Bridge construction control method and system based on deep reinforcement learning

The invention discloses a bridge construction control method and system based on deep reinforcement learning. Comprising the steps that an intelligent agent based on a deep neural network is constructed and trained, and the intelligent agent is used for determining a construction instruction of a next working condition; obtaining a safety target characteristic value, a quality target characteristic value and an environment characteristic value in the construction process; and inputting the safety target characteristic value, the quality target characteristic value and the environment characteristic value into the trained intelligent agent to obtain a current optimal construction instruction of the next working condition. The method is an intelligent decision algorithm with robustness, can be suitable for various bridge scenes such as cable-stayed bridges, suspension bridges and arch bridges, and has universality; according to the method, the potential problems of construction safety, quality, progress and the