Graph Convolutional Network-Based Multi-Agent Collaborative Task Allocation Optimization Algorithm

Hao Yang

doi:10.54097/dtmc8r41

Authors

Hao Yang

DOI:

https://doi.org/10.54097/dtmc8r41

Keywords:

Graph Convolutional Network, Multi-agent, Task Allocation, Cooperative Optimization, Utility Function

Abstract

Due to the complex interdependencies between tasks, the dynamic state changes of the tasks, and the lack of flexibility of the traditional static matching method in the collaborative task allocation with multi-agent, this study aims to explore an optimization algorithm for collaborative task allocation driven by a graph convolutional network. The paper details the construction of multi-agent task graphs, the design of node features and edge weights as well as the graph convolutional feature aggregation mechanism. It also introduces updating of dynamic collaborative relationship, construction of allocation utility function and collaborative optimization process. The results of comparative and ablation experiments in a two-dimensional simulation platform show that the algorithm has a 97.4% task completion ratio and an average utility of 86.7 with an excellent allocation efficiency and convergence stability.

Downloads

Download data is not yet available.

References

[1] Li, Y., Ren, S., Wu, P., Chen, S., Feng, C., & Zhang, Q. (2021). Learning distilled collaboration graph for multi-agent perception. Advances in Neural Information Processing Systems, 34, 29541-29552.

[2] Chen, S., Dong, J., Ha, P., Li, Y., & Labi, S. (2021). Graph neural network and reinforcement learning for multi-agent cooperative control of connected autonomous vehicles. Computer-Aided Civil and Infrastructure Engineering, 36(7), 838-857. https://doi.org/10.1111/mice.12673

[3] Li, M., Chen, S., Shen, Y., Liu, G., Zhou, X., & Wu, X. (2022). Online multi-agent forecasting with interpretable collaborative graph neural networks. IEEE Transactions on Neural Networks and Learning Systems, 35(4), 4768-4782.

https://doi.org/10.1109/TNNLS.2022.3187316

[4] Fawaz, H., Lesca, J., Quang, P. T. A., & Tran, P. (2022). Graph convolutional reinforcement learning for collaborative queuing agents. IEEE Transactions on Network and Service Management, 20(2), 1363-1377.

https://doi.org/10.1109/TNSM.2022.3225981

[5] Goeckner, A., Sui, Y., Martinet, N., Fang, Z., Zhang, Y., & Duan, W. (2024). Graph neural network-based multi-agent reinforcement learning for resilient distributed coordination of multi-robot systems. In Proceedings of the 2024 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) (pp. 5732-5739). IEEE.

[6] Tang, B., Zhong, Y., Neumann, U., Wang, G., Chen, Y., & Zhang, L. (2021). Collaborative uncertainty in multi-agent trajectory forecasting. Advances in Neural Information Processing Systems, 34, 6328-6340.

[7] Duan, W., Lu, J., & Xuan, J. (2024). Group-aware coordination graph for multi-agent reinforcement learning. arXiv, arXiv:2404.10976. https://doi.org/10.48550/arXiv.2404.10976

[8] Wang, Y., Wu, J., Hua, X., Wang, C., & Zhang, Y. (2023). Air-ground spatial crowdsourcing with UAV carriers via geometric graph convolutional multi-agent deep reinforcement learning. In Proceedings of the 2023 IEEE 39th International Conference on Data Engineering (ICDE) (pp. 1790-1802). IEEE. https://doi.org/10.1109/ICDE55515.2023.00148

[9] Rokhforoz, P., Montazeri, M., & Fink, O. (2023). Multi-agent reinforcement learning with graph convolutional neural networks for optimal bidding strategies of generation units in electricity markets. Expert Systems with Applications, 225, 120010. https://doi.org/10.1016/j.eswa.2023.120010

[10] Bocheng, Z., Huo, M., Li, Z., Wang, X., & Liu, Y. (2025). Graph-based multi-agent reinforcement learning for collaborative search and tracking of multiple UAVs. Chinese Journal of Aeronautics, 38(3), 103214.

https://doi.org/10.1016/j.cja.2024.08.011

[11] Hong, S., Liu, Y., Li, Z., Wang, T., & Chen, Q. (2024). Multi-agent collaborative perception via motion-aware robust communication network. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (pp. 15301-15310). IEEE.

[12] Gao, X., Li, X., Liu, Q., Wang, H., Chen, Y., & Zhang, Y. (2022). Multi-agent decision-making modes in uncertain interactive traffic scenarios via graph convolution-based deep reinforcement learning. Sensors, 22(12), 4586. https://doi.org/10.3390/s22124586

[13] Mo, X., Huang, Z., Xing, Y., Liu, W., & Lv, Y. (2022). Multi-agent trajectory prediction with heterogeneous edge-enhanced graph attention network. IEEE Transactions on Intelligent Transportation Systems, 23(7), 9554-9567.

https://doi.org/10.1109/TITS.2021.3121249

[14] Galliera, R., Venable, K. B., Bassani, M., Piro, G., & Boggia, G. (2024). Collaborative information dissemination with graph-based multi-agent reinforcement learning. In International Conference on Algorithmic Decision Theory (pp. 160-173). Springer Nature Switzerland.

[15] Ahmed, A. N., Mercelis, S., & Anwar, A. (2024). CollabGAT: Collaborative perception using graph attention network. IEEE Access, 12, 142380-142393.

https://doi.org/10.1109/ACCESS.2024.3412803