RPA controller framework based on the"large model subtask"structure
The RPA controller is an important module that can demonstrate the high-level abilities of RPA.With the development of large models and artificial intelligence technology,intelligence is expected to become one of the important dimensions to measure the performance of RPA controllers.This paper proposes an RPA controller framework based on the"large model-subtask"structure,which references algorithms such as HuggingGPT.This framework integrates the subtask library into the large model and supports intelligent task decomposition,task planning and scheduling with autonomous decision-making capabilities.Additionally,this framework supports an intelligent display of assignment result.
NETL
NSTL
万方数据
