On the afternoon of the 13th, the CCF-GAIR Conference welcomed the fifth lecture of the “Special Field of Robotics and Dronesâ€, Academician of the Chinese Academy of Engineering, Major General of the People’s Liberation Army of the Chinese People’s Liberation Army, Senior American Visiting Scholar, Postdoctoral Supervisor, Domestic Exploration UAV and None The bottom-level technical expert and founder of the field of man-machine operations, Professor Fan Bangkui brought us a keynote entitled "Interpretation of Drones and Artificial Intelligence."
Professor Fan Bangkui first reviewed the development history of UAV automation, then analyzed the various requirements of UAVs for artificial intelligence technology, and finally looked into the future of UAV+AI integration.
Professor Fan Bangkui said that drone automation has undergone the following three stages of development:
1. The era of manual control with gyro control and basic radar navigation as the core technologies;
2. The era of semi-automatic control with inertial navigation system control and terrestrial radio positioning system as the core technology;
3. Finally reached today's self-navigation with a satellite/inertial combination, using a dedicated data link remote control and tracking and positioning, can be modified in real time the era of automated air drones.
Although we have ushered in the era of drone automation, Professor Fan Bangkui believes that the core of drones is the control system. Automation does solve the problem of automatic control of drones, but it also makes the task more meaningful for users. There is no help, and this is exactly why drones need artificial intelligence.
As for the demand of artificial intelligence for drones, Professor Fan Bangkui believes that it is mainly divided into the following three parts:
1. Smart flight. Including awareness and avoidance, flexible and fault-tolerant, target tracking, path planning and other specific functions, the key technologies involved are the intelligent control during flight, the flexible integration of airspace, and the complete environment adaptation.
2. Smart collaboration. Including the implementation of task coordination and location sharing and other specific functions, related to the coordination of command and control, collaborative situation generation and assessment, collaborative semantic interaction and other key technical points.
3. Smart task. That is, drones can autonomously determine what needs to be done next, and then give orders to themselves. The key technologies involved include pattern recognition control of speech, text, and images, neural control of artificial neural networks, and fuzzy control of fuzzy sets.
With regard to the future of drones + artificial intelligence, Professor Fan Bangkui said: In the era of manned machines, today is an era in which drones and manned machines are mutually complementary and complementary. In the future, it will be an era of drones. More specifically, drones will affect our future from the following three aspects:
1. Intelligent drones will change people's lifestyles, such as sending drones to express delivery.
2. Intelligent drones will change industrial models, such as automated drone industrial production.
3. Intelligent drones will change the form of future warfare, such as U.S. dead and Predators and other unmanned reconnaissance aircraft and bombers.
Finally, Professor Fan Bangkui said: "The future will be an era where everybody has a robot and everyone has a drone," and we are also destined to coexist with these intelligent products.
The following is a speech record:
Distinguished guests, good afternoon. I am very happy and very honored to come to the summit organized by the association. First of all, I would like to thank the high academician for giving me this opportunity to report on the development of drones and artificial intelligence. In fact, I was a pupil of artificial intelligence, but I feel that artificial intelligence will play a crucial role in the development of drones. It can be said that it ranks first in disruptive technologies. I also learned from young people in this year's Lijia, catching up with a sleek, and talk about artificial intelligence today. Then report from three aspects. First, briefly review the development process of drones in the automation process; second, talk about the technical requirements of drones for artificial intelligence; and third, talk about drones once they have artificial intelligence. What kind of impact on the future society.
The UAV is driven by power, and is drone on the aircraft. The aircraft that can be used repeatedly is abbreviated as "reusable". It is distinguished from missiles and other equipment. What is a UAV system? It is a platform, a load, a data link, an information processing device, and an integrated support device. It controls the execution of a flight system. The essence of a robot is a machine. It contains three elements. The first is the sensor, which is the detection and perception. The second is to deal with the so-called artificial intelligence. Third, to make changes to the environment based on various environments and perceptions, which is often referred to as execution. Drones also have these three elements, so drones are also called aerial robots and drones. It is actually a category of robots. In recent years, drones are the most popular robots. The most advanced technology is the robot with the highest penetration rate.
The development of drone automation, drones are unmanned on the plane, mainly through the ground control, in the past is the use of people to manipulate the flight, so the control technology is the core key. Then, humans experienced manual control, semi-automatic control and automatic control during the development of drones. The next step will be intelligent control. A simple review of the manual control phase used a simple gyroscope to control the stability of the aircraft. This was the earliest time Spey used the first gyro principle to use the first drone. Commands and sending remote commands at the same time. Earlier electronic radios used ground radar to track the operation of the aircraft to control drone flight. It could not do track planning, and most of them were used as drones.
With the rapid development of information technology, semi-automatic control has emerged, and gyro navigation systems are used to control the stability of the aircraft. There are various gyros. But now there are more gyros, fiber optics, and laser shedding. The radio receivers on the ground receive remote control commands, and the terrestrial radio system confirms the position of the drone. The airshow can be planned on the ground. However, it cannot be modified by air to begin military operations. In the Battle of Bekaa in 1982, this was a drone. A typical experience on the battlefield is where control technology plays an important role. During the fighting in Vietnam, the U.S. deployed a "Fire Breeze" drone, flying a total of 3,435 flights and obtaining 80% of air surveillance photos.
At the end, we are now in the stage where we are now, or it was an earlier era of fully automatic. Using WeChat and inertial integrated navigation for positioning, autonomous positioning. In the past, a lot of positioning was through external positioning. Now that you have satellite navigation, you can position yourself. Remote telemetry, tracking and information transmission using dedicated data links. This is a predator drone data link, which is very complex, with radio waves at the C-end, and emergency extra-high-level waves, etc. This development has been greatly improved. This time can be planned in real time, with CPU, DSP and other digital control system. This is the X-47B of the United States. It is a drone that is used on a voltmeter aircraft carrier and can automatically take off and land on an aircraft carrier. On the right is the U.S. "Death" drone that we have often heard of. The operation can now be completely automated. With a variety of equipment to perform various tasks, it is not detailed here.
Simply put, in 1991, the Gulf War, the first real-time transmission drone war, the Kosovo War in 1999, the first application of SAR detection drones, the Afghanistan war in 2001 and the 2003 Iraq war, UAVs were put into use. From 2003 to now, the U.S. global fight against terrorism has seen drones. This is a U.S. ultra-high-speed drone, and the other one is a large plane that is brought up to carry out group operations.
The core of the UAV is control, and its automation development mainly solves the flight automation control and has not yet solved the problem of task autonomy. This requires artificial intelligence to solve, and later I report this issue on the second question. In my opinion, artificial intelligence technology will be the top of the subversive technology for the next phase of UAV development. We are also engaged in drones and are now paying attention to artificial intelligence.
Drone demand for artificial intelligence. There are differences between the intelligence of drone systems and others. I think that there are three aspects to the intelligentization of drones. The first is intelligent flight, the second is intelligent coordination, and the third is intelligent tasks. Smart flight is the foundation, and intelligent collaboration is the way to realize our intelligent tasks. Finally, we must achieve this goal of smart missions.
What is smart flight? I use a simple animation to explain that drones are rapidly falling, agile maneuvering, and how to fly. Avoid the obstacles, can go to the community, and then change the state from indoor to outdoor. This time requires a variety of tasks to complete, there are several characteristics of intelligent flying.
To have the ability to perceive and evade, there are many ways to use it. You can use LAB, you can also use microwave radar, you can also use photoelectric, these days we all introduced. This is intelligence perception and avoidance. This is the basis of intelligence. At present, we have a drone that we can do in China. It can stick to a fixed-distance flight and test flights under the bridge. Second, to be flexible and fault-tolerant. In the process of flying, one wing is missing, or the wing is damaged. If a wing is missing, it can still fly. This is called fault tolerance. Third, tracking. This is the most basic. It can track any target during the flight and requires target recognition, including tracking technology. This place has to be tracked differently and the technology is different. Fourth, path planning. This I use a "bird". This bird is called "fine bird". It can guarantee seven rounds of flights around itself. In the future, the planning of a single drone must be done at random. And when I talk about collaboration later, I need each coordinated action to be a random plan like a bird.
For intelligent flight, the human brain and cerebellum are all running, and drones need to be stable from flight to navigation control. The key technologies for intelligent flight, intellisense and avoidance, reconnaissance, detection, communication, sensors, and intelligent path planning are all used in artificial intelligence. Based on the feedback data at each node of the neural network, we solved it with a genetic algorithm. Everyone here is an expert. I just know a little bit about it, and it is also a trick. Intelligent flight technology requires a combination of robustness, openness, autonomous decision making, and intelligent learning. The combination of airspace and airspace must also be intelligent. These can be adapted to the rapid development of drones. If drones are industrialized, the drone's air traffic control will not be able to support it as it is now. Airspace intelligence is also inevitable. Intelligent aircraft, bionic aircraft, rotors, fixed aircraft compliant aircraft, variant aircraft, cross-media vehicles. The layout of the aircraft itself also needs to be solved intelligently, all of which constitute the basis of intelligent flight.
Smart collaboration. One is the ability to cooperate when performing tasks; the use of shared cross-domain UAS sensors to transmit information. The U.S. military’s navy should engage in drone grouping. The key technologies in this area involve coordinated command and control. Dynamic, ad hoc network communication technology, control of flight, mission planning and target allocation technology. Synergetic situational language evaluation is also needed to be solved in smart collaboration. It is very important for smart collaboration. Collaborative path planning, on the basis of the previous individual planning, also requires collaborative planning. Collaborative semantic interaction, as we all know, in future collaboration, coordination between UAVs and drones, cooperation between drones and manned aircraft, cooperation between drones and other unmanned systems, including ground Control System. Machine understanding of natural language really makes semantics interactive.
Smart task. This is a U.S. military warfare now that drones can track a target. However, the drone could not find the target on its own and gave orders to itself. The vice chairman of the U.S. Joint Staff Association pointed out that today drone operators are sitting there and staring at death TV for hours, trying to find a goal or see something moving or to determine that it is a goal. This is The waste of human resources is inefficient. In the end, the drones have to do smart tasks, that is, the tasks must be judged by themselves. To change the architecture of drones, the current architecture is a ground warfare center. There are a lot of people in the operation. Next, Its center must be moved directly to the plane and truly accomplish its own tasks. This is the ultimate goal. The future battle will be machine warfare, which is between machine and machine. It needs to be intelligent. In the film "Top Secret Flight," how does it do intelligent tasks? Air refueling, do not let it go in, this time the machine is smart, you can add oil, but also to destroy the probe. This task is accomplished intelligently. No one needs to know how to perform it. What the place needs is the learning control of pattern recognition (voice, text, and image). Pattern recognition is the recognition of images. For people, everybody sees them at a glance and no matter how they change, they are all apples. However, it is difficult for the machine to analyze. In the future, during the development of drones, it will be very important for drone control if we can recognize such things as voice, images, and text. The second is the control of the artificial neural network. Seeing these screens can generate control models through artificial neural networks. Expert system, rule control. Combining with expert systems, in the future when it comes to intelligent tasks, as a drone, although deep learning of intelligent machines is very important, it is also very important for expert systems. It must be able to work together to achieve intelligent tasks. Fuzzy set of fuzzy control.
Prospective to the future. Once drones have the breakthrough technology of artificial intelligence, what impact will it have on the development of drones? The development rules of air vehicles have been dominated by man-machines in the past. In our present era, drones are an important complement to man-machines; the next step is that drones and manned aircraft depend on each other; if they go down, there is no man-machine The important supplement of the machine, the final stage, must be UAV unified world.
Since drones will dominate the world, smart drones will change their lifestyle. Using drones to send water, this is primary. What will be the future? If Prof. Kumar likes to drink tea and another person likes to drink coffee, big data will come into play at this time. After adaption, it can be escorted to the past and it can be adapted to machine services. The transformation of human services brought about by the intelligence of drones. This is a drone father. His children have to insist on going to school on their own, but his dad is not at ease. What should he do? He brought him a drone following him. It is still a machine. What to do next? When the child walks on the street, the drone can follow him, recognize him, and keep a certain distance from him. The status of his child can be seen on the parents' mobile phone at any time. . For example, it is now feared that children will be trafficked when they go out and that drone tracking can solve them. This is an example.
Artificial intelligence will enable people to “look at the worldâ€, and now large-scale competitions are conducted by drones in real time. This duck-like drone can be opened at any time to fly. Smart drones will change the industrial model, drones will be transformed into courier brother, this era is coming soon. At present, in addition to the traditional industry research aircraft drones, we can count, now the world engaged in information, the Internet, which one is not involved in drones? Is there any drone that involves any international company, including domestic big companies, engaged in information? At present, drones are also the future of robotics. If you don't get involved in this field, how can you say it? Recently, the unit recently fired at Houchangcun Road in Beijing. The opposite is the Phase I and Phase II of the software park. This backcourt is well-known for the village road. The well-known information is all in this software park. When you settled in, you are too embarrassed to say that you are engaged in the Internet. Later many years later, the company did not engage in drones to engage in information technology. On March 25 this year, the United States listed a city in Nevada as the pilot city for the first drone express delivery service. With the release of the low altitude, this business will certainly develop.
If the houses are structured, will drones not build houses in the future? Smart drones will change the form of war. This is also what Obama’s military advisor Singh said. The machine war represented by drones will change the form of warfare for 5,000 years. The emergence of drones has made the boundaries between strategy and tactics more and more blurred. Strategic corporals and strategic soldiers have emerged. Strategic corporals will only act in battle and will mobilize large systems to combat this goal. In turn, an Admiral and Lieutenant can command a very small operational unit. The front and the back are integrated. This is the case of U.S. warfare in the U.S. war in Afghanistan and the war in Iraq. The real control station is the U.S. base in U.S.A. It is a remote-controlled war through satellite missions.
Intelligence will make group operations possible. The U.S. military is now studying group operations. This is the future development of drones, especially the key to smart technology in this industry. The role of the drone changes from the supporting role to the protagonist. More than half of Al-Qaida from Afghanistan was destroyed by drones. In 9-12, more than 1,300 missiles were fired by drones. The number of U.S. military operations over 11 years exceeded that of manned aircraft, and the number of UAV purchases in 12 years exceeded the previous. The form of war changed, and no prophet missiles were fired. The Afghan leaders said that the US tanks and armored units are nothing at all. The real fear is drones, but at the same time they laugh at the US military as a coward who dared not show up. At this time, the emergence of drones will change the mentality of human combat.
Everyone has robots, and everyone has drones coming. Ladies and gentlemen, are you ready? Thank you!
Moderator: Thank you, Academician Fan, because Academician Fan is very modest. He said that he is rushing in fashion. These fashions are created by top national experts such as Academician Fan. When we talked about drones, we also did service, what was the part of man-machine interaction that was definitely needed for service, human-computer interaction, and how important was the human-computer interaction during the three-dimensional sensing period?
Fan Bangkui: The three-dimensional sensor is closely related to the three elements of the drone that I talked about earlier. The first element is sensing, no data, no language, how to be smart, and finally how to do it. This is the first element that must be possessed.
Moderator: Thank you again, Academician Fan.
Fan Bangkui: I am very excited today. Mr. Lu Xun once said one sentence. Unjustifiably wasting other people's time is tantamount to financial loss. Today, so many artificial intelligences, including the robot brain, occupy everyone for so long. I am very sorry, thank you.
UCOAX manufactures a wide spectrum of wire and cable assemblies, each custom-designed for each individual client. Coaxial cable, at the industrial level, is used for instrumentation, control, RF, and microwave applications. It is also commonly used to connect home video equipment and cable television. These cable assemblies are available both rigid and flexible options, and consist of four layers: a round conducting wire, insulating spacer, cylindrical conducting sheath, and a final insulating layer. For any cable fabrication needs, contact UCOAX.
Customized Cables,Custom Pc Cables,Custom Computer Cables,Custom Network Cables
UCOAX , https://www.ucoax.com