@inproceedings{MarcondosSantos2021a,
  abstract = {Agents can individually devise plans and coordinate to achieve common goals.
Methods exist to factor planning problems into separate tasks and distribute the plan synthesis process, while reducing the overall planning complexity.
Merging distributedly generated plans becomes computationally costly when task plans are tightly coupled, and conflicts arise due to dependencies between plan actions.
New plan merging algorithms allow factoring and solving large problems with a growing number of agents and tasks, but are yet to be demonstrated in physical real-world systems. 
This Demo presents an architecture that deploys plan merging algorithms in a physical multi-robot setting and emulates a First Response Domain.},
  address = {London, UK},
  author = {Marcon dos Santos, Gilberto and Adams, Julie A.},
  booktitle = {International Conference on Autonomous Agents and Multiagent Systems},
  keywords = {mine},
  month = may,
  pages = {1776--1778},
  title = {Scalable Multiple Robot Task Planning with Plan Merging and Conflict Resolution},
  url = {https://www.ifaamas.org/Proceedings/aamas2021/pdfs/p1776.pdf},
  year = 2021
}

Agents can individually devise plans and coordinate to achieve common goals. Methods exist to factor planning problems into separate tasks and distribute the plan synthesis process, while reducing the overall planning complexity. Merging distributedly generated plans becomes computationally costly when task plans are tightly coupled, and conflicts arise due to dependencies between plan actions. New plan merging algorithms allow factoring and solving large problems with a growing number of agents and tasks, but are yet to be demonstrated in physical real-world systems. This Demo presents an architecture that deploys plan merging algorithms in a physical multi-robot setting and emulates a First Response Domain.

@article{McCammon2020,
  added-at = {2020-05-24T20:36:48.000+0200},
  author = {McCammon, Seth and dos Santos, Gilberto Marcon and Frantz, Matt and Welch, T. P. and Best, Graeme and Shearman, R. Kipp and Nash, Jonathan and Barth, J. A. and Adams, Julie A. and Hollinger, Geoffrey A.},
  biburl = {https://www.bibsonomy.org/bibtex/257d0e2210520afcadb953c90ea0f5da6/marcondg},
  interhash = {81288e336713ced983882871663ab8d1},
  intrahash = {57d0e2210520afcadb953c90ea0f5da6},
  journal = {Journal of Field Robotics},
  keywords = {mine myown},
  timestamp = {2020-05-24T20:36:48.000+0200},
  title = {Ocean Front Detection and Tracking using a Team of Heterogeneous Marine Vehicles},
  pages = {854--881},
  volume = 38,
  number = 6,
  year = 2021
}

@phdthesis{MarcondosSantos2020e,
  abstract = {This dissertation incorporates coalition formation and probabilistic planning towards a domain-independent automated planning solution scalable to multiple heterogeneous robots in complex domains. The first research direction investigates the effectiveness of Task Fusion and introduces heuristics that improve task allocation and result in better quality plans, while requiring lower computational cost than the baseline approaches. The heuristics incorporate relaxed plans to estimate coupling and determine which tasks to fuse. As a result, larger temporal continuous planning problems involving multiple robots can be solved. The second research direction introduces new coordination methods to merge plans and resolve conflicts while extending the framework to domains with stochastic action duration. Merging distributedly generated plans becomes computationally costly when task plans are tightly coupled, and conflicts arise due to dependencies between plan actions. Existing methods either scale poorly as the number of agents and tasks increases, or do not minimize makespan, the overall time necessary to execute all tasks. A new family of plan coordination and conflict resolution algorithms is introduced to merge independently generated plans, minimize the resulting makespan, and scale to a large number of tasks and agents in complex problems. A thorough algorithmic analysis and empirical evaluation demonstrates how the new conflict identification and resolution models can impact the resulting plan quality and computational cost across three heterogeneous multiagent domains and outperform the baseline algorithms.},
  added-at = {2020-06-15T20:19:59.000+0200},
  author = {Marcon dos Santos, Gilberto},
  biburl = {https://www.bibsonomy.org/bibtex/2d39781ac507bd01f8fe6929410b2dc39/marcondg},
  interhash = {f489c500a996062202c9dedec50bea46},
  intrahash = {d39781ac507bd01f8fe6929410b2dc39},
  keywords = {mine},
  month = {June},
  school = {Oregon State University},
  timestamp = {2020-06-15T20:22:12.000+0200},
  title = {Coordination for Scalable Multiple Robot Planning Under Temporal Uncertainty},
  url = {https://ir.library.oregonstate.edu/downloads/gb19fd28x},
  year = 2020
}

This dissertation incorporates coalition formation and probabilistic planning towards a domain-independent automated planning solution scalable to multiple heterogeneous robots in complex domains. The first research direction investigates the effectiveness of Task Fusion and introduces heuristics that improve task allocation and result in better quality plans, while requiring lower computational cost than the baseline approaches. The heuristics incorporate relaxed plans to estimate coupling and determine which tasks to fuse. As a result, larger temporal continuous planning problems involving multiple robots can be solved. The second research direction introduces new coordination methods to merge plans and resolve conflicts while extending the framework to domains with stochastic action duration. Merging distributedly generated plans becomes computationally costly when task plans are tightly coupled, and conflicts arise due to dependencies between plan actions. Existing methods either scale poorly as the number of agents and tasks increases, or do not minimize makespan, the overall time necessary to execute all tasks. A new family of plan coordination and conflict resolution algorithms is introduced to merge independently generated plans, minimize the resulting makespan, and scale to a large number of tasks and agents in complex problems. A thorough algorithmic analysis and empirical evaluation demonstrates how the new conflict identification and resolution models can impact the resulting plan quality and computational cost across three heterogeneous multiagent domains and outperform the baseline algorithms.

@inproceedings{MarcondosSantos2020b,
  abstract = {Agents can individually devise plans and coordinate to achieve common goals. Methods exist to factor planning problems into separate tasks and distribute the plan synthesis process, while reducing the overall planning complexity. However, merging distributedly generated plans becomes computationally costly when task plans are tightly coupled, and conflicts arise due to dependencies between plan actions. Existing methods either scale poorly as the number of agents and tasks increases, or do not minimize makespan, the overall time necessary to execute all tasks. A new algorithm, the Temporal Optimal Conflict Resolution Algorithm (TCRA*), is introduced to merge independently generated plans and optimally minimize the resulting makespan. A proof of optimality is provided and the algorithm is empirically evaluated across two heterogeneous multiagent domains against two baseline algorithms. The TCRA* results in better makespan across the problems solved, and a search relaxation constant allows the TCRA* to generate better plans with competitive processing time and memory usage.},
  added-at = {2020-05-24T20:36:48.000+0200},
  address = {Auckland, New Zealand},
  author = {Marcon dos Santos, Gilberto and Adams, Julie A.},
  biburl = {https://www.bibsonomy.org/bibtex/29ddc20acd02b6688719499428fa6e2c0/marcondg},
  booktitle = {International Conference on Autonomous Agents and Multiagent Systems},
  interhash = {9a641ba2aed56777e7b19878eb758f34},
  intrahash = {9ddc20acd02b6688719499428fa6e2c0},
  keywords = {mine},
  month = may,
  pages = {851--859},
  timestamp = {2020-05-24T20:36:48.000+0200},
  title = {Optimal Temporal Plan Merging},
  url = {http://ifaamas.org/Proceedings/aamas2020/pdfs/p851.pdf},
  year = 2020
}

Agents can individually devise plans and coordinate to achieve common goals. Methods exist to factor planning problems into separate tasks and distribute the plan synthesis process, while reducing the overall planning complexity. However, merging distributedly generated plans becomes computationally costly when task plans are tightly coupled, and conflicts arise due to dependencies between plan actions. Existing methods either scale poorly as the number of agents and tasks increases, or do not minimize makespan, the overall time necessary to execute all tasks. A new algorithm, the Temporal Optimal Conflict Resolution Algorithm (TCRA*), is introduced to merge independently generated plans and optimally minimize the resulting makespan. A proof of optimality is provided and the algorithm is empirically evaluated across two heterogeneous multiagent domains against two baseline algorithms. The TCRA* results in better makespan across the problems solved, and a search relaxation constant allows the TCRA* to generate better plans with competitive processing time and memory usage.

@article{MarcondosSantos2020a,
  abstract = {Automating planning for large teams of heterogeneous robots is a growing challenge, as robot capabilities diversify and domain complexities are incorporated. Temporal and continuous features accurately model real-world constraints, but add computational complexity. Distributed planning methods, such as the Coalition Formation then Planning framework, allocate tasks to robot teams and plan each task separately to accelerate planning. However, the task decomposition limits cooperation between coalitions allocated to different tasks and results in lower quality plans that require more actions and time to complete. Task Fusion estimates couplings between tasks and fuses coupled coalition-task pairs to improve cooperation and produce higher quality plans. Task Fusion relies on existing heuristics, which were ineffective and often resulted in worse results than the baseline framework. This manuscript introduces new heuristics that outperform the existing methods in two complex heterogeneous multi-robot domains that incorporate temporal and continuous constraints.},
  added-at = {2020-05-24T20:36:48.000+0200},
  author = {Marcon dos Santos, Gilberto and Adams, Julie A.},
  biburl = {https://www.bibsonomy.org/bibtex/2c04ec05c37b956ae1e0a1317f5f4c016/marcondg},
  interhash = {beaa161d6afdb907aaf6604f0ea66136},
  intrahash = {c04ec05c37b956ae1e0a1317f5f4c016},
  journal = {Multiagent and Grid Systems},
  keywords = {mine myown},
  number = 2,
  pages = {171--192},
  timestamp = {2020-09-11T07:03:56.000+0200},
  title = {{Plan Distance Heuristics for Task Fusion in Distributed Temporal Continuous Planning}},
  url = {http://people.oregonstate.edu/~marcondg/publications/mgs.pdf},
  volume = 16,
  year = 2020
}

Automating planning for large teams of heterogeneous robots is a growing challenge, as robot capabilities diversify and domain complexities are incorporated. Temporal and continuous features accurately model real-world constraints, but add computational complexity. Distributed planning methods, such as the Coalition Formation then Planning framework, allocate tasks to robot teams and plan each task separately to accelerate planning. However, the task decomposition limits cooperation between coalitions allocated to different tasks and results in lower quality plans that require more actions and time to complete. Task Fusion estimates couplings between tasks and fuses coupled coalition-task pairs to improve cooperation and produce higher quality plans. Task Fusion relies on existing heuristics, which were ineffective and often resulted in worse results than the baseline framework. This manuscript introduces new heuristics that outperform the existing methods in two complex heterogeneous multi-robot domains that incorporate temporal and continuous constraints.

@inproceedings{MarcondosSantos2018,
  added-at = {2020-05-24T20:36:48.000+0200},
  address = {Stockholm, Sweden},
  author = {Marcon dos Santos, Gilberto and Adams, Julie A.},
  biburl = {https://www.bibsonomy.org/bibtex/296f86bc195479e5d5ad7a83762e43874/marcondg},
  booktitle = {International Conference on Autonomous Agents and Multiagent Systems},
  interhash = {2ff5ac32495a40dd34fa8fdaf1d9462a},
  intrahash = {96f86bc195479e5d5ad7a83762e43874},
  keywords = {mine myown},
  month = jul,
  pages = {2198--2200},
  timestamp = {2020-05-24T20:36:48.000+0200},
  title = {{Task Fusion Heuristics for Coalition Formation and Planning}},
  url = {http://ifaamas.org/Proceedings/aamas2018/pdfs/p2198.pdf},
  year = 2018
}

@inproceedings{Tuzel2018,
  added-at = {2020-05-24T20:36:48.000+0200},
  address = {Rome, Italy},
  author = {Tuzel, Ovunc and Marcon dos Santos, Gilberto and Fleming, Chlo\"{e} and Adams, Julie A.},
  biburl = {https://www.bibsonomy.org/bibtex/20c2ba6bc0775813049aad69bc8898007/marcondg},
  booktitle = {International Conference on Swarm Intelligence},
  doi = {10.1007/978-3-030-00533-7\_33},
  interhash = {a7b6d1daef3b951b41467bdba344dd7e},
  intrahash = {0c2ba6bc0775813049aad69bc8898007},
  isbn = {978-3-030-00532-0},
  keywords = {mine myown},
  month = {October},
  pages = {385--394},
  timestamp = {2020-05-24T20:36:48.000+0200},
  title = {{Learning Based Leadership in Swarm Navigation}},
  url = {http://people.oregonstate.edu/~marcondg/publications/ANTS2018.pdf},
  year = 2018
}

@article{MarcondosSantos2016b,
  added-at = {2020-05-24T20:36:48.000+0200},
  author = {Marcon dos Santos, Gilberto and Ferrao, Victor and Noronha Vinhal, Cassio D. and Cruz Junior, Gelson},
  biburl = {https://www.bibsonomy.org/bibtex/281346ce3bc345be8b29a26556e87ba96/marcondg},
  interhash = {0b951bfec1d1c03a78b589e769844104},
  intrahash = {81346ce3bc345be8b29a26556e87ba96},
  journal = {Pattern Recognition Letters},
  keywords = {mine myown},
  pages = {192--198},
  timestamp = {2020-05-24T20:36:48.000+0200},
  title = {{Fast algorithm for real-time ground extraction from unorganized stereo point clouds}},
  url = {http://people.oregonstate.edu/~marcondg/publications/PRL16.pdf},
  year = 2016
}

@inproceedings{MarcondosSantos2016a,
  added-at = {2020-05-24T20:36:48.000+0200},
  address = {Fukuoka, Japan},
  author = {Marcon dos Santos, Gilberto and Ferrao, Victor and Noronha Vinhal, Cassio D. and Cruz Junior, Gelson},
  biburl = {https://www.bibsonomy.org/bibtex/26292f1255d432297071f09b1042de06b/marcondg},
  booktitle = {2015 International Conference of Soft Computing and Pattern Recognition, So{CP}aR 2015},
  interhash = {f9a54b5d6948a9de24ed04b3595e1b2b},
  intrahash = {6292f1255d432297071f09b1042de06b},
  keywords = {mine myown},
  month = {November},
  pages = {76--83},
  timestamp = {2020-09-16T20:01:12.000+0200},
  title = {{An adaptive algorithm for embedded real-time point cloud ground segmentation}},
  url = {http://people.oregonstate.edu/~marcondg/publications/SocPar15.pdf},
  year = 2015
}

@article{MarcondosSantos2015,
  added-at = {2020-05-24T20:36:48.000+0200},
  author = {Marcon dos Santos, Gilberto and Ferrao, Victor and Vinhal, Cassio and da Cruz Junior, Gelson},
  biburl = {https://www.bibsonomy.org/bibtex/2b225dd21c39a94854ec716349e29c26e/marcondg},
  interhash = {74d3735329f460b1e140f9be0915068e},
  intrahash = {b225dd21c39a94854ec716349e29c26e},
  journal = {International Journal of Hybrid Intelligent Systems},
  keywords = {mine myown},
  number = 4,
  pages = {229--243},
  publisher = {IOS Press},
  timestamp = {2020-05-24T20:36:48.000+0200},
  title = {{Performance analysis for a novel adaptive algorithm for real-time point cloud ground segmentation}},
  url = {http://people.oregonstate.edu/~marcondg/publications/IJHIS16.pdf},
  volume = 12,
  year = 2015
}

@inproceedings{MarcondosSantos2014,
  added-at = {2020-05-24T20:36:48.000+0200},
  address = {Philadelphia, Pennsylvania},
  author = {Marcon dos Santos, Gilberto and Barnes, Zachary and Lo, Eric and Ritoper, Bryan and Nishizaki, Lauren and Tejeda, Xavier and Ke, Alex and Lin, Han and Schurgers, Curt and Lin, Albert and Kastner, Ryan},
  biburl = {https://www.bibsonomy.org/bibtex/2240df39776f6748b113de86e91466cbd/marcondg},
  booktitle = {{IEEE} International Conference on Mobile Ad Hoc and Sensor Systems},
  interhash = {1b56a055a53ef4becbc7dac8f53c7df4},
  intrahash = {240df39776f6748b113de86e91466cbd},
  keywords = {mine myown},
  month = {October},
  pages = {761--766},
  timestamp = {2020-05-24T20:36:48.000+0200},
  title = {{Small unmanned aerial vehicle system for wildlife radio collar tracking}},
  url = {http://people.oregonstate.edu/~marcondg/publications/MASS14.pdf},
  year = 2014
}

@article{MarcondosSantos2020d,
  added-at = {2020-05-24T20:36:48.000+0200},
  author = {Marcon dos Santos, Gilberto and Adams, Julie A.},
  biburl = {https://www.bibsonomy.org/bibtex/20295933d80b0b0297dba8e0cf521a1dc/marcondg},
  interhash = {454a6c8f566ffc4ea65a847b0387e20a},
  intrahash = {0295933d80b0b0297dba8e0cf521a1dc},
  keywords = {mine myown},
  timestamp = {2020-05-24T20:36:48.000+0200},
  title = {Decision Support for Multi-Robot Ocean Feature Detection},
  year = {in review}
}

@article{MarcondosSantos2020c,
  abstract = {Agents can individually devise plans and coordinate to achieve common goals. Methods exist to factor planning problems into separate tasks and distribute the plan synthesis process, while reducing the overall planning complexity. However, merging distributedly generated plans becomes computationally costly when task plans are tightly coupled, and conflicts arise due to dependencies between plan actions. Existing methods either scale poorly as the number of agents and tasks increases, or do not minimize makespan, the overall time necessary to execute all tasks. A new family of plan coordination and conflict resolution algorithms is introduced to merge independently generated plans, minimize the resulting makespan, and scale to a large number of tasks and agents in complex problems. A thorough algorithmic analysis and empirical evaluation demonstrates how the new conflict identification and resolution models can impact the resulting plan quality and computational cost across two heterogeneous multiagent domains and outperform the baseline algorithms. },
  added-at = {2020-05-24T20:36:48.000+0200},
  author = {Marcon dos Santos, Gilberto and Adams, Julie A.},
  biburl = {https://www.bibsonomy.org/bibtex/2907f8c7d42a23f90c5345fe0a9f2f39f/marcondg},
  interhash = {773af77bb3fe68fa669772dbf013b5e5},
  intrahash = {907f8c7d42a23f90c5345fe0a9f2f39f},
  keywords = {mine},
  timestamp = {2020-05-24T20:36:48.000+0200},
  title = {Scalable Temporal Plan Merging},
  year = {in review}
}

Agents can individually devise plans and coordinate to achieve common goals. Methods exist to factor planning problems into separate tasks and distribute the plan synthesis process, while reducing the overall planning complexity. However, merging distributedly generated plans becomes computationally costly when task plans are tightly coupled, and conflicts arise due to dependencies between plan actions. Existing methods either scale poorly as the number of agents and tasks increases, or do not minimize makespan, the overall time necessary to execute all tasks. A new family of plan coordination and conflict resolution algorithms is introduced to merge independently generated plans, minimize the resulting makespan, and scale to a large number of tasks and agents in complex problems. A thorough algorithmic analysis and empirical evaluation demonstrates how the new conflict identification and resolution models can impact the resulting plan quality and computational cost across two heterogeneous multiagent domains and outperform the baseline algorithms.