He is born in the 1970s. The new trend of reform and opening up and the spring dew of science have deeply implanted in his bones the sense of mission and responsibility of "carrying forward the past and serving the country"; Shandong University, Huazhong University of Science and Technology, Southeast University , Georgia Tech University in the United States, and the National University of Singapore... He has been influenced and tempered by famous Chinese and foreign universities, and has always kept his mission in mind and forged ahead on the research path of artificial intelligence and robotics technology.
As one of the most outstanding young experts in the field of robotics in China, he has been deeply involved in the scientific research and teaching of robotics for more than ten years. He has personally participated in and witnessed the development, cultivation, and development of robotics in our country. The whole process from development to widespread application.
He is Li Xinde, a determined explorer and pure scientist in the field of robotics.
He said that scientific research is not only about in-depth cultivation on the basis of predecessors, but also about bravely exploring unknown areas. For the future world of robots, we need to continue to make breakthroughs and bold innovations.
"There is a nursing robot that can be used in elderly care and other nursing service application scenarios."
"There is an intelligent welding robot that can perform automatic welding and greatly reduce the cost of It reduces labor intensity and saves labor force."
"There is an amphibious robot whose design concept is to combine multiple legs and flapping wings to achieve dual use on land and air."
... …
Professor Li Xinde from the School of Automation of Southeast University is usually a quiet and reserved person, even silent, but whenever the topic of artificial intelligence or robots comes up, he seems to have opened up his "chatterbox" and seems to be a different person. Like people, they have all kinds of strange ideas. In his laboratory, various robots and systems mentioned above are being designed intensively.
Not long ago, the "Key Technologies and Applications of Complex Positioning and Navigation Control of Mobile Medical Service Robots" that he participated in as an important member won the first prize of the Science and Technology Progress Award of the Chinese Society of Automation. This project is oriented to the major needs of national development, especially the major anti-epidemic needs. It focuses on the three key technical issues of mobile service robots: complex environment perception, independent planning and positioning, and intelligent navigation control. It uses a combination of industry, academia and research to develop key technologies for mobile service robots. tackle key problems. Research in these fields is precisely Li Xinde’s “speciality”. During the 2020 COVID-19 epidemic prevention and control period, autonomous medical service robots developed using this core technology performed disinfection tasks in hospitals, nursing homes, rehabilitation centers, communities and other places, greatly reducing the workload of medical staff and reducing cross-infection. risks, ensuring the normal life of the people during the epidemic prevention and control period, and achieving extremely high social and economic benefits.
In fact, in the past ten years or so, Li Xinde has undertaken more than 20 national and provincial scientific research projects, authorized 17 national invention patents, 5 software copyrights, and obtained Remarkable results and awards. But compared to the various honors that scientific research work has brought him, Li Xinde values ????the difficulty level of the challenges he encounters more. In recent years, he has devoted himself to overcoming cutting-edge, world-class professional problems, forgetting sleep and eating, working day and night. Even if he encountered bottlenecks, he never gave up.
Li Xinde comes from a small village in Yantai City, Shandong Province. Back then, if a rural child wanted to change his life trajectory, the only way was to study. When he was a sophomore in high school, he enrolled in a liberal arts class. After he entered his senior year, he felt that he had little interest in liberal arts, so he transferred to a science class. This was an extremely risky choice because the learning contents of liberal arts and sciences were very different. In order to successfully pass the "one-plank bridge" of the college entrance examination, Li Xinde used all available time to cram in science courses and followed the teacher's pace for sprint review. In the end, through his own efforts, he was successfully admitted to the chemical equipment and machinery major of Shenyang University of Chemical Technology. It can be seen that Li Xinde is a person who is unwilling to follow the steps and has a very challenging spirit. This quality laid a good foundation for him to continuously challenge himself in academic research and achieve one achievement after another.
During his college years, in order to save living expenses, Li Xinde only ate two meals a day. Even though material conditions were scarce, it did not affect his determination to learn knowledge. Because of his excellent academic performance and performance, he was rated as the school's outstanding graduate when he graduated.
After graduation, in order to relieve his parents' financial pressure, the sensible Li Xinde did not choose to continue his studies. Instead, he worked as a technician in a profitable company in Jinan for three years.
In the eyes of relatives and friends in his hometown, this job is an enviable and stable "iron rice bowl", but Li Xinde is still not satisfied. He has higher requirements for himself and has higher expectations for the future. Great vision. He decided to apply for a master's degree. At that time, his workplace did not support postgraduate entrance examinations for working people. He could only work during the day. After get off work at night, he would ride on his old bicycle and rush all the way to Jinan University for self-study in the evening, without even bothering to eat dinner. At that time, the roads in the suburbs of Jinan were full of potholes and there were no street lights for a section of the road. Once, in order to save review time, Li Xinde rode his bicycle too fast and accidentally fell into a manhole with a loose manhole cover. He suffered multiple injuries on his body and his face. A big piece of skin fell off.
The hardships of life did not stop Li Xinde, but instead tempered his mind, carving a piece of raw jade into a gleaming one. Hard work paid off, and he was eventually admitted to Shandong University for a master's degree in mechanical design and theory. During his several years of further study at Shandong University, with his solid theoretical foundation and unyielding learning attitude, he gained the favor of Professor Song Kongjie, a master in the field of mechanical vibration, and was selected as his disciple. Under Professor Song's teaching and subtle influence, Li Xinde's knowledge and insights have increased a lot, and he has also achieved rich scientific research results in his studies. He was awarded the "President's Scholarship", the highest scholarship of Shandong University.
From the whims of his youth to his interest as an adult, artificial intelligence and robots have always been a dream that Li Xinde has been pursuing. After entering the 21st century, artificial intelligence in China has gone from a stage of confusion and difficulties to breaking through the restrictions and beginning to see the light. At that time, Li Xinde relied on his forward-looking vision to keenly grasp that artificial intelligence would be the ultimate technological field in the next ten or twenty years. Therefore, he made up his mind to apply for the Department of Control Science and Engineering at Huazhong University of Science and Technology, and successfully became a doctoral student of Professor Huang Xinhan. This year, he was 28 years old. In his heart, Li Xinde inherits the integrity and tenacity of Shandong people. He is unwilling to talk too much about the hard work he has put in, because in his opinion, being able to participate in such a huge and rapid development of artificial intelligence is a great opportunity for the people of Shandong. The country's contribution of its meager strength is an opportunity that can only be encountered in life.
His mentor, Professor Huang Xinhan, once served as a member of the intelligent robot theme expert group of the original "863" plan of the country. He has been working on the front line of artificial intelligence research and often participated in domestic and foreign academic exchanges. Under the teacher's words and deeds, Li Xinde got many opportunities to exercise, which led to greater research interest in artificial intelligence and automation, and he also did more in-depth and detailed scientific research work. Of course, difficulties always exist. The first is the knowledge gap caused by cross-majors. The automation and artificial intelligence majors are very different from the mechanical majors, and there is a lot of homework that needs to be made up for. Moreover, Professor Huang Xinhan is well-known as a strict teacher in the industry. If his academic requirements are not met, he will never give any accommodation.
After thinking about it, Li Xinde believed that the only way he could persevere was to work harder. He chose the most "stupid" way, such as insisting on arriving at the laboratory at the earliest every morning and leaving at the last at night, so that he could spend at least 12 hours of scientific research in the laboratory; he spent almost all his spare time in the library reading I searched almost all the information about artificial intelligence, relying on one or two references, one or two professional books, and slowly accumulated my own unique knowledge in this field. Know. He discusses automation and robotics issues with his mentor and fellow students every day, humbly asks for advice, and is always thinking about how his project will progress.
Li Xinde always grasps the forefront of the field of artificial intelligence, and opportunities always favor those who are prepared. By chance, Li Xinde read the recently released Paradoxical Theory (DSmT). He was deeply fascinated by it and tried to study it in depth. From this, he got acquainted with French scientist Jean Dezert, the founder of Specious Theory (DSmT) and an internationally renowned information fusion expert, and Professor Florentine Smarandache from the United States. After long-term exchanges and discussions, Li Xinde formed a profound friendship with them.
Among them, Jean Dezert was invited to visit Southeast University twice, and Florentine Smarandache also awarded Li Xinde an International Scientific Contribution Certificate in the name of his unit for his special contribution to the theory of plausibility.
Li Xinde's stupid method of "building a stronghold and fighting a stupid war" made him unknowingly one of the students with the best research results in the tutor team. It usually takes 5 to 6 years to complete. He graduated from his doctoral studies in just 4 years. When he graduated, he also won the highest scholarship of Huazhong University of Science and Technology - the "Rockwell" scholarship. Due to his outstanding performance in professional research, many well-known domestic universities have offered him an olive branch. Faced with the choice of "squandering flowers that gradually become charming," Li Xinde had his own opinion, and he finally chose Southeast University, a century-old institution.
"Southeast University is a century-old university, known as the 'Holy Land of Learning.' The construction foundation is very solid and the strength is very strong. I think this platform is very suitable for me!” Facts have proved that Li Xinde once again made the right choice.
Robotics have long been an important research field in artificial intelligence. Since the invention of the first robot in the 1950s, scientists have spent more than 60 years researching various robots that can imitate the movements of humans and creatures. Robot technology has continued to advance. In the past science fiction works, Many imaginations of robots have gradually become reality.
However, it has to be admitted that robots have not been able to escape the category of automated machines. Humans and robots are still defined as the relationship between use and being used, substitution and substitution. Robots are more of a A functional replacement tool emerges. In the design process, robots seldom consider the close coordination and cooperation with people in the same space, so that basic issues such as intrinsic safety, human-machine collaborative cognition and behavioral mutual assistance are not well solved, and flexibility and intelligent decision-making capabilities are not well solved. Still insufficient.
Integration, collaboration, sharing and governance are the distinctive features of the future intelligent society. With the rapid development of artificial intelligence technology, intelligent robots have gradually entered human life and are working hard to become one of them. An indispensable member. In communities, hospitals, nursing homes, factories... people expect robots to play multiple roles and perform cross-domain tasks in different scenarios. However, for robots to achieve intelligent tasks, seamless communication and harmonious integration with humans, they also rely on revolutionary breakthroughs in the ability to perceive, understand and interact with the environment.
Li Xinde concluded that intelligent robots currently face three main problems in place perception and understanding, namely, the lack of integrated utilization of multi-source information on highly complex places, the lack of effective clues for identifying places, and the lack of perception processes. Semantic interpretability results in current methods generally having low recognition rates and making it difficult to integrate into intelligent tasks. In order to overcome the above-mentioned shortcomings and effectively expand and extend human capabilities, the future development direction of robots - integrated robots - has emerged. However, if we want to achieve the goal of "integrated integration", we also need multi-faceted innovations in robotics technology.
The so-called "integrated robots" refer to robots that can naturally interact with the operating environment, people, and other robots, independently adapt to complex dynamic environments, and work collaboratively.
Li Xinde emphasized that "integration" specifically contains three meanings: The first is the natural interaction between the robot and the environment. Traditional robots work in specific areas with simple goals and single objects. However, it is difficult to have such a pure environment in actual production and life. Letting robots go out of the workshop, into mountains, snow, and onto streets where people come and go, and even let robots and people work together in a dynamic environment is the first goal of integrated robots.
The second is the mutual assistance and complementation between robots. Two people can lift heavier objects, a team of three F1 racing service personnel can change a tire in two seconds, and a complex medical operation requires the cooperation of multiple doctors and nurses. How can robots achieve breakthroughs in individual capabilities through the collaboration of multiple "people"? This is the second goal of the development of integrated robots, to achieve collaboration and complementarity between robots, and to enhance the overall function and performance of robots through cooperation.
The third is the collaborative work between robots and humans.
Cooperating with people doesn’t just mean listening to people’s instructions. Fusion robots also need to have their own “judgments” and “ideas.” Humans are complex and dynamic, and collaboration with humans is a higher requirement for robots. At this time, the robot no longer passively waits for human assistance, no longer blindly relies on human programming, but must be able to actively perceive and solve various problems. This requires the robot to be able to sense the environment in which it is located, so as to determine what tasks it should perform. On the other hand, in the era of human-machine integration, in order to allow robots and humans to communicate seamlessly, the expression of target semantics and place semantics in natural language interaction is very common. When robots/unmanned systems are oriented towards interaction and operations based on these concepts, , but it is difficult to understand and perceive these semantic concepts, "For example, going to the ward to take care of patients, and going to the kitchen to boil water. In addition to performing specific actions or tasks such as taking care of and boiling water, the robot must also be able to analyze the references of the ward and the kitchen. What kind of place is it?" Li Xinde said that, therefore, intelligent robots’ perception of places, and further their understanding of places, play a supporting role in the execution decomposition of intelligent tasks and good interaction between humans and machines.
The reason why human beings can adapt to various environments is inseparable from their keen perception of different environments. Fusion robots also need this diverse perception ability. "It is worth noting that diversified sensing is not as simple as adding a few sensors. There is also a lot of work to be done in signal transmission, calculation, recognition, etc. In many cases, multiple models are needed to identify the external environment and the intentions of others. The fusion processing of dynamic signals requires real-time perception, understanding and even continuous learning by the robot,” Li Xinde emphasized.
With the support of the National Natural Science Foundation of China’s major research projects, Li Xinde led the team to work hard and proposed a unified deep neural network with multi-modal fusion, from the "prototype" clustering of place semantics, semantics From the perspective of graph correlation reasoning and multi-modal fusion interactive correlation reasoning, we conducted in-depth research on robot place perception and understanding algorithms and robot autonomous motion operation control. Aiming at the problem of how to enhance the semantic interpretability of the place perception process, from the perspective of natural language description, the definition of place concept under semantic clues is proposed and the corresponding recognition methods are studied. In order to solve the problem of place semantic understanding and identification under complex environmental information, from From the perspective of simulating the human cognitive process of place, research and propose methods based on the combination of images and their natural language representation, and also study the related image description generation technology and multi-modal fusion technology; from the perspective of the place understanding framework Starting from this, we researched and proposed a place identification and understanding framework based on knowledge graph reasoning, and provided a knowledge graph construction method and reasoning model in the place field. With this as the core basis, it successfully applied for and was approved for the Guangdong Province Pearl River Talent Innovation Team Project "Research and Development and Industrial Application of Key Technologies for Cross-Domain Unmanned Systems in Multi-Mission Scenarios".
Li Xinde’s team conducted in-depth research on this project to further carry out tasks such as human-computer interaction, service strategies, task-level planning and autonomous operations related to the place, and to help intelligent robots smoothly enter human homes and daily life. , laying a solid foundation for seamless communication and harmonious coexistence with human beings.
Currently, the world's major powers are competing around the core technology of artificial intelligence. Whoever runs faster will be able to seize the commanding heights first. From this point of view alone, research around artificial intelligence is of great significance and is related to China’s strategic deployment of building an innovative country and a world science and technology power.
In recent years, China has included robots and intelligent manufacturing into priority areas of national scientific and technological innovation, and has stepped up planning and solid advancement.
Li Xinde is eager to accept difficult challenges, and at the same time, he can analyze the situation calmly. In his opinion, in the field of artificial intelligence and robotics, compared with its foreign counterparts, my country's disciplines started relatively late. However, with the development of the economy, the country has significantly improved its funding investment and emphasis, and there is great potential. As we catch up, the gap with the international community is gradually narrowing.
Over the years, Li Xinde has worked hard and devoted all his efforts to rooted what he has learned in the soil of the motherland, so that artificial intelligence and robotics technology can thrive on Chinese soil. stand up.
Li Xinde often laments: "Scientific and technological workers need to have a sense of social responsibility, and do some research and development and experiments with the emotion of benefiting the country and the people, produce first-class results, solve some practical problems of major national projects, and improve people's lives." A standard of living is what science and technology workers dream of!”
As an explorer and practitioner of robotics technology, Li Xinde said this and does this as well. In many areas related to the national economy and people's livelihood, the results of his wisdom and hard work are at work.
——For intelligent manufacturing and modern logistics, such as disordered sorting, intelligent welding, dynamic assembly and other robot application fields. During the operation process, the robot is required to conduct inspections on the work site and the shape, posture, etc. of the workpiece. Dynamic high-precision three-dimensional measurement poses new challenges to structured light sensors, requiring them to have the characteristics of high precision, small size, fast speed, strong environmental adaptability, and strong anti-interference ability. In the past, smart welding equipment systems were completely monopolized by foreign countries, and the high costs of hundreds of thousands or hundreds of thousands of yuan often seriously hindered the development of the industry. With the funding of Jiangsu Province's key scientific and technological achievements transformation project - "Welding Robot Complete Equipment", Li Xinde's team independently developed a set of structured light vision sensors and welding seam tracking systems, breaking the high-price market monopoly of foreign tracking equipment and creating a huge market. Social and economic benefits. With strong funding from the key project of the Jiangsu Provincial Key R&D Plan - "Key Technologies and Applications of High-Precision Structured Light Dynamic Measurement Systems Based on MEMS Chips", Li Xinde's team forged ahead and opened a new era of 3D stereoscopic visual dynamic perception.
——In the field of health services, according to WHO, 400 million people in the world have mental disorders. Mental illness has become an important disease in the 21st century. Depression is now leading to people losing their health. The fifth leading cause of mobility impairment. But what is gratifying is that the daily actions of patients with early-stage depression can well reflect their symptoms, making it possible to monitor and analyze their behavior with machine vision. Therefore, Li Xinde's team uses non-contact visual perception methods and deep learning technology to conduct intelligent analysis of human behavior, emotions, mental state, etc., to achieve estimation of the human body's mental state. This research was funded by a national key project. Its results are aimed at intelligent monitoring scenarios, using cameras laid out in intelligent environments or mounted on robots to acquire non-contact information in the natural state of the human body (interactive/non-interactive), combining prior knowledge and Scene/place understanding, and then analyze emotional and depressive states by fusing multi-modal information such as expressions, micro-expressions, eye movements, and gait, and even fusing gaze and character interaction behavior during character interaction, which has great application in depression technology. Originality. The results also laid a solid foundation for service robots to analyze the online status of service objects and provide high-quality services.
——In the field of national defense and security, Li Xinde has carried out in-depth cooperation with a number of research institutes and undertaken a number of national defense pre-research projects, including small target detection, identification, and tracking, individual abnormal behavior detection, and group monitoring. Rich research results have been achieved in abnormal behavior detection, violent behavior detection, multi-machine collaboration, swarm intelligence, air-ground cross-domain collaborative operations, etc., and it has broad application prospects in national border defense, street fighting, reconnaissance assault, security and other application fields.
Artificial intelligence is not only accelerating its development, but also upgrading. It is a strategic technology that will lead the future. Countries around the world have realized that this will be a key arena for competition between countries in the future, and they have regarded the development of artificial intelligence as a major strategy to enhance national competitiveness and maintain national security.
The large population and complete industrial structure provide China with the potential to create massive data and a vast market. But China still needs to do a lot of basic work, such as a more open data environment and well-trained scientific talents. In particular, continued efforts on talent will be a decisive factor.
"Whoever has first-class innovative talents and scientists can take advantage of scientific and technological innovation." Li Xinde naturally understands the power of the team and the importance of talent echelon construction. In fact, since he accepted the invitation to serve at Southeast University in 2007, he has taken root in this fertile soil for more than ten years. He not only engages in scientific research, but also takes on the important task of cultivating talents.
In Li Xinde’s mind, his relationship with his students is not only that of teachers and students, but also friends. They are also teammates who will make China lead the world in the field of automation research in the future. When dealing with team members, Li Xinde shows his unique kindness and candor. Rather than calling him "Teacher Li" or "Professor Li," he prefers to hear students call him "Old Li."
Li Xinde often says that every student is unique, "Never underestimate any of your students, they are all capable." This is a sentence that Li Xinde often talks about. . He will teach students in accordance with their aptitude according to their personality and characteristics, and always explore students' potential.
"The most important thing is not your learning ability, but your ability to analyze, solve and innovate problems." Li Xinde repeatedly warned team members, "Your future competitors will not be people, but machines. If you ask it to learn, it will be faster than you; if you ask it to repeat the work, it will be harder than you. So what do you have to compete with it? You must have a forward-looking vision, and don't just focus on it. In order to broaden students’ horizons and increase their knowledge, Li Xinde will provide them with many opportunities to participate in various conferences, communicate and give reports, so that they can receive training and encourage them. If there are opportunities for learning and exchange at home and abroad, be brave enough to go out and breathe more "fresh air."
"Exchanging abroad is a good thing to broaden their horizons, and it is also a very important part of the growth of young people, which will make their thinking more active." At that time, Li Xinde served as a nationally-sponsored visiting scholar and postdoctoral fellow. He visited Georgia Tech University in the United States and the National University of Singapore, where he met Professor Donald Arkin, the father of American robotics, winner of the U.S. Time Magazine's Best Invention Award, and a famous robotics expert, and Professor Ge Shuzhi, an academician of the Singapore Academy of Engineering.
"They are all people of great wisdom, especially Mr. Ge. Our chat is not limited to academic aspects, but also involves life philosophy. When communicating with someone like him who stands tall and sees far, we will learn more. It gave me a lot of inspiration," Li Xinde sighed.
From the initial days of working alone and with nothing to now becoming increasingly large and powerful, Li Xinde led this rapidly growing team onto the scientific research highway. The management model of Li Xinde's team is very open. Although he is the "top leader" of the team, he never engages in "one-note speech". Instead, he sets up a management working committee, which is divided into several groups according to different projects and different directions. Things are decided by committee. Li Xinde’s knowledge and conduct have been unanimously recognized and respected by team members. Under his leadership, a robotic automated iron army with strong cohesion and combat effectiveness was born. In 2019, the graduate student team guided by Li Xinde went out for the first time and won the special prize in the China Aeronautical Society Target Detection and Recognition Challenge.
"Last year's epidemic had a great impact on us. We planned to have 6 projects to be completed. Since the school was closed in the first half of the year, we basically discussed various issues during the implementation of the projects online." Li Xinde Recalling last year's scene is still thrilling, "But at the critical moment, the team showed its combat effectiveness, found ways to overcome various difficulties, and finally completed the task on time and with high quality!" Li Xinde's words revealed his pride in the team. In his Under his leadership, such a team with enthusiasm for exploration, innovative spirit and international vision has made solid progress on the road of scientific research, step by step, and has achieved outstanding results.
But Li Xinde is not satisfied with the achievements. “On the one hand, in scientific research, we have to solve a specific problem. We must encourage cross-integration of teams and have the courage to do things that we have not done before. Climb the peak of science and lead the trend of scientific and technological development; on the other hand, in engineering applications, we must integrate theory with practice, pay attention to the transformation of industry-university-research results, and implement them into practical applications. At this time, we must break through the boundaries of our abilities, learn, and apply. Communicate and cooperate with people who have new knowledge.” In his view, only in this way can the researcher himself or even a team become stronger and stronger.
There are different opinions on how far the rapidly changing artificial intelligence and robotics technologies will develop in the future. “There’s actually no way to predict it accurately.
Li Xinde changed the topic, "Scientific research is not a written script, but to bravely explore the unknown world. You may succeed or you may fail. Isn't the joy of scientific researchers in the process of exploration?" ? ”
Li Xinde, professor and doctoral supervisor at Southeast University, concurrently serves as deputy director of the Intelligent Robot Committee of the Chinese Society of Artificial Intelligence and deputy director of the Intelligent Products and Industry Working Committee of the Chinese Society of Artificial Intelligence. Long-term commitment For research in machine vision, machine perception, intelligent robots, human-computer interaction, unmanned systems, etc., he has been selected into the Jiangsu Province Qinglan Engineering Talent Training Program, Jiangsu Province Six Peak Talent Training Program, Southeast University National Outstanding Youth Training Program and Chief scientist of key projects of Jiangsu Provincial Key R&D Plan, etc., Guangdong Provincial Pearl River Talent Innovation Team, Jiangsu Provincial University Excellent Innovation Team, etc.