Modern defence requires intelligent perception algorithms to possess incremental and continuous learning capabilities in complex open scenarios， while traditional deep learning methods are based on closed training with the entire dataset， which limits their application ability and usage scope. Existing continuous learning algorithms face the problem of catastrophic knowledge forgetting. This paper proposes for the first time an incremental continuous target detection method based on the distillation of high-level semantic features， which guides the selection of high-value underlying features through high-level semantic features and distills the feature from the teacher model to the student model， thus effectively facilitating the transfer of knowledge of the old task and alleviating catastrophic knowledge forgetting. Experiments on the public image dataset MS COCO show that this method outperforms the previous best method for target detection in all types of continuous learning scenarios， which is expected to promote the generation of continuous learning capability and autonomous attempts of intelligent systems in open-world setting.

Multi-agent cooperation and competition has the characteristics of real-time and action continuity， incomplete information， huge search space， multi-complex tasks and time-space inference， etc. It is one of the most challenging problems in the current artificial intelligence field. Aiming at the problem of long training time for large-scale multi-agent reinforcement learning， this paper proposes an Actor-Critic-based cooperative confrontation framework， which uses meta curriculum reinforcement learning method to extract meta-models of basic tasks for small-scale scenarios. We carry out model migration to large-scale scenarios based on the curriculum learning， which continues training based on the meta-models and finally obtains a better collaboration strategy. This paper conducts simulation experiments on the "Star-Craft II" platform. The results show that the multi-agent cooperative confrontation technology based on the meta curriculum reinforcement learning can effectively accelerate the training process， and can achieve a higher win rate within a shorter time compared with the traditional training methods. The training speed is increased by about 40%. This method can effectively support the efficient generation of multi-agent cooperative confrontation strategies.