Mechanical arm path planning method and device based on SAC reinforcement learning and medium

The invention discloses a mechanical arm path planning method and device based on SAC reinforcement learning and a medium. The method comprises the following steps that the current state of a mechanical arm is obtained; the current state of the mechanical arm is input into the trained soft actor-commentator SAC reinforcement learning model, the next action of the mechanical arm is obtained, path planning of the mechanical arm is completed, and by means of the method, the equipment and the medium, path planning of the mechanical arm can be more reasonable. 本发明公开了一种基于SAC强化学习的机械臂路径规划方法、设备及介质，包括以下步骤：获取机械臂的当前状态；将所述机械臂的当前状态输入到训练后的软性演员-评论家SAC强化学习模型中，得到机械臂的下一步动作，完成机械臂的路径规划，该方法、设备及介质能够使得机械臂路径规划更加合理。