Distributed reinforcement learning training method and device for super real-time simulation environment

The embodiment of the invention provides a distributed reinforcement learning training method and device for a super real-time simulation environment. The method comprises the following steps: deploying a super real-time simulation environment and an action device on the same virtual machine; controlling the super real-time simulation environment to add an additional information stamp containing the latest feedback time limit information of the action instruction when the environment observation is sent to the action device; controlling the action device to output an action decision accordingto the environment observation and converting the action decision into an action instruction; meanwhile, controlling the action device to judge whether the action instruction is sent to the super real-time simulation environment within the limit of the latest feedback time of the action instruction according to the latest feedback time limit information of the action instruction, and if not, controlling the action device