Man-machine heterogeneous data fusion and multi-agent task soft cooperation allocation method and device

The invention provides a man-machine heterogeneous data fusion and multi-agent task soft cooperation allocation method and device. The data fusion method comprises the steps of obtaining multi-source heterogeneous data generated by humans and agents; modeling the data into a factor graph based on the multi-modality of the data and the incidence relation between the data; based on the factor graph, screening the data by applying an attention mechanism or a filtering method; and fusing the effective heterogeneous data obtained by screening on the factor graph by applying a message passing algorithm to obtain enhanced heterogeneous data. The allocation method comprises the steps of establishing a multi-agent cooperation generalization model, taking enhanced heterogeneous data as input, establishing a multi-agent task decision based on a task cooperation factor graph, making a task allocation strategy, and effectively allocating tasks to different agents. According to the method, more effective data input can be