Multi-Agent based task allocation method for control systems in malicious environment

To guarantee the execution of control system, the task allocation in clustered control systems with malicious nodes was researched. The multi-Agent method was used to model and analyze the clustered control systems, where the function modules within nodes were modeled as agents and the clusters were modeled as multi-agent coalition sub-systems. Based on the multi-Agent modeling, the new concept of cluster reputation was presented to measure the reliability and ability of cluster's implementation. to execute tasks. A safe task allocation method of multi-Agent clustered control system was presented based on cluster reputation and control unit reputation, which could improve the security and reliability for tasks execution.