双足机器人运动技能突破！谷歌DeepMind创新深度强化学习框架，赋能仿人机器人全身控制，征战足球赛场。机器人展现惊人动态技能，跌倒自恢复，战术防守样样精通。

一项佐治亚理工学院的研究团队发表在国际顶级期刊《Science Robotics》的研究，试图利用深度域适应框架解决这个痛点。据悉，他们提出的方法能够将外骨骼专用标注数据的需求量减少95%，同时保持接近最优模型的控制性能。

月球，这颗距离地球最近的天体邻居，不仅承载着人类对宇宙的无限遐想，更藏着解读宇宙起源的科学密码，其丰富的自然资源和独特的空间环境，使其成为深空探测的关键跳板。在月球众多神秘特征中，由地下空洞坍塌形成的坑洞和洞穴，被认为是未来人类 ...

QUT robotics researchers have developed a new robot navigation system that mimics neural processes of the human brain and uses less than 10 per cent of the energy required by traditional systems. In a ...

Swiss scientists have published research showing AI-informed robots can learn how to self correct and teach other robots how to behave. It raises questions of consciousness in artificial intelligence.

Purdue University Colleges of Science, Engineering, and Agriculture together with the Purdue Office of Industry Partnerships and Purdue Polytechnic Institute proudly present Robotics Day, a ...