Robot walking method and device based on neural network

The invention discloses a robot walking method and device based on a neural network. The method comprises the following steps: acquiring target position information; determining at least two pieces of secondary target position information according to the target position information; generating route information from the target position information and the secondary target position information through a neural network model; and according to the route information, executing the robot walking operation. The invention can solve the technical problems that in the prior art, path calculation is conducted only through a single algorithm in the walking process of the robot, the path making efficiency is low, and the walking precision is low. 本发明公开了一种基于神经网络的机器人行走方法及装置。其中，该方法包括：获取目标位置信息；根据目标位置信息，确定至少两个次目标位置信息；将所述目标位置信息以及所述次目标位置信息通过神经网络模型生成路线信息；根据所述路线信息，执行机器人行走操作。本发明解决了现有技术中机器人行走过程中仅通过单一算法进行路径计算，路径制定效率低，行走精度低的技术问题。