After the break, everyone sat down again and the meeting continued.
But this time, everyone seemed to have forgotten the dispute and unhappiness just now, and focused all their attention on this project.
Academician Tang Shizhen put on her reading glasses and, with the help of her assistant, started using PPT to speak to everyone.
“No matter which control method we choose, whether it is remote control technology, automatic control technology, or a combination of the two.
All automatic control technologies have a prerequisite, that is, the platform they are equipped with must be sufficiently automated, digital, and intelligent.
To put it simply, the aircraft model targeted must be sufficiently advanced and must have strong dynamic perception capabilities.
Otherwise, no matter how advanced our intelligent driverless system is, we can't use it on old models like Yun 5 or Yun 7.
The control of these aircraft is completely controlled by mechanical power transmission, or even purely mechanical transmission. It is impossible to modify these aircraft.
Therefore, I think it is very appropriate for Deputy Director Qi to propose using the MA 700 model as our modification experimental platform.
Of course, it can only meet our current needs, and we still need to test the versatility of our technology on different models in the future."
Having said this, Academician Tang Shizhen changed a PPT, which showed a picture of a company's MA700 aircraft in flight.
"Although this MA700 regional passenger aircraft is advanced enough and the digital flight control system meets our requirements, it is not enough for our project. It cannot be used directly. We need to modify it to meet the requirements."
Our experimental needs.
For example, it lacks the dynamic perception capability of the outside world that we need. The so-called dynamic perception capability refers to the aircraft's ability to acquire external information. This includes the radar on the aircraft and various flight sensors, such as airspeed indicator, altimeter, level gauge, and air pressure.
Plan and so on.
It even includes some surveillance cameras outside the aircraft and radio receiving devices, which are all part of the dynamic perception system.
The external data information transmitted back by these data will be transmitted and summarized to the cockpit instruments to assist the pilot in controlling the aircraft and meet the pilot's needs for flying the aircraft.
But this is far from enough for our intelligent driverless system.
First of all, the system is not a human being and does not have the sensory organs of a human body. For example, vision, touch, hearing, and smell, the most important of which should be vision and hearing.
The system cannot directly obtain these data information and needs to receive these data through various devices.
In fact, during air flight, pilots rely on radar to observe the situation ahead most of the time, and rarely use vision directly. Vision is generally mainly used for takeoff and landing, as well as handling some emergency situations.
Of course, what I am talking about here is civil aviation passenger aircraft and transport aircraft, not fighter jets.
This is not to say that vision is not important, in fact it is very important, especially when dealing with some emergency situations, as well as when taking off and landing, which requires pilots to have sharp visual observation capabilities.
The same is true for the system, which also requires strong visual observation capabilities."
"In this regard, we have synthetic aperture radar and photoelectric pods that are widely used on UAVs. The related technologies are very mature and we can install them directly on the aircraft." Director Song from the flight test center interjected.
Academician Tang Shizhen put down the teacup in his hand, then nodded and said: "Yes, but it is not enough. This type of photoelectric pod can satisfy general small drones or aircraft, but it cannot satisfy medium-sized and large aircraft. For example, medium-sized and large aircraft
Its huge body can block the large field of view of this type of photoelectric pod, resulting in incomplete observations.
No matter how advanced the photoelectric observation equipment is, they are still different from what we see with our real human eyes.
Therefore, if you want to realize automatic control or remote control, the vision problem must be solved first.
Regarding this aspect, we specially set up a photoelectric observation system research team, and they proposed a preliminary design plan.
That is, multiple photoelectric observation devices are installed at different locations on the entire aircraft to facilitate the system and our remotely controlled ground pilots to better perceive and grasp the dynamic information outside the aircraft."
A PPT was immediately displayed on the big screen, which showed a MA700 regional airliner with markings on it at different locations.
Looking at the annotated information on the picture, Director Song of the Flight Test Center immediately shook his head and raised his voice to express his different opinions: "I don't think it is necessary to be so complicated. You are making it too complicated.
The acquisition of data does not necessarily require so many photoelectric observation equipment. In fact, we can completely rely on some sensors to solve these problems.
For example, takeoff and landing, these data information can be obtained through relevant sensors. A key data for successful takeoff and landing is speed.
Whether it is taking off or landing, the speed is controlled. When taking off, the speed must be fast enough so that the aircraft can fly off the runway.
When landing, the speed needs to be slow enough so that the aircraft can land smoothly. We can do this according to the established programming.
At present, the take-off and landing technology of drones is also very mature, and some drones have even achieved autonomous take-off and landing.
In this regard, we can learn from the technology in this area, and there is no need to follow human thinking patterns."
Wu Hao nodded when he heard this, and then said with a smile: "It is true to fully obtain the dynamic data information outside the aircraft, but it does not need to be so complicated.
Adding such complex equipment will increase costs on the one hand, and on the other hand, the more and more complex the equipment, the more likely it is to malfunction.
In the field of photoelectric observation technology, we have actually been conducting research in this area, and have achieved good results. For example, the intelligent image recognition technology we use on various types of drones is very suitable for use as drones.
s eyes'.
This technology can very clearly identify the data information in the images observed by photoelectric equipment and process it accordingly. Its ability to capture screen information can fully meet the needs of our project."
Hearing Wu Hao's words, Director Song looked at Wu Hao and said, "I have heard about your image recognition technology for a long time. We will have more exchanges in this regard. I think there is great potential that can be tapped."
"No problem, I will ask someone to compile a detailed information and send it to everyone later." Wu Hao smiled and nodded.