In a bamboo forest in Anji, Huzhou, a group of micro intelligent aerial robots shuttle freely, sometimes skirting low shrubs, sometimes crossing flight and changing formation... On May 5, the Dr. Gao Fei team of the school of control science and engineering of Zhejiang University and Professor Xu Chao team of Zhejiang Great Lakes Research Institute published the latest research results in the journal Science · robot: after more than two years of research, The scientific research team has solved a series of core technologies such as intelligent navigation and fast obstacle avoidance methods for single robots and groups in unknown and complex environments, and developed a small, lightweight, autonomous and controllable flight system that can be in groups. The large-scale robot air formation performance is no longer novel. Through the early satellite positioning and trajectory coding, the robot can pose a variety of shapes. However, once they lose the control of the ground computer, these robots will be headless and even hit obstacles and fall down. Flying in groups in unknown complex environment has always been regarded as a major technical bottleneck in the field of robot and artificial intelligence. "A single aerial robot is only the size of a palm, which is lighter than a can of coke." Goofy said that this new robot has achieved a number of key technological breakthroughs, such as sensing surrounding obstacles, locating its position, generating flight path and multi-agent communication in the complex environment of wild forests, using only airborne vision and airborne computing resources. Xu Chao said that the team has also created a "smart brain" for the aerial robot. Although the processor is only the size of a thumb, it can independently calculate and process a large amount of information encountered during flight. Referring to future applications, goofy believes that in search and rescue scenes such as fire, micro cluster robots can better achieve search and rescue objectives and reduce the risk of search and rescue personnel; In topographic exploration, it can also quickly model the areas that cannot be reached by personnel. "The existing sweeping or service robot will be smarter if it is equipped with the 'smart brain' carried by this robot." (Xinhua News Agency)