is arbitrary; results depend on this threshold.
  • standard math HAPPO (reference [27]) provides a valid joint-policy optimization for this homogeneous-agent setting.
    The paper relies on the published HAPPO algorithm without verifying its convergence guarantees for this specific exploration task.
  • domain assumption The occupancy grid and action abstraction (one cell per action) captures the robot's continuous motion sufficiently for policy transfer to Gazebo.
    The policy is trained on discrete grid steps and then applied to continuous TurtleBot navigation; the paper itself shows transfer issues in some setups.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Investigating the Impact of Communication-Induced Action Space on Exploration of Unknown Environments with Decentralized Multi-Agent Reinforcement Learning." pith.science (2026). https://pith.science/paper/TA2VHL6P

@misc{pith2026241220075,
  author       = {Pith},
  title        = {Pith review of: Investigating the Impact of Communication-Induced Action Space on Exploration of Unknown Environments with Decentralized Multi-Agent Reinforcement Learning},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/TA2VHL6P}},
  note         = {Machine review of arXiv:2412.20075}
}
read the original abstract

This paper introduces a novel enhancement to the Decentralized Multi-Agent Reinforcement Learning (D-MARL) exploration by proposing communication-induced action space to improve the mapping efficiency of unknown environments using homogeneous agents. Efficient exploration of large environments relies heavily on inter-agent communication as real-world scenarios are often constrained by data transmission limits, such as signal latency and bandwidth. Our proposed method optimizes each agent's policy using the heterogeneous-agent proximal policy optimization algorithm, allowing agents to autonomously decide whether to communicate or to explore, that is whether to share the locally collected maps or continue the exploration. We propose and compare multiple novel reward functions that integrate inter-agent communication and exploration, enhance mapping efficiency and robustness, and minimize exploration overlap. This article presents a framework developed in ROS2 to evaluate and validate the investigated architecture. Specifically, four TurtleBot3 Burgers have been deployed in a Gazebo-designed environment filled with obstacles to evaluate the efficacy of the trained policies in mapping the exploration arena.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

34 extracted references · 31 canonical work pages

  1. [1]

    In: 2020 6th International Conference on Control, Automation and Robotics (ICCAR), pp

    Huang, Y., Wu, S., Mu, Z., Long, X., Chu, S., Zhao, G.: A multi-agent rein- forcement learning method for swarm robots in space collaborative exploration. In: 2020 6th International Conference on Control, Automation and Robotics (ICCAR), pp. 139–144 (2020)

    work page 2020

  2. [2]

    IEEE Robotics and Automa- tion Letters 6(2), 2091–2098 (2021)

    Clark, L., Galante, J., Krishnamachari, B., Psounis, K.: A queue-stabilizing framework for networked multi-robot exploration. IEEE Robotics and Automa- tion Letters 6(2), 2091–2098 (2021)

    work page 2021

  3. [3]

    2023 China Automation Congress (CAC), 7639–7644 (2023)

    Zhu, L., Cheng, J., Liu, Y.: Multi-robot autonomous exploration in unknown environment: A review. 2023 China Automation Congress (CAC), 7639–7644 (2023)

    work page 2023

  4. [4]

    iHERO: Interactive Human-oriented Exploration and Supervision Under Scarce Communication

    Tian, Z., Zhang, Y., Wei, J., Guo, M.: iHERO: Interactive Human-oriented Explo- ration and Supervision Under Scarce Communication (2024). https://arxiv.org/ abs/2405.12571

    work page Pith review arXiv 2024

  5. [5]

    IEEE Robotics and Automation Letters 4, 610–617 (2019) 25

    Niroui, F., Zhang, K., Kashino, Z., Nejat, G.: Deep reinforcement learning robot for search and rescue applications: Exploration in unknown cluttered environments. IEEE Robotics and Automation Letters 4, 610–617 (2019) 25

    work page 2019

  6. [6]

    In: 2022 China Automation Congress (CAC), pp

    Deng, L., Gong, W., Li, L.: Multi-robot exploration in unknown environments via multi-agent deep reinforcement learning. In: 2022 China Automation Congress (CAC), pp. 6898–6902 (2022). IEEE

    work page 2022

  7. [7]

    IEEE Transactions on Vehicular Technology 69(12), 14413–14423 (2020)

    Hu, J., Niu, H., Carrasco, J., Lennox, B., Arvin, F.: Voronoi-based multi- robot autonomous exploration in unknown environments via deep reinforcement learning. IEEE Transactions on Vehicular Technology 69(12), 14413–14423 (2020)

    work page 2020

  8. [8]

    Current Robotics Reports 1(4), 227–237 (2020)

    Quattrini Li, A.: Exploration and mapping with groups of robots: Recent trends. Current Robotics Reports 1(4), 227–237 (2020)

    work page 2020

Show all 34 references
  1. [9]

    arXiv preprint arXiv:2307.08690 (2023)

    Adeleye, A.: Robotic exploration for mapping. arXiv preprint arXiv:2307.08690 (2023)

    2023 arXiv

  2. [10]

    In: Proceedings of the Second International Conference on Autonomous Agents, pp

    Yamauchi, B.: Frontier-based exploration using multiple robots. In: Proceedings of the Second International Conference on Autonomous Agents, pp. 47–53 (1998)

    1998

  3. [11]

    In: Pro- ceedings of the Seventeenth National Conference on Artificial Intelligence and Twelfth Conference on Innovative Applications of Artificial Intelligence, pp

    Simmons, R.G., Apfelbaum, D., Burgard, W., Fox, D., Moors, M., Thrun, S., Younes, H.L.S.: Coordination for multi-robot exploration and mapping. In: Pro- ceedings of the Seventeenth National Conference on Artificial Intelligence and Twelfth Conference on Innovative Applications...

    2000

  4. [12]

    In: Intelligent Robotics and Applications: 5th International Conference, ICIRA 2012, Montreal, Canada, October 3-5, 2012, Proceedings, Part II 5, pp

    Bautin, A., Simonin, O., Charpillet, F.: Minpos: A novel frontier allocation algo- rithm for multi-robot exploration. In: Intelligent Robotics and Applications: 5th International Conference, ICIRA 2012, Montreal, Canada, October 3-5, 2012, Proceedings, Part II 5, pp. 496–508 (...

    2012

  5. [13]

    In: ISR 2010 (41st International Symposium on Robotics) and ROBOTIK 2010 (6th German Conference on Robotics), pp

    Holz, D., Basilico, N., Amigoni, F., Behnke, S.: Evaluating the efficiency of frontier-based exploration strategies. In: ISR 2010 (41st International Symposium on Robotics) and ROBOTIK 2010 (6th German Conference on Robotics), pp. 1–8 (2010). VDE

    2010

  6. [14]

    IF AC-PapersOnLine 53(2), 9682–9687 (2020)

    Batinovi´ c, A., Orˇ suli´ c, J., Petrovi´ c, T., Bogdan, S.: Decentralized strategy for cooperative multi-robot exploration and mapping. IF AC-PapersOnLine 53(2), 9682–9687 (2020)

    2020

  7. [15]

    The International Journal of Robotics Research 33(2), 215–236 (2014)

    Keidar, M., Kaminka, G.A.: Efficient frontier detection for robot exploration. The International Journal of Robotics Research 33(2), 215–236 (2014)

    2014

  8. [16]

    In: 2018 13th Annual Conference on System of Systems Engineering (SoSE), pp

    Mahdoui, N., Fr´ emont, V., Natalizio, E.: Cooperative frontier-based exploration strategy for multi-robot system. In: 2018 13th Annual Conference on System of Systems Engineering (SoSE), pp. 203–210 (2018)

    2018

  9. [17]

    In: 2017 IEEE/RSJ International 26 Conference on Intelligent Robots and Systems (IROS), pp

    Bai, S., Chen, F., Englot, B.: Toward autonomous mapping and exploration for mobile robots through deep supervised learning. In: 2017 IEEE/RSJ International 26 Conference on Intelligent Robots and Systems (IROS), pp. 2379–2384 (2017)

    2017

  10. [18]

    In: Proceedings 2000 ICRA

    Burgard, W., Moors, M., Fox, D., Simmons, R., Thrun, S.: Collaborative multi- robot exploration. In: Proceedings 2000 ICRA. Millennium Conference. IEEE International Conference on Robotics and Automation. Symposia Proceedings, vol. 1, pp. 476–481 (2000). IEEE

    2000

  11. [19]

    https://arxiv.org/abs/1905.12127

    Iqbal, S., Sha, F.: Coordinated Exploration via Intrinsic Rewards for Multi-Agent Reinforcement Learning (2021). https://arxiv.org/abs/1905.12127

    2021 arXiv

  12. [20]

    IEEE Robotics and Automation Letters 7(2), 3435–3442 (2022)

    Zhang, H., Cheng, J., Zhang, L., Li, Y., Zhang, W.: H2GNN: Hierarchical-hops graph neural networks for multi-robot exploration in unknown environments. IEEE Robotics and Automation Letters 7(2), 3435–3442 (2022)

    2022

  13. [21]

    Entropy 21(3), 294 (2019)

    Geng, M., Xu, K., Zhou, X., Ding, B., Wang, H., Zhang, L.: Learning to coop- erate via an attention-based communication neural network in decentralized multi-robot exploration. Entropy 21(3), 294 (2019)

    2019

  14. [22]

    https://arxiv.org/abs/2011

    Li, Q., Lin, W., Liu, Z., Prorok, A.: Message-Aware Graph Attention Networks for Large-Scale Multi-Robot Path Planning (2021). https://arxiv.org/abs/2011. 13219

    2021

  15. [23]

    In: 2020 IEEE International Conference on Robotics and Automation (ICRA), pp

    Han, R., Chen, S., Hao, Q.: Cooperative multi-robot navigation in dynamic environment with deep reinforcement learning. In: 2020 IEEE International Conference on Robotics and Automation (ICRA), pp. 448–454 (2020)

    2020

  16. [24]

    IEEE Intelligent Systems 32(6), 48–57 (2017)

    Amigoni, F., Banfi, J., Basilico, N.: Multirobot exploration of communication- restricted environments: A survey. IEEE Intelligent Systems 32(6), 48–57 (2017)

    2017

  17. [25]

    Springer Proceedings in Advanced Robotics, pp

    Jensen, E., Gini, M.: Effects of Communication Restriction on Online Multi- robot Exploration in Bounded Environments. Springer Proceedings in Advanced Robotics, pp. 469–483 (2019)

    2019

  18. [26]

    https://arxiv.org/abs/ 2011.01119

    Tolstaya, E., Paulos, J., Kumar, V., Ribeiro, A.: Multi-Robot Coverage and Exploration using Spatial Graph Neural Networks (2021). https://arxiv.org/abs/ 2011.01119

    2021 arXiv

  19. [27]

    Journal of Machine Learning Research25(1-67), 1 (2024)

    Zhong, Y., Kuba, J.G., Feng, X., Hu, S., Ji, J., Yang, Y.: Heterogeneous-agent reinforcement learning. Journal of Machine Learning Research25(1-67), 1 (2024)

    2024

  20. [28]

    : Pettingzoo: Gym for multi-agent reinforcement learning

    Terry, J., Black, B., Grammel, N., Jayakumar, M., Hari, A., Sullivan, R., Santos, L.S., Dieffendahl, C., Horsch, C., Perez-Vicente, R., et al. : Pettingzoo: Gym for multi-agent reinforcement learning. Advances in Neural Information Processing Systems 34, 15032–15043 (2021)

    2021

  21. [29]

    https://arxiv.org/ abs/2202.11931

    Xu, Y., Yu, J., Tang, J., Qiu, J., Wang, J., Shen, Y., Wang, Y., Yang, H.: 27 Explore-Bench: Data Sets, Metrics and Evaluations for Frontier-based and Deep- reinforcement-learning-based Autonomous Exploration (2022). https://arxiv.org/ abs/2202.11931

    2022 arXiv

  22. [30]

    arXiv preprint arXiv:2110.09619 (2021)

    Costa, L.d.F.: Further generalizations of the jaccard index. arXiv preprint arXiv:2110.09619 (2021)

    2021 arXiv

  23. [31]

    Science Robotics 7(66), 6074 (2022)

    Macenski, S., Foote, T., Gerkey, B., Lalancette, C., Woodall, W.: Robot operating system 2: Design, architecture, and uses in the wild. Science Robotics 7(66), 6074 (2022)

    2022

  24. [32]

    In: 2004 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) (IEEE Cat

    Koenig, N., Howard, A.: Design and use paradigms for gazebo, an open-source multi-robot simulator. In: 2004 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) (IEEE Cat. No.04CH37566), vol. 3, pp. 2149–21543 (2004)

    2004

  25. [33]

    https://arxiv.org/abs/1606.01540

    Brockman, G., Cheung, V., Pettersson, L., Schneider, J., Schulman, J., Tang, J., Zaremba, W.: OpenAI Gym (2016). https://arxiv.org/abs/1606.01540

    2016 arXiv

  26. [34]

    Journal of Open Source Software 6(61), 2783 (2021) 28

    Macenski, S., Jambrecic, I.: Slam toolbox: Slam for the dynamic world. Journal of Open Source Software 6(61), 2783 (2021) 28

    2021

Pith tools

Reviewed August 10, 2026 · model on record in the stance chip above.