CN201918032U - An anti-collision device for aircraft flying at low altitude - Google Patents

Abstract

The utility model provides a low-altitude flying anti-collision device of an aircraft, which comprises a multi-detector device, a computer system, a man-machine interface and an automatic flying control system. The multi-detector device comprises a GPS (global position system), a height gauge, radars and a photographic device, wherein the GPS is provided with an ultrasonic sensor and an infrared sensor, the radars includes a microwave radar and a laser radar, the computer system comprises a terrain matching calculating module, a target identification module, a situation threat assessment module and a decision module which are connected successively, the terrain matching calculating module is further respectively connected with the GPS and the height gauge, the target identification module is respectively connected with the radars and the photographic device, and the decision module is connected to the man-machine interface and the automatic flying control system. The automatic flying control system comprises an automatic collision avoidance system which is connected with the decision module. The low-altitude flying anti-collision device can be operated for fire extinguishment of complex places such as land, water and forests, can automatically take a collision avoidance measure, increases survivability of the aircraft, and reduces burden of pilots.

Description

An anti-collision device for aircraft flying at low altitude

technical field

The utility model relates to the technical field of aircraft flight equipment, in particular to an anti-collision device for aircraft flying at low altitude.

Background technique

Mountains, buildings, etc. may pose a potential threat to aircraft flying at low altitude and high speed. Statistical analysis by the Aeronautical Society of Japan shows that 50% of helicopter flight accidents are due to direct collision with low-observable obstacles. The Ground Near Collision Avoidance System is designed to provide safety protection for aircraft flying at low altitudes and avoid the danger of colliding with ground obstacles. Solving the problem of near-ground collision avoidance has become a major problem that many countries in the world pay attention to and strive to solve. At present, both at home and abroad are actively developing the ultra-low altitude collision avoidance system for aircraft.

In the process of emergency rescue and rescue, aircraft often have to perform missions under extreme conditions such as low visibility and ultra-low altitude of 60m-200m, and its flight safety is extremely vulnerable to threats from low-altitude obstacles. If the detection capabilities on the aircraft are limited to conventional observations such as human eyes or CCD cameras, the possibility of the aircraft colliding with natural and artificial obstacles is very high. Therefore, aircraft used for rescue and rescue put forward higher requirements for the detection capability of the near ground collision avoidance system.

In addition, the current domestic research on the ground proximity avoidance system is limited to the ground proximity warning system, ignoring the research on the aircraft automatically taking self-rescue measures in critical conditions. Taking the Jiaolong 600, whose main tasks are forest fire fighting and water rescue, as an example, higher requirements are put forward for research in this area. First of all, while fighting the fire, the flight safety of the pilot and the aircraft itself must be ensured. On the other hand, we cannot only consider potential threats. Rescue operations must count against every second. Therefore, it is very important to assess the existing threats and decide whether to take evasive actions. necessary. Secondly, the water source generally used for forest firefighting comes from the nearest lake or sea, so the firefighting aircraft must have the ability to identify water or land in a short period of time in order to adopt different operating modes.

To sum up, it is imminent to develop a set of near-earth collision avoidance devices with strong detection capabilities, which can identify detected obstacles as soon as possible and assess threats.

Utility model content

The problem to be solved by the utility model is to provide an anti-collision device capable of identifying obstacles and evaluating threats as soon as possible. The technical scheme that the utility model takes for solving its technical problem is to provide a kind of device of aircraft low-altitude flight anti-collision, and this device is made up of the computer system that multi-detector is connected with it, man-machine interface system, flight automatic control system:

Described GPS is connected with RS-422, is used for determining the three-dimensional coordinate of aircraft; Described altimeter is connected with RS-422, is used for determining the altitude of aircraft and ground; Described radar is that microwave radar or lidar are installed on aircraft head The external position is connected with MAX485 to generate a wave-to-wave transmission path, and the information detected by the detector is sent to the computer system through the MIL-STD-1553 bus.

Described computer system comprises the terrain matching computation module that matches successively, target recognition module, situation threat assessment module, decision-making module; Terrain matching computation module is connected with GPS and altimeter respectively, target recognition module is connected with radar and camera respectively, decision-making The module is connected to the human-machine interface and the flight automation control system.

The utility model can be operated in the complex occasions of water, land and forest fire fighting. It firstly detects obstacles through a multi-detector device, and then classifies the obstacles according to the D-S evidence theory algorithm, and then conducts threat assessment according to the target classification results. And make a comprehensive response decision based on the threat level, and finally display the result or inform the pilot through voice.

Compared with the prior art, the low-altitude flight collision avoidance device of the utility model adopts multi-detector configuration, so the detection capability is strong, and the obstacle identification and threat assessment capabilities are effectively improved. The utility model guarantees the personal safety of the pilots and reduces the accident rate of the plane to a certain extent. It can be operated in special occasions such as emergency rescue and rescue, which fills up the application of anti-collision technology in the aviation field in my country, and also contributes to the further development and improvement of our country. The ground proximity warning and collision avoidance system provides strong support.

Description of drawings

Fig. 1 is a kind of structural schematic diagram of the device structure of aircraft low-altitude flight anti-collision of the utility model

Fig. 2 is a schematic diagram of the structure of the Bayesian network model adopted by the device situational threat assessment module of a kind of aircraft low-altitude flight collision avoidance device of the utility model

Figure 3 is a schematic diagram of the structure of the flight automatic control system

Detailed ways

Below in conjunction with accompanying drawing, specifically illustrate the utility model.

Please refer to Fig. 1, which is a schematic structural diagram of a device for avoiding collisions in low-altitude flight of the utility model;

It consists of a computer system connected with multiple detectors, a human-machine interface system, and an automatic flight control system:

Described GPS is connected with RS-422, is used for determining the three-dimensional coordinate of aircraft; Described altimeter is connected with RS-422, is used for determining the altitude of aircraft and ground; Described radar is that microwave radar or lidar are installed on aircraft head The internal position is connected with MAX485 to generate a wave-to-wave transmission path, and the information detected by the detector is sent to the computer system through the MIL-STD-1553 bus.

Computer system 2 comprises sequentially matching terrain library 21, terrain matching calculation module 22, target identification module 23, situational threat assessment module 24 and decision-making module 25, terrain library 21, terrain matching calculation module 22, target identification module 23, situational threat Evaluation module 24 and decision-making module 25 are connected sequentially; Terrain matching calculation module 22 is also connected with GPS11 and altimeter 12 respectively; Target recognition module 23 is connected with radar 13 and camera device 14 respectively; Automatic control system4. The monitor is respectively connected with the above-mentioned decision-making module 25 and the man-machine interface 3 .

The man-machine interface 3 is connected with the display device, earphones and warning lights of the aircraft, and outputs the decision support provided by the decision module 25 to the pilot. The utility model all adopts prior art in this part, therefore, does not do detailed introduction here again. The multi-detector 1, the computer system 2 and the flight automatic control system 4 will be described respectively below.

As the detection device of the whole system, the multi-detector 1 is mainly responsible for obtaining the measured information, and its ultimate purpose is to meet the requirements of information transmission, processing, storage, display, recording and control. For the entire ground near-collision avoidance system, detection is the first and most important link to realize all functions.

In detection technology, commonly used active sensors mainly include ultrasonic, infrared, microwave radar and laser radar. The following table makes a horizontal comparison of these four sensors:

Table 1 Transverse analysis and comparison table of sensors

The utility model aims at the special working environment of large-scale fire extinguishing/rescue amphibious aircraft, combined with the above-mentioned analysis and comparison of various sensors, by simulating the function of the human brain to comprehensively deal with complex problems, making full use of multiple sensor resources, through the analysis of various sensors and The rational domination and use of its observation information combines the complementary and redundant information of various sensors in space and time according to a certain optimization criterion to produce a consistent interpretation and description of the observation environment.

The utility model multi-detector 1 comprises GPS11, altimeter 12, radar 13 and video camera 14, and GPS11 and altimeter 12 are all provided with ultrasonic sensor and infrared sensor, and radar 13 comprises microwave radar and lidar. Multi-detector 1 rationally and effectively configures the entire multi-dimensional detector system, the layout scheme of multiple multi-dimensional detectors and the connection relationship among them and the way of providing information, breaking through the traditional mode of relying on a single sensor to provide information in the past, and has strong detection capabilities.

The computer system 2 performs target identification and situational threat assessment based on the information provided by the multi-detector 1, and provides decision support for the pilot based on the determined threat level of the target.

Target recognition is a prerequisite for situational threat assessment. The target recognition module 23 is connected to the radar 13, camera device 14, and terrain matching calculation module 22 to extract the Doppler frequency, polarization characteristics, target frequency response, and harmonics of the radar echo signal. Target. Terrain matching calculation module 22 is connected with GPS11, altimeter 12 and terrain storehouse 21 respectively again, and this target identification module carries out based on terrain database and GPS real-time data respectively, obtains information by terrain matching calculation module 22, as building, mountain range, forest etc. Target recognition; radar target recognition and target recognition based on multi-source image fusion.

Most of the images provided by the detector are multi-focus images. That is, when the camera shoots multiple targets at different distances from the lens, it cannot focus on these targets at the same time to make them clear. Multiple images are obtained by focusing on each target separately. Therefore, the utility model adopts the boundary invariant moment to identify the potential target in the image, first extracts the edge feature, and then uses the boundary moment feature to describe each target feature.

In addition, according to the types and influences of the environment and obstacles, the utility model divides the target types into nine categories: airplanes, buildings, mountains, wires and cables, birds, trees, fire, water, and land. The properties of the object are divided into five properties: position, speed, time, quantity and strength.

Target recognition is to separate the target from the background according to the characteristics of the target object obtained in the feature extraction process, and to determine the type, location and other useful information of the target. Target recognition usually has methods such as neural network, support vector machine, and D-S evidence theory. Compared with neural network, support vector machine can seek the best compromise between the complexity of the model and the learning ability according to the limited sample information, and obtain more High generalization ability. D-S evidence theory can effectively deal with uncertain information, but under conflict information, its normalization process often produces invalid results that are counterintuitive. Therefore, the utility model adopts a support vector machine (SVM, Support Vector Machine) method to carry out target recognition and classification. At the same time, in order to increase the computing speed of the SVM, the utility model improves the SVM algorithm in terms of support vectors and decision functions. First, select the base vector, and find a set of base vectors in the high-dimensional feature space. The number of base vectors is smaller than the support vector and all samples can be reconstructed by linear mixing of this set of sample subsets, so as to improve the operation speed. Then determine the decision function, according to the selected basis vector, determine the corresponding simplified decision function.

Situational threat assessment provides the basis for decision support. The situational threat assessment module 24 of the present utility model adopts a situational assessment method based on Bayesian networks, using Bayesian networks as a model, combined with a computable situational assessment method constructed by expert knowledge, and finally The target threat level is defined as four levels: high, medium, low, and none. As shown in Figure 2, it is the Bayesian network model structure schematic diagram that the utility model adopts, at first, by multi-detector device 1 detection environment information, after obtaining GPS information, height information, radar information and image information, then by target identification module 23 Determine the target type, target location, target speed, target time, target number, and target strength based on the above information. Finally, the situational threat assessment module 24 determines the threat level of the target according to the type and attribute of the target.

The decision-making module 25 is respectively connected with the situational threat assessment module 24 and the man-machine interface 3 , and provides decision support for the pilot according to the target threat degree determined by the situational threat assessment module 24 . As shown in Fig. 1, the decision-making module 25 is also connected with the automatic flight control system 4, in addition, it is also connected with a monitor, and the monitor is connected with the man-machine interface 3 again, when the threat level exceeds the boundary line, once the pilot fails to specify If the corresponding action is taken within a certain period of time and the threat still exists, at this time, the decision-making module 25 outputs the specific information of the obstacle to the flight automatic control system 4, and the aircraft will automatically take corresponding avoidance measures.

Each module of the computer system 2 has been described above, and it should be noted that the above modules are usually interactive and parallel, and there is no clear boundary for clear division. Please refer to Fig. 3, it is a kind of anti-collision device of aircraft low-altitude flight of the present invention, its MIL-STD-1553 bus line is connected with flight automatic control system, is used for the automatic manipulation and instruction driving of aircraft, the present utility model is in prior art On the basis of the system, an automatic collision avoidance system 41 is added to the system, and the automatic collision avoidance system 41 is connected with the decision-making module 25 and the aircraft control system 42 respectively. The obstacle information detected by the multi-detector device 1 passes through the target identification module 23, the situational threat assessment module 34 and the decision-making module 25 in sequence. action and the threat remains. In this case, the decision-making module 25 inputs the specific information of the obstacle (such as category, geometric size, etc.) The signal is sent to the track control system 42, and the track control system 42 controls the autonomous steering gear 44 through the autopilot 43, and finally realizes the change of the aircraft heading through the control operation of the rudder surface 45 to achieve the effect of collision avoidance.

The utility model relates to an anti-collision device for low-altitude flight of an aircraft, which has strong detection ability, and can perform obstacle identification and threat assessment, and can operate in complex occasions of water, land and forest fire extinguishing. In addition, the utility model also adds an automatic collision avoidance system. When the threat exceeds the warning line, the aircraft can automatically take collision avoidance measures, which improves the survivability of the aircraft and reduces the burden on the pilot.

The above disclosures are only a few specific embodiments of the utility model, but the utility model is not limited thereto, and any changes conceivable by those skilled in the art should fall within the protection scope of the utility model.

Claims (4)

1. device that the aircraft low-latitude flying is crashproof, it is characterized in that, the computer system that this device is attached thereto by multidetector, man-machine interface system, automatic flight control system are formed: described GPS links to each other with RS-422, is used for determining the three-dimensional coordinate of aircraft; Described altitude gauge links to each other with RS-422, is used for determining aircraft and floor level; Described radar is that microwave radar or laser radar are installed in the Vehicle nose position and link to each other with MAX485, produces the delivering path that involves ripple, and the information that detector detects is delivered to computer system by the MIL-STD-1553 bus.

2. the crashproof device of a kind of aircraft low-latitude flying as claimed in claim 1 is characterized in that, described computer system comprises terrain match computing module, Target Recognition module, situation threat assessment module, the decision-making module that is complementary successively; The terrain match computing module links to each other with altitude gauge with GPS respectively, and the Target Recognition module links to each other with camera head with radar respectively, and decision-making module links to each other with automatic flight control system with man-machine interface.

3. the crashproof device of a kind of aircraft low-latitude flying as claimed in claim 1 is characterized in that described computer system also comprises terrain lib, and described terrain lib links to each other with described terrain match computing module.

4. the crashproof device of a kind of aircraft low-latitude flying as claimed in claim 1, it is characterized in that, described radar is that microwave radar or laser radar are installed in the Vehicle nose position and link to each other with MAX485, and the information that detector detects is delivered to computer system by the MIL-STD-1553 bus.

Images