Home > Research > Publications & Outputs > Real-time Learning and Planning in Environments...

Electronic data

  • aamas20

    Rights statement: © ACM, 2020. This is the author's version of the work. It is posted here by permission of ACM for your personal use. Not for redistribution. The definitive version was published in AAMAS '20: Proceedings of the 19th International Conference on Autonomous Agents and MultiAgent Systems, 2020 https://dl.acm.org/doi/abs/10.5555/3398761.3398880

    Accepted author manuscript, 1.01 MB, PDF document

    Available under license: CC BY-NC: Creative Commons Attribution-NonCommercial 4.0 International License

Links

View graph of relations

Real-time Learning and Planning in Environments with Swarms: A Hierarchical and a Parameter-based Simulation Approach

Research output: Contribution in Book/Report/Proceedings - With ISBN/ISSNConference contribution/Paperpeer-review

Published

Standard

Real-time Learning and Planning in Environments with Swarms : A Hierarchical and a Parameter-based Simulation Approach. / Pelcner, Lukasz; Li, Shaling; Do Carmo Alves, Matheus; Soriano Marcolino, Leandro; Collins, Alex.

Proceedings of the 19th International Conference on Autonomous Agents and Multiagent Systems (AAMAS 2020). ACM, 2020. p. 1019–1027.

Research output: Contribution in Book/Report/Proceedings - With ISBN/ISSNConference contribution/Paperpeer-review

Harvard

Pelcner, L, Li, S, Do Carmo Alves, M, Soriano Marcolino, L & Collins, A 2020, Real-time Learning and Planning in Environments with Swarms: A Hierarchical and a Parameter-based Simulation Approach. in Proceedings of the 19th International Conference on Autonomous Agents and Multiagent Systems (AAMAS 2020). ACM, pp. 1019–1027. <https://dl.acm.org/doi/abs/10.5555/3398761.3398880>

APA

Pelcner, L., Li, S., Do Carmo Alves, M., Soriano Marcolino, L., & Collins, A. (2020). Real-time Learning and Planning in Environments with Swarms: A Hierarchical and a Parameter-based Simulation Approach. In Proceedings of the 19th International Conference on Autonomous Agents and Multiagent Systems (AAMAS 2020) (pp. 1019–1027). ACM. https://dl.acm.org/doi/abs/10.5555/3398761.3398880

Vancouver

Pelcner L, Li S, Do Carmo Alves M, Soriano Marcolino L, Collins A. Real-time Learning and Planning in Environments with Swarms: A Hierarchical and a Parameter-based Simulation Approach. In Proceedings of the 19th International Conference on Autonomous Agents and Multiagent Systems (AAMAS 2020). ACM. 2020. p. 1019–1027

Author

Pelcner, Lukasz ; Li, Shaling ; Do Carmo Alves, Matheus ; Soriano Marcolino, Leandro ; Collins, Alex. / Real-time Learning and Planning in Environments with Swarms : A Hierarchical and a Parameter-based Simulation Approach. Proceedings of the 19th International Conference on Autonomous Agents and Multiagent Systems (AAMAS 2020). ACM, 2020. pp. 1019–1027

Bibtex

@inproceedings{b6bf8d1e0575453bbb74566a1cc67c74,
title = "Real-time Learning and Planning in Environments with Swarms: A Hierarchical and a Parameter-based Simulation Approach",
abstract = "Swarms can be applied in many relevant domains, such as patrolling or rescue. They usually follow simple local rules, leading to complex emergent behavior. Given their wide applicability, an agent may need to take decisions in an environment containing a swarm that is not under its control, and that may even be an antagonist. Predicting the behavior of each swarm member is a great challenge, and must be done under real time constraints, since they usually move constantly following quick reactive algorithms. We propose the first two solutions for this novel problem, showing integrated on-line learning and planning for decision-making with unknown swarms: (i) we learn an ellipse abstraction of the swarm based on statistical models, and predict its future parameters using time-series; (ii) we learn algorithm parameters followed by each swarm member, in order to directly simulate them. We find in ourexperiments that we are significantly faster to reach an objective than local repulsive forces, at the cost of success rate in some situations. Additionally, we show that this is a challenging problem for reinforcement learning.",
author = "Lukasz Pelcner and Shaling Li and {Do Carmo Alves}, Matheus and {Soriano Marcolino}, Leandro and Alex Collins",
note = "{\textcopyright} ACM, 2020. This is the author's version of the work. It is posted here by permission of ACM for your personal use. Not for redistribution. The definitive version was published in AAMAS '20: Proceedings of the 19th International Conference on Autonomous Agents and MultiAgent Systems, 2020 https://dl.acm.org/doi/abs/10.5555/3398761.3398880",
year = "2020",
month = may,
day = "9",
language = "English",
isbn = "9781450375184",
pages = "1019–1027",
booktitle = "Proceedings of the 19th International Conference on Autonomous Agents and Multiagent Systems (AAMAS 2020)",
publisher = "ACM",

}

RIS

TY - GEN

T1 - Real-time Learning and Planning in Environments with Swarms

T2 - A Hierarchical and a Parameter-based Simulation Approach

AU - Pelcner, Lukasz

AU - Li, Shaling

AU - Do Carmo Alves, Matheus

AU - Soriano Marcolino, Leandro

AU - Collins, Alex

N1 - © ACM, 2020. This is the author's version of the work. It is posted here by permission of ACM for your personal use. Not for redistribution. The definitive version was published in AAMAS '20: Proceedings of the 19th International Conference on Autonomous Agents and MultiAgent Systems, 2020 https://dl.acm.org/doi/abs/10.5555/3398761.3398880

PY - 2020/5/9

Y1 - 2020/5/9

N2 - Swarms can be applied in many relevant domains, such as patrolling or rescue. They usually follow simple local rules, leading to complex emergent behavior. Given their wide applicability, an agent may need to take decisions in an environment containing a swarm that is not under its control, and that may even be an antagonist. Predicting the behavior of each swarm member is a great challenge, and must be done under real time constraints, since they usually move constantly following quick reactive algorithms. We propose the first two solutions for this novel problem, showing integrated on-line learning and planning for decision-making with unknown swarms: (i) we learn an ellipse abstraction of the swarm based on statistical models, and predict its future parameters using time-series; (ii) we learn algorithm parameters followed by each swarm member, in order to directly simulate them. We find in ourexperiments that we are significantly faster to reach an objective than local repulsive forces, at the cost of success rate in some situations. Additionally, we show that this is a challenging problem for reinforcement learning.

AB - Swarms can be applied in many relevant domains, such as patrolling or rescue. They usually follow simple local rules, leading to complex emergent behavior. Given their wide applicability, an agent may need to take decisions in an environment containing a swarm that is not under its control, and that may even be an antagonist. Predicting the behavior of each swarm member is a great challenge, and must be done under real time constraints, since they usually move constantly following quick reactive algorithms. We propose the first two solutions for this novel problem, showing integrated on-line learning and planning for decision-making with unknown swarms: (i) we learn an ellipse abstraction of the swarm based on statistical models, and predict its future parameters using time-series; (ii) we learn algorithm parameters followed by each swarm member, in order to directly simulate them. We find in ourexperiments that we are significantly faster to reach an objective than local repulsive forces, at the cost of success rate in some situations. Additionally, we show that this is a challenging problem for reinforcement learning.

M3 - Conference contribution/Paper

SN - 9781450375184

SP - 1019

EP - 1027

BT - Proceedings of the 19th International Conference on Autonomous Agents and Multiagent Systems (AAMAS 2020)

PB - ACM

ER -