Journal of System Simulation
Abstract
Abstract: In recent years, the environment in which agents perform tasks has become more open and dynamic, which puts forward higher requirements for the robustness of task planning and behavior scheduling of agents. As a classic behavior control architecture, behavior tree has the characteristics of modularity, behavior parameterization, and structure of both plan representation and reaction, which can effectively support the behavior representation, decision making and scheduling of agents. Based on the hybrid behavior strategy , this paper proposes a robust behavior tree control architecture for dynamic task environment to realize the prudent decision-making and reactive control of agents. The robust behavior tree control architecture combines the prudent planning algorithm and reactive rules of the agent. A behavior tree code framework of the architecture is accordingly designed, and the planning subtree and the Loopnode are customized to support the fast implementation of the behavior tree to complete the switch between planning and reaction within the tree, which supports users to implement behavior tree quickly and standardly based on the customized structure. By the modeling and simulating of the agent object in a typical attack and defense game case, the paper verifies the rationality of the proposed behavior tree control architecture. Compared with the classical reactive behavior tree strategy, the robust behavior tree can flexibly switch between execution of the plan and reactive strategies, improve the blue agent's penetration win rate to 90% in typical scenarios, shorten task execution time by 24.88%, and improve the effectiveness of agent behavior strategy and modeling efficiency.
Recommended Citation
Wang, Qiwei; Zhang, Qi; Yang, Shuo; and Peng, Yong
(2025)
"Design of Robust Behavior Tree Control Architecture for Agents in Dynamic Task Environment,"
Journal of System Simulation: Vol. 37:
Iss.
1, Article 5.
DOI: 10.16182/j.issn1004731x.joss.24-0584
Available at:
https://dc-china-simulation.researchcommons.org/journal/vol37/iss1/5
First Page
54
Last Page
65
CLC
TP391.9
Recommended Citation
Wang Qiwei, Zhang Qi, Yang Shuo, et al. Design of Robust Behavior Tree Control Architecture for Agents in Dynamic Task Environment[J]. Journal of System Simulation, 2025, 37(1): 54-65.
DOI
10.16182/j.issn1004731x.joss.24-0584
Included in
Artificial Intelligence and Robotics Commons, Computer Engineering Commons, Numerical Analysis and Scientific Computing Commons, Operations Research, Systems Engineering and Industrial Engineering Commons, Systems Science Commons