Man-machine cooperation method and device, intelligent terminal and storage medium

Abstract

The invention provides a man-machine cooperation method and device, an intelligent terminal and a storage medium, and particularly relates to the technical field of man-machine collaboration, and the scheme comprises the steps: carrying out the three-dimensional posture parameterization of a target depth image through employing a three-dimensional posture parameterization network in a human body digital twin model, and constructing a human body posture; based on a human body posture, extracting a human body joint point sequence and performing feature extraction to obtain global joint point features; performing risk assessment on the global joint point features by adopting a risk assessment network in a human body digital twin model to obtain an ergonomics risk index; and a behavior intention recognition network in a human body digital twin model is adopted to recognize and classify the global joint point features and the target video stream, and a classification result of behavior intentions is obtained, so that a behavior decision is made and a motion track is planned. According to the scheme, real-time comprehensive understanding and response to the human body state can be achieved, it is ensured that the robot efficiently and safely executes tasks, and therefore smoother man-machine cooperation is achieved. 本发明提供的人机协作方法、装置、智能终端及存储介质，具体涉及人机协作技术领域，方案包括：采用人体数字孪生模型中的三维姿态参数化网络对目标深度图像进行三维姿态参数化，构建人体姿态；基于人体姿态，提取出人体关节点序列并进行特征提取，获得全局关节点特征；采用人体数字孪生模型中的风险评估网络对全局关节点特征进行风险评估，获得人机工程学风险指数；采用人体数字孪生模型中的行为意图识别网络对全局关节点特征和目标视频流进行识别和分类，获得行为意图的分类结果，从而做出行为决策并规划运动轨迹。该方案能够实现对人体状态的实时全面理解和响应，确保机器人高效且安全地执行任务，从而实现更为顺畅的人机协作。