China's key sensors for reconnaissance unmanned vehicles need to be imported and localized in 3-5 years

• Categories:Industry information
• Time of issue:2016-10-21 16:21

(Summary description)　　On October 17, the finals of the "Over Dangers 2016" Ground Unmanned System Challenge held by the Chinese Army Equipment Department was held in Beijing. Tang Jin, deputy director of the Scientific Research Ordering Bureau of the Army Armament Department, stated at the final press conference that unmanned technology has broad application prospects in the military field and will profoundly change the shape and style of future wars. In layman's terms, people expect that unmanned systems can complete some high-risk tasks or tasks that humans cannot complete in non-war military operations such as war or rescue and disaster relief.   　　Tang Jin said that this competition represents the highest level of unmanned technology in China in terms of scale and subject difficulty. According to the reporter's understanding, a total of 22 teams stood out from 73 teams from all over the country and advanced to the finals. Their scientific research strength is at the leading level in China. Therefore, industry leaders who come to participate and watch the competition have a considerable say in the development status of unmanned systems and the prospects of their application in future military operations.   　　Modeling the unknown environment is a worldwide problem 　　The guest, Deng Zongquan, vice president of Harbin Institute of Technology, was one of the earliest scholars engaged in lunar rover research. In his view, unmanned systems still have a long way to go before they become tools on the battlefield. 　　Wu Tao, an associate professor at the National University of Defense Technology, interprets the development status of unmanned systems in a popular way as "it can be applied in a known environment, but cannot be modeled in an unknown environment." In this research field, the adaptability of unmanned systems to unstructured environments is a recognized problem, that is, how to improve the adaptability of unmanned systems to unknown environments. 　　In the competition, Professor Xu Youchun from the Military Transportation Academy led the "Mengshi Intelligent No. 1" team to take the lead in the field battlefield reconnaissance category A and B competitions. In fact, this car not only performed well in this competition, but its previous generations of principle prototypes have already won 3 championships and 3 runner-ups in the "China Smart Car Future Challenge". 　　Xu Youchun told a reporter from China Youth Daily and China Youth Online that “every cross-country road that can be driven by humans can basically be walked”, and the “Fierce Lion” also has a “walking” function, that is, it can also be used when GPS positioning and navigation are temporarily lost. "Try to find your way" yourself.But in general, the "Fierce Lion" is still in the cub stage, and can only complete the set tasks in the known environment, and does not have the corresponding adaptive ability to the unknown environment.        For example, if the reconnaissance target is set as an oil barrel, Xu Youchun and his teammates can take hundreds of images of oil barrels with different shapes, positions, and lighting conditions, etc., and save them in the system for comparison and recognition by the "Fierce Lion". By designing a fixed route and procedures for how to overcome known obstacles, the "Fierce Lion" can basically complete the task successfully. But the reality is that people cannot predict all the routes and obstacles in the area to be surveyed. 　　This is not only a problem facing Xu Youchun's team, but the world's unmanned technology research field is facing the dilemma that it cannot model the unknown environment. 　　Yang Jingyu, a professor at Nanjing University of Science and Technology, has been engaged in unmanned systems research for 25 years. He explained that structured environments can be simulated by non-stop modeling. For example, environmental data on highways can be measured, and machines can be collected and identified. And judgment. "Artificial intelligence has something to do with structured environments. But unstructured environments are difficult to simulate, and the world has not done a good job of modeling in unstructured environments." 　　For this reason, interviewed professionals such as Wu Tao, Deng Zongquan, and Yang Jingyu all said that from this perspective, China and the United States, the leader in unmanned technology, are not much different. They all need to strengthen theoretical design research on unstructured environments, that is, unknown environments.    　　My country's R&D strength of basic components needs to be improved 　　Specific to the current status of domestic unmanned system research and development, the main research and development forces who came to participate and watch the competition clearly saw their own advantages and shortcomings. 　　Professionals usually use three dimensions to investigate unmanned systems, including environmental perception, planning and decision-making, and low-level co