CN111044052B - Unmanned aerial vehicle self-adaptive navigation system and method based on intelligent sensing - Google Patents

Unmanned aerial vehicle self-adaptive navigation system and method based on intelligent sensing Download PDF

Info

Publication number
CN111044052B
CN111044052B CN201911408053.6A CN201911408053A CN111044052B CN 111044052 B CN111044052 B CN 111044052B CN 201911408053 A CN201911408053 A CN 201911408053A CN 111044052 B CN111044052 B CN 111044052B
Authority
CN
China
Prior art keywords
path
target
navigation
information
risk
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN201911408053.6A
Other languages
Chinese (zh)
Other versions
CN111044052A (en
Inventor
连峰
张鸣益
张光华
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Xian Jiaotong University
Original Assignee
Xian Jiaotong University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Xian Jiaotong University filed Critical Xian Jiaotong University
Priority to CN201911408053.6A priority Critical patent/CN111044052B/en
Publication of CN111044052A publication Critical patent/CN111044052A/en
Application granted granted Critical
Publication of CN111044052B publication Critical patent/CN111044052B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
    • G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
    • G01C21/20—Instruments for performing navigational calculations
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
    • G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
    • G01C21/005—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 with correlation of navigation data from several sources, e.g. map or contour matching

Landscapes

  • Engineering & Computer Science (AREA)
  • Radar, Positioning & Navigation (AREA)
  • Remote Sensing (AREA)
  • Automation & Control Theory (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Traffic Control Systems (AREA)

Abstract

The invention provides an unmanned aerial vehicle self-adaptive navigation system and method based on intelligent perception, and the system comprises: the environment and target intelligent sensing module is used for acquiring sensing information of a ground environment and a target, exchanging information with a remote end and sensing the environment and the target by intelligent reasoning to obtain a battlefield situation sensing map; the self-adaptive navigation module is used for planning a self-adaptive navigation path and selecting an optimal navigation path according to the minimum criterion of natural risk and war risk on the basis of a battlefield situation perception map; and the data display module is used for displaying the real-time detection data, the risk coefficient and the path distance of each feasible path and the current optimal navigation path of the unmanned aerial vehicle. The invention also discloses an adaptive navigation method based on intelligent perception, which comprises the steps of selecting scenes in a simulation system, completing the identification and tracking of targets and intention deduction, then adaptively selecting a navigation path according to the change of conditions, and finally utilizing possible navigation information source fusion to carry out intelligent navigation.

Description

Unmanned aerial vehicle self-adaptive navigation system and method based on intelligent sensing
Technical Field
The invention belongs to the technical field of tracking identification, and particularly relates to an unmanned aerial vehicle self-adaptive navigation system and method based on intelligent sensing.
Background
In systematic operation of future development, particularly in systematic operation of an unmanned aerial vehicle, an operation platform faces the problems of instantaneous change of battlefield situation, multiple interference factors of battlefield environment and the like, multiple uncertain factors are formed, so that a driving path planned in advance is not optimal any more, and the online self-adaptive sensing environment and the route re-planning technology of the unmanned aerial vehicle are important directions of military development, so that the unmanned aerial vehicle can really realize autonomous navigation flight in a complex and variable environment. Because modern weapon systems have characteristics such as mobility height, disguise are good, electronic antagonism performance is strong, and the early warning system for reconnaissance and tracking enemy's target must adopt many sensing detection such as radar, infrared, video, audio frequency, contacts different data of each check point simultaneously and fuses, and the problem of emphatic research includes reliable accurate target acquisition, tracking, identification, intelligent processing, logistics plan, maintenance plan, commander and control etc.. And an information fusion technology is adopted to achieve the purposes of accurate target acquisition, identification and tracking. The existing intelligent perception technology is mainly focused on the engineering application fields of intelligent vehicles, intelligent robot self-adaptive navigation and the like, the military field does not relate to the self-adaptive navigation application based on intelligent perception, and the self-adaptive navigation theory based on intelligent perception is not completely formed. In modern air-ground integrated battles, particularly unmanned aerial vehicle system battles, because many links do not need direct participation of people, the research on an intelligent perception-based adaptive navigation theoretical method is strongly needed.
Disclosure of Invention
In order to overcome the defects of the prior art and solve the problem that the prior art related to battlefield environment adaptive navigation has no systematic development, the invention aims to provide an unmanned aerial vehicle adaptive navigation system and method based on intelligent perception, which can intelligently perceive battlefield situations and perform instant navigation and provide a solution for a moving body to adaptively select a traveling route according to ever-changing battlefield situations on the basis of acquiring situation perception maps. On the premise of having the intelligent sensing function, battlefield situations including the categories, positions, motion rules and intentions of targets of both enemies and my parties are obtained through intelligent sensing, and therefore a navigation possible path vector diagram and the like of the battlefield situation diagram are established according to battlefield needs.
In order to achieve the purpose, the invention adopts the technical scheme that:
an unmanned aerial vehicle adaptive navigation system based on intelligent perception, comprising:
the environment and target intelligent sensing module is used for acquiring sensing information of a ground environment and a target, exchanging information with a remote end and sensing the environment and the target by intelligent reasoning to obtain a battlefield situation sensing map;
the self-adaptive navigation module is used for planning a self-adaptive navigation path and selecting an optimal navigation path according to the minimum criterion of natural risk and war risk on the basis of a battlefield situation perception map;
and the data display module is used for displaying real-time detection data including the radar and SAR sensors, the risk coefficient and the path distance of each feasible path and the current optimal navigation path of the unmanned aerial vehicle.
The environment and target intelligent sensing module acquires sensing information of a ground environment and a target by using a radar, an optical sensor and an infrared sensor which are installed on an unmanned aerial vehicle based on battlefield environment, important nodes and feasible path information, exchanges sensing information with a military supply station of one party and a maritime base through a data chain, and senses the environment and the target by using intelligent reasoning; the data display module comprises a detection data module, a single-path risk and distance module and a current optimal path display module; the detection data module detects the target category and the target position in real time by utilizing sensors including radar and SAR sensors, a sensor data list is given in a table, the sensor data list comprises enemy aircraft warships, our aircraft warships and longitude and latitude heights of unmanned planes, and the detection probability and the false alarm probability of the enemy aircraft warships and the my aircraft warships are displayed at the same time; the single-path risk and distance module displays the risk coefficient and the distance of the single path in real time; and the current optimal path display module displays risk evaluation of all feasible paths at the current moment, wherein the risk evaluation consists of a total distance and a total risk, and simultaneously displays the optimal path.
The environment and target intelligent perception module comprises a battlefield situation perception module and a navigation path vector diagram planning module; the battlefield situation perception module is used for producing a battlefield environment, and acquiring battlefield information including defense arrangement of both enemies and my parties in a battlefield and activity targets of both enemies and my parties by utilizing an airborne radar and optical imaging to generate a battlefield situation perception map; the navigation path vector diagram planning module is used for selecting all possible passing nodes from the starting node to the destination node according to the battlefield situation perception result, preliminarily evaluating the possibility of passing each node and planning all possible navigation path vectors.
The self-adaptive navigation module comprises an initial optimal path planning module and a self-adaptive path transformation module; the initial optimal path planning module carries out quantitative processing on the risk sum to generate an initial optimal path according to natural risks and war risks, wherein the natural risks comprise natural barriers and weather conditions of the node, and the war risks comprise the attack degree of an enemy and the detection possibility of the enemy; and the self-adaptive path transformation module performs self-adaptive adjustment according to the function sensing result and selects the optimal path at the current moment.
The invention also provides an unmanned aerial vehicle self-adaptive navigation method based on intelligent perception, which comprises the following steps:
environment and target intelligent perception: acquiring perception information of a ground environment and a target, exchanging information with a remote end, and obtaining a battlefield situation perception map comprising important nodes of both the enemy and the my, and motion tracks and intentions of active targets by using an intelligent reasoning perception environment and the target, and generating all possible paths from an initial node to a target node of one party on the basis;
self-adaptive navigation path selection: on the basis of a battlefield situation perception map, self-adaptive navigation path planning is carried out according to the minimum criterion of natural risk and war risk, and an optimal navigation path is selected.
The environment and target intelligent perception step comprises the following steps:
target identification and tracking based on multi-source heterogeneous information fusion: extracting effective characteristic information with different types from various sensing information, identifying a target and an environment by utilizing effective fusion of the characteristic information, and effectively tracking the target;
target intent inference: the intention is clarified according to the results of target recognition and tracking;
and (3) generating a battlefield situation perception map: on the basis of target identification and tracking and intention inference, generating a battlefield situation perception map by utilizing geographic information resources and perception information;
intelligent perception of deep learning and evidence reasoning: and establishing an application framework suitable for feature extraction by taking the actual problem as a background, driving a learning process by using actual data, and verifying the reasonability or correctness of the feature extraction result by using real data.
In the target identification and tracking step based on multi-source heterogeneous information fusion, the characteristic attribute of the sensing information is analyzed, the unified description of the perception information is realized from a characteristic level and a decision level by utilizing a random set theory, and necessary preparation is provided for information fusion;
in the target intention deduction step, planning and predicting a combat mission, and making an auxiliary decision on accurate attack of an enemy time-sensitive target;
in the intelligent sensing step of deep learning and evidence reasoning, the system operation result is collected for a long time, the system learning result and the actual detection result are continuously collected in the long-term operation process of the system, and the system learning result is continuously verified by using the actual detection result.
In the target identification and tracking step based on multi-source heterogeneous information fusion, the perception information source comprises: radar information, optical imaging information, acoustic detection sensor information, electronic reconnaissance information, wide-range imaging satellite information and technical reconnaissance information; on the aspect of feature level information fusion, feature space evaluation is realized by using a random set theory and a generalized rough set theory, the status and the effect of the features with clearer class distribution are highlighted, the influence of the features with fuzzy class distribution is weakened, and a unified feature space is effectively constructed;
in the target intention inference step, functions of threat assessment and auxiliary decision are realized through situation analysis, threat estimation, tactical decision, striking effect assessment and man-machine interface management, on the basis of target identification and tracking of networked information fusion, an intention inference method for target intelligent information processing is established to form an intention inference system based on knowledge base trajectory classification, when target tracking is carried out, tracks obtained through tracking are processed, track segment estimation which is rough compared with actual tracks but has clearer concept is achieved, and the intention inference system is used for target intention inference;
in the battlefield situation awareness graph generating step, marking a vector set from all initial nodes to target nodes on the battlefield situation awareness graph, wherein each vector is a directional connecting line for linking one possible node to the next possible node;
in the intelligent sensing step of deep learning and evidence reasoning, a detection result in system operation is used as a basis for supervised learning to judge the accuracy and precision of a learning result; continuously correcting the characteristic relation table according to the verification result, and establishing an application framework suitable for characteristic extraction by taking the actual problem as the background; driving a learning process by using actual data, and verifying the reasonability or correctness of the feature extraction result by using real data;
in the step of selecting the self-adaptive navigation path, natural risks comprise natural obstacles and weather conditions of the nodes, war risks comprise the degree of attack of an enemy, the possibility of monitoring by the enemy and the like, the sum of the risks is subjected to quantitative processing, the length of each vector mainly comprises the traveling distance between the vector nodes, and the sum of the lengths of the vectors is also subjected to quantitative processing.
The transformation criteria for the adaptive navigation path planning are as follows:
assuming that the unmanned aerial vehicle has moved to the ith node at present, the path vector linked with the ith node is called as the ith path vector, and is expressed as:
Figure BDA0002349217400000051
wherein p isif，pieRespectively, the ith path vector siAnd a start node and a stop node of, and (p)ifx,pify)，(piex,piey) Are each pifAnd pieCoordinate values on the two-dimensional war zone perception map, set up enemy approach siOne moving object of is tjThe coordinate value at the current time is (q)jx,qjy) Then enemy target tjFor path vector siThe risk of (a) is:
Figure BDA0002349217400000052
wherein A (t)j) Is tjIs a function of the type of object, and is set in the presentation system to:
Figure BDA0002349217400000053
τ(tj) Is tjIs also a function of the type of object, set in the presentation system to:
Figure BDA0002349217400000054
α(tj) Is a target tjThe intensity of the risk caused by the intention of (1) is set in the presentation system as:
Figure BDA0002349217400000055
maxd(tj,si) Representing a target tjTo path vector siDistance of all points is maximum, and mind (t)j,si) Representing a target tjMinimum distance to all points of the path vector; thus, all enemy target-to-path vectors siThe sum of the risks of (a) is:
R(si)＝∑jR(tj,si)
thus, the slave path vector siInitially, the path with the least risk is selected as:
Figure BDA0002349217400000061
where m is the slave path vector siStarting to the total number of feasible paths of the destination node, wherein l is one path from {1,2, …, m }; r(s)kl) Selecting the kl path vector s of the first feasible path in the futureklThe resulting risk, summed up is the total risk of selecting the l-th feasible path,
Figure BDA0002349217400000062
i.e. slave vectorsiThe first with the least risk of initial selection
Figure BDA0002349217400000063
The total path vector sum of the feasible paths;
path vector siThe path length of (a) is:
d(si)＝[(piex-pifx)2+(piey-pify)2]1/2
select the first
Figure BDA0002349217400000064
More than one feasible path is selected, and the path with the shortest path length is selected as the optimal path finally selected according to the shortest path criterion
Figure BDA0002349217400000065
After selecting the optimal navigation path, the method further comprises the following steps:
optimizing, scheduling and fusing multi-navigation source information: and performing optimized scheduling and fusion on the satellite navigation information and navigation source information including the satellite navigation information, inertial navigation information and astronomical navigation information to obtain an optimal navigation instruction, and performing intelligent navigation according to the navigation instruction.
The invention provides a method theory and a demonstration system for unmanned aerial vehicle self-adaptive navigation transformation in military battlefield environment, which are more targeted compared with the existing engineering navigation scheme and provide a specific scheme for unmanned system combat.
Drawings
FIG. 1 is a schematic diagram of the system of the present invention.
Fig. 2 is a flow chart of the system usage of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention are clearly and completely described below, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
As shown in fig. 1, the present invention provides an adaptive navigation system for an unmanned aerial vehicle based on intelligent sensing, which implements adaptive navigation based on risk coefficients and distance coefficients obtained by intelligent sensing, and includes:
the environment and target intelligent sensing module is used for giving out information such as battlefield environment, important nodes, feasible paths and the like, then acquiring battlefield sensing information of ground environment and targets by utilizing sensors such as radars, optics, infrared and the like installed on an unmanned aerial vehicle, preliminarily detecting and sensing moving targets of both sides of the battlefield, exchanging sensing information with remote centers such as military supply stations of our party and offshore bases through data chains, sensing the environment and the targets by utilizing intelligent reasoning, selecting all nodes which are possibly passed by a starting node to a target node according to sensing results, obtaining a battlefield situation perception map, and expressing the battlefield situation perception map as moving targets such as aircraft warships of both sides of the enemy and the like and selected intermediate nodes.
The system specifically comprises a battlefield situation perception module and a navigation path vector diagram planning module; the battlefield situation perception module is used for producing a battlefield environment, battlefield information including defense arrangement of both enemies and my parties in a battlefield area, moving targets of both enemies and my parties and the like is obtained through airborne radar and optical imaging to generate a battlefield situation perception map, and battlefield situations and important nodes are represented in the map; the navigation path vector diagram planning module is used for selecting all possible passing nodes from the starting node to the destination node according to the battlefield situation perception result, preliminarily evaluating the possibility of passing each node and planning all possible navigation path vectors. After selecting the intermediate nodes, the first step is to plan all possible navigation path vectors (excluding vectors with natural barriers and important risks among the nodes), and display a navigation possible path vector diagram.
The self-adaptive navigation module evaluates the risks of all possible path vectors according to situation perception information on the basis of a battlefield situation perception map, carries out self-adaptive navigation path planning according to the minimum criterion of natural risk and war risk and selects an optimal navigation path according to the natural risk and war risk, and adaptively adjusts the driving path along with the change of the battlefield situation in the process of navigation.
The system specifically comprises an initial optimal path planning module and a self-adaptive path transformation module; the initial optimal path planning module carries out quantitative processing on the risk sum according to natural risks and war risks, wherein the natural risks comprise natural barriers, weather conditions and the like of the node, the war risks comprise the degree of attack of an enemy, the detection possibility of the enemy and the like, and the initial optimal path is generated; an optimal path obtained by initial path planning cannot be guaranteed to be kept unchanged all the time, and self-adaptive adjustment is needed in the flight process according to the intelligent sensing result. And the self-adaptive path transformation module performs self-adaptive adjustment according to the function sensing result and selects the optimal path at the current moment.
And the data display module is used for displaying real-time detection data (such as target types, target positions and the like) including radar and SAR sensors and the like, the risk coefficient and the path distance of each feasible path and the current optimal navigation path of the unmanned aerial vehicle.
The system specifically comprises a detection data module, a single-path risk and distance module and a current optimal path display module; the detection data module utilizes sensors including radar sensors, SAR sensors and the like to detect the types and the positions of targets in real time, a sensor data list is given in a table, data including enemy aircraft warships, our aircraft warships, unmanned planes about to fly by one party, the longitude and latitude heights of the unmanned planes and the like are displayed, the detection probability and the false alarm probability are displayed at the same time, and the display form is shown in table 1.
TABLE 1 Protect data Module display results
Target Time of detection Longitude (G) Dimension (d) of Height Object classes Probability of detection Probability of false alarm
Blue warship A 15:16:40 123.58 25.63 0.00 Battleship 0.90 0.15
Blue warship B 15:16:40 124.04 24.95 0.00 Battleship 0.90 0.13
Blue warship C 15:16:40 124.91 24.33 0.00 Battleship 0.90 0.15
Lanfang warplane A 15:16:40 122.66 25.61 10000.00 Warplane 0.90 0.11
Lanfang warplane B 15:16:40 125.44 25.59 10000.00 Warplane 0.90 0.12
Red warship A 15:16:40 122.33 25.49 0.00 Battleship 0.90 0.13
Red warship B 15:16:40 123.46 24.45 0.00 Battleship 0.90 0.16
Red square warplane A 15:16:40 121.87 26.89 10000.00 Warplane 0.90 0.16
Unmanned plane 15:16:40 120.82 27.00 8000.00 Unmanned plane 0.90 0.16
The single-path risk and distance module displays the risk coefficient and the distance of the single path in real time; and the current optimal path display module displays risk evaluation of all feasible paths at the current moment, the risk evaluation consists of a total distance and a total risk, and the optimal paths are displayed at the same time in a display form shown in table 2.
Namely, the data display module displays the feasible path risk and the optimal path selection process, including the risk and distance of each path vector, and the total risk and total path length of each feasible path, and provides a basis for the optimal path selection or the adaptive adjustment of the optimal path in the flight process.
TABLE 2 display results of current optimal path display module
Optimal navigation path Current feasible route Total distance Total risk Risk assessment
1 0->1->5->9 1238.63 23 168.88
2 0->1->8->4->6->9 1524.23 9 190.83
3 0->1->8->4->7->9 1424.13 4 174.42
4 0->1->8->5->9 1294.84 21 173.86
5 0->1->8->7->9 1247.19 4 153.18
6 0->2->4->6->9 1265.74 9 159.81
7 0->2->4->7->9 1165.64 4 143.40
8 0->2->8->7->9 1159.54 4 142.67
The invention also provides an unmanned aerial vehicle self-adaptive navigation method based on intelligent perception, namely, firstly, a situation awareness graph is generated according to the selected scene, the situation awareness graph is matched with a corresponding feasible path, and an initial optimal driving route is planned. And secondly, changing the optimal feasible path in real time according to the change of the situation awareness graph and the combination of the minimum risk criterion and the shortest path criterion. The method specifically comprises the following steps:
environment and target intelligent perception: acquiring perception information of a ground environment and a target, exchanging information with a remote end, and obtaining a battlefield situation perception graph by using intelligent reasoning perception environment and the target, wherein all possible paths from an initial node to a target node of one party are generated on the basis that the battlefield situation perception graph comprises important nodes (key fixed points) of the two parties of the enemy and the my and the motion trail and intention of a movable target;
self-adaptive navigation path selection: on the basis of a battlefield situation perception map, self-adaptive navigation path planning is carried out according to the minimum criterion of natural risk and war risk, and an optimal navigation path is selected.
Wherein, the environment and target intelligent perception specifically comprises:
target identification and tracking based on multi-source heterogeneous information fusion: effective characteristic information with different types is extracted from various sensing information, a large number of heterogeneous characteristics exist in the battlefield sensing information, the characteristic attributes of the sensing information are analyzed, and unified description of the sensing information is realized from a characteristic level and a decision level by utilizing a random set theory, so that necessary preparation is provided for information fusion. The target and the environment are identified by utilizing the effective fusion of the characteristic information, and the target is effectively tracked to form an important component of situation perception. Wherein, its perception information source includes: radar information, optical imaging information, acoustic detection sensor information, electronic reconnaissance information, wide-range imaging satellite information and technical reconnaissance information; in the aspect of feature level information fusion, feature space evaluation can be realized by using a random set theory and a generalized rough set theory, the status and the effect of the features with clearer class distribution are highlighted, the influence of the features with fuzzy class distribution is weakened, and a unified feature space is effectively constructed.
The optimized scheduling and fusion of the multi-navigation source information specifically comprises the following steps:
step 1: differential preprocessing of multi-navigation source data: data used for navigation fusion estimation should have complementarity, so the primary problem after obtaining the navigation data is to measure the difference, analyze the redundant or complementary characteristics of various measurement data in space or time, and comprehensively consider the error formation mechanism, error transmission, correlation, update rate and the like of the measurement data to determine the availability of the data. Common methods of measuring dissimilarity are: q statistical method, correlation coefficient method, inconsistency measure method, double error method, new difference measure method based on sample information, and the like. The specific method is determined according to the type of the selected sensor. The resulting data type and sector are available.
Step 2: preprocessing the space-time consistency of the multi-navigation source data: the problem of time-space consistency refers to the problem of time inconsistency caused by different sampling rates and time starting points when a plurality of sensors measure the position, the speed and the like of a forwarding platform; and spatial inconsistency due to errors such as measurement bias when converting to a common coordinate system. The so-called temporal registration problem is to align the data to be used for estimating the fusion at the same point in time. The spatial registration problem is a process of estimating and compensating the systematic deviation of each sensor so that the error is as small as possible. Because the available navigation information for adaptive navigation based on intelligent perception may be of various types, the coordinate systems adopted by the information sources are very different, and the error levels are completely different, the preprocessing of the information of multiple navigation sources is very necessary. Different coordinate systems can be selected according to the positioning purpose, and all navigation measurement information is converted into measurement in the coordinate system. And then preprocessing the difference data of multiple navigation sources, processing the space-time consistency and the like, and finally performing fusion estimation.
And step 3: and (4) application of an optimal fusion algorithm. The inertial navigation information is information for measuring acceleration, speed and the like (inertial quantity) of the carrier. The inertial navigation system outputs navigation parameters of the carrier, such as instantaneous speed, acceleration, attitude position and the like. The general inertial navigation system comprises a gyroscope, an accelerometer, a platform structure, a circuit, a computer part and the like. The position and the heading of the carrier can be calculated by measuring the high angle and the azimuth angle of the celestial body relative to the reference datum of the aircraft through astronomical navigation. The astronomical conductor system is an autonomous system, does not radiate electromagnetic waves outwards, has good concealment, high orientation and positioning precision and unrelated positioning error with time, and utilizes inertial navigation and a satellite-sensitive sensor on a platform to carry out combined positioning and attitude determination.
Target intent inference: according to the results of target identification and tracking, the intention is made clear so as to realize situation evaluation and auxiliary decision making; the method is mainly used for planning and predicting combat missions and making auxiliary decisions on accurate attack of enemy time-sensitive targets. Specifically, the functions of threat assessment and aid decision making are realized through situation analysis, threat estimation, tactical decision making, attack effect assessment and man-machine interface management, wherein target intention inference on enemies is an important component of situation awareness. On the basis of target identification and tracking of networked information fusion, an intention inference method for target intelligent information processing is established to form an intention inference system based on knowledge base trajectory classification, and when target tracking is carried out, tracks obtained by tracking are processed to form track segment estimation which is rough compared with actual tracks but has a clearer concept and is used for intention inference of targets.
And (3) generating a battlefield situation perception map: on the basis of target recognition and tracking and intention inference, a battlefield situation perception graph is generated by utilizing geographic information resources and perception information, the battlefield situation perception graph comprises a possible path vector graph of intelligent navigation, namely, a vector set from all starting nodes to destination nodes is marked on the battlefield situation perception graph, and each vector is a directional connecting line for linking one possible node to the next possible node. Because the possible nodes for navigation are limited in a battlefield, the vectors in the possible path vector diagram are also limited.
Intelligent perception of deep learning and evidence reasoning: the method takes a detection result in the operation of the system as a basis for supervised learning to judge the accuracy and precision of a learning result; continuously correcting the characteristic relation table according to the verification result, and establishing an application framework suitable for characteristic extraction by taking the actual problem as the background; the actual data is used for driving the learning process, the reasonability or the correctness of the feature extraction result is verified by the real data, and the learning method is modified to achieve a practical degree. By collecting the system operation result for a long time, the system learning result and the actual detection result can be continuously collected in the long-term operation process of the system, and the system learning result can be continuously verified by using the actual detection result.
In the self-adaptive navigation path selection step, natural risks comprise natural obstacles and weather conditions of the nodes, war risks comprise the degree of attack of an enemy, the possibility of monitoring by the enemy and the like, the sum of the risks is subjected to quantitative processing, the length of each vector mainly comprises the traveling distance between the vector nodes, and the sum of the lengths of the vectors is also subjected to quantitative processing, so that an initial optimal path is obtained; and the optimization calculation adopts a standard method to obtain the optimal navigation feasible path. Namely, an optimal path obtained by initial path planning cannot be guaranteed to be kept unchanged all the time, and self-adaptive adjustment needs to be performed according to an intelligent sensing result in the flight process. The transformation criteria for adaptive navigation path planning are as follows:
assuming that the unmanned aerial vehicle has moved to the ith node at present, the path vector linked with the ith node is called as the ith path vector, and is expressed as:
Figure BDA0002349217400000121
wherein p isif，pieRespectively, the ith path vector siAnd a start node and a stop node of, and (p)ifx,pify)，(piex,piey) Are each pifAnd pieCoordinate values on the two-dimensional battlefield perception map. Let enemy approach siOne moving object of is tjThe coordinate value at the current time is (q)jx,qjy),. Then the enemy target tjFor path vector siThe risk of (a) is:
Figure BDA0002349217400000122
wherein A (t)j) Is tjIs a function of the type of object, and is set in the presentation system to:
Figure BDA0002349217400000123
τ(tj) Is tjRisk decline ofThe subtraction factor, which is also a function of the target type, is set in the presentation system to:
Figure BDA0002349217400000124
α(tj) Is a target tjThe intensity of the risk caused by the intention of (1) is set in the presentation system as:
Figure BDA0002349217400000125
max d(tj,si) Representing a target tjTo path vector siDistance maximum of all points, and min d (t)j,si) Representing a target tjTo path vector siThe distance of all points is the minimum. Thus, all enemy target-to-path vectors siThe sum of the risks of (a) is:
R(si)＝∑jR(tj,si)
thus, the slave path siInitially, the path with the least risk is selected as:
Figure BDA0002349217400000131
where m is the slave vector siStarting to the total number of feasible paths of the destination node, wherein l is one path from {1,2, …, m }; r(s)kl) Selecting the kl path vector s of the first feasible path in the futureklThe resulting risk, summed up is the total risk of selecting the l-th feasible path,
Figure BDA0002349217400000132
is simply the slave vector siThe first with the least risk of initial selection
Figure BDA0002349217400000133
The total path vector sum of the feasible paths.
Path vector siThe path length of (a) is:
d(si)＝[(piex-pifx)2+(piey-pify)2]1/2
select the first
Figure BDA0002349217400000134
More than one feasible path is selected, and the path with the shortest path length is selected as the optimal path finally selected according to the shortest path criterion
Figure BDA0002349217400000135
After selecting the optimal navigation path, the method may further include:
optimizing, scheduling and fusing multi-navigation source information: and performing optimized scheduling and fusion on the satellite navigation information and various navigation source information including the satellite navigation information, the inertial navigation information and the astronomical navigation information to obtain an optimal navigation instruction, and performing intelligent navigation according to the navigation instruction. When multi-navigation source information is utilized, a multi-source information fusion method based on a biological perception mechanism can be established, and a new fusion method is established by utilizing a random theory.
As shown in fig. 2, the specific using process of the present invention includes the following steps:
step 1: opening a demonstration system application program and entering an unmanned aerial vehicle self-adaptive navigation demonstration system starting interface
Step 2: clicking the upper left corner of the page to select a scene, and selecting any scene.
And step 3: clicking the next step to enter a task background interface, and showing the task background.
And 4, step 4: clicking the next step, displaying a map of the Taiwan strait in China, and showing a starting point and an end point of the unmanned aerial vehicle of the same party and longitude and latitude high coordinates of the starting point and the end point on the map
And 5, clicking the next step, and displaying the intermediate nodes of the possible paths of the unmanned aerial vehicle and longitude and latitude coordinates of the nodes on an interface.
Step 6: clicking on the next step, the interface displays the possible paths between all nodes.
And 7: clicking the next step, displaying an enemy target, and displaying the name of the target and longitude and latitude height coordinates of the position of the target above the target.
And 8: clicking the next step, displaying the target of the party, and displaying the name of the target and longitude and latitude height coordinates of the position of the target above the target.
And step 9: and (4) clicking a display button in a menu bar to select RADAR or SAR and popping up a RADAR data table and a SAR data table in combination with the table 1. The target category, longitude and latitude height and detection probability detected by the sensor are displayed in the table.
Step 10: and clicking the operation button, starting the self-adaptive navigation process by the unmanned aerial vehicle, clicking the operation button again, and stopping the unmanned aerial vehicle from moving.
Step 11: clicking a display button in a menu bar, selecting path risk evaluation, and popping up a path risk evaluation form; and selecting the optimal path and popping up an optimal path selection menu by combining the table 2. And the path risk evaluation table displays the risk coefficient and the distance of each path in the animation running process in real time. And the optimal path selection table displays the minimum risk path of the unmanned aerial vehicle in real time.

Claims (9)

1. An unmanned aerial vehicle self-adaptation navigation based on intelligence perception, its characterized in that includes:
the environment and target intelligent sensing module is used for acquiring sensing information of a ground environment and a target, exchanging information with a remote end and sensing the environment and the target by intelligent reasoning to obtain a battlefield situation sensing map;
the self-adaptive navigation module is used for planning a self-adaptive navigation path and selecting an optimal navigation path according to the minimum criterion of natural risk and war risk on the basis of a battlefield situation perception map;
the data display module is used for displaying real-time detection data including radar and SAR sensors, a risk coefficient and a path distance of each feasible path and a current optimal navigation path of the unmanned aerial vehicle;
the transformation criteria for the adaptive navigation path planning are as follows:
assuming that the unmanned aerial vehicle has moved to the ith node at present, the path vector linked with the ith node is called as the ith path vector, and is expressed as:
Figure FDA0003024884610000011
wherein p isif，pieRespectively, the ith path vector siAnd a start node and a stop node of, and (p)ifx,pify)，(piex,piey) Are each pifAnd pieCoordinate values on the two-dimensional war zone perception map, set up enemy approach siOne moving object of is tjThe coordinate value at the current time is (q)jx,qjy) Then enemy target tjFor path vector siThe risk of (a) is:
Figure FDA0003024884610000012
wherein A (t)j) Is tjIs a function of the type of object, and is set in the presentation system to:
Figure FDA0003024884610000013
τ(tj) Is tjIs also a function of the type of object, set in the presentation system to:
Figure FDA0003024884610000014
α(tj) Is a target tjThe intensity of the risk caused by the intention of (1) is set in the presentation system as:
Figure FDA0003024884610000021
maxd(tj,si) Representing a target tjTo path vector siDistance of all points is maximum, and mind (t)j,si) Representing a target tjMinimum distance to all points of the path vector; thus, all enemy target-to-path vectors siThe sum of the risks of (a) is:
R(si)＝∑jR(tj,si)
thus, the slave path vector siInitially, the path with the least risk is selected as:
Figure FDA0003024884610000022
where m is the slave path vector siStarting to the total number of feasible paths of the destination node, wherein l is one path from {1,2, …, m }; r(s)kl) Selecting the kl path vector s of the first feasible path in the futureklThe resulting risk, summed up is the total risk of selecting the l-th feasible path,
Figure FDA0003024884610000026
i.e. the slave vector siThe first with the least risk of initial selection
Figure FDA0003024884610000023
The total path vector sum of the feasible paths;
path vector siThe path length of (a) is:
d(si)＝[(piex-pifx)2+(piey-pify)2]1/2
select the first
Figure FDA0003024884610000024
More than one feasible path is selected according to the shortest path criterionThe shortest one is the optimal path finally selected
Figure FDA0003024884610000025
2. The unmanned aerial vehicle adaptive navigation system based on intelligent sensing of claim 1, wherein the environment and target intelligent sensing module is based on battlefield environment, important nodes and feasible path information, and utilizes radar, optical and infrared sensors installed on the unmanned aerial vehicle to acquire sensing information of ground environment and targets, and simultaneously exchanges sensing information with the military supply station of our party and the offshore base through a data chain, and then senses the environment and the targets by intelligent reasoning; the data display module comprises a detection data module, a single-path risk and distance module and a current optimal path display module; the detection data module detects the target category and the target position in real time by utilizing sensors including radar and SAR sensors, a sensor data list is given in a table, the sensor data list comprises enemy aircraft warships, our aircraft warships and longitude and latitude heights of unmanned planes, and the detection probability and the false alarm probability of the enemy aircraft warships and the my aircraft warships are displayed at the same time; the single-path risk and distance module displays the risk coefficient and the distance of the single path in real time; and the current optimal path display module displays risk evaluation of all feasible paths at the current moment, wherein the risk evaluation consists of a total distance and a total risk, and simultaneously displays the optimal path.
3. The unmanned aerial vehicle adaptive navigation system based on intelligent perception according to claim 1 or 2, wherein the environment and target intelligent perception module comprises a battlefield situation perception module and a navigation path vector diagram planning module; the battlefield situation perception module is used for producing a battlefield environment, and acquiring battlefield information including defense arrangement of both enemies and my parties in a battlefield and activity targets of both enemies and my parties by utilizing an airborne radar and optical imaging to generate a battlefield situation perception map; the navigation path vector diagram planning module is used for selecting all possible passing nodes from the starting node to the destination node according to the battlefield situation perception result, preliminarily evaluating the possibility of passing each node and planning all possible navigation path vectors.
4. The unmanned aerial vehicle adaptive navigation system based on intelligent perception of claim 3, wherein the adaptive navigation module comprises an initial optimal path planning module and an adaptive path transformation module; the initial optimal path planning module carries out quantitative processing on the risk sum to generate an initial optimal path according to natural risks and war risks, wherein the natural risks comprise natural barriers and weather conditions of the node, and the war risks comprise the attack degree of an enemy and the detection possibility of the enemy; and the self-adaptive path transformation module performs self-adaptive adjustment according to the function sensing result and selects the optimal path at the current moment.
5. An unmanned aerial vehicle self-adaptive navigation method based on intelligent perception is characterized by comprising the following steps:
environment and target intelligent perception: acquiring perception information of a ground environment and a target, exchanging information with a remote end, and obtaining a battlefield situation perception map comprising important nodes of both the enemy and the my, and motion tracks and intentions of active targets by using an intelligent reasoning perception environment and the target, and generating all possible paths from an initial node to a target node of one party on the basis;
self-adaptive navigation path selection: on the basis of a battlefield situation awareness graph, self-adaptive navigation path planning is carried out according to the minimum criterion of natural risk and war risk, and an optimal navigation path is selected;
wherein the transformation criteria for the adaptive navigation path planning are as follows:
assuming that the unmanned aerial vehicle has moved to the ith node at present, the path vector linked with the ith node is called as the ith path vector, and is expressed as:
Figure FDA0003024884610000041
wherein p isif，pieRespectively, the ith path vector siAnd a start node and a stop node of, and (p)ifx,pify)，(piex,piey) Are each pifAnd pieCoordinate values on the two-dimensional war zone perception map, set up enemy approach siOne moving object of is tjThe coordinate value at the current time is (q)jx,qjy) Then enemy target tjFor path vector siThe risk of (a) is:
Figure FDA0003024884610000042
wherein A (t)j) Is tjIs a function of the type of object, and is set in the presentation system to:
Figure FDA0003024884610000043
τ(tj) Is tjIs also a function of the type of object, set in the presentation system to:
Figure FDA0003024884610000044
α(tj) Is a target tjThe intensity of the risk caused by the intention of (1) is set in the presentation system as:
Figure FDA0003024884610000045
maxd(tj,si) Representing a target tjTo path vector siDistance of all points is maximum, and mind (t)j,si) Representing a target tjMinimum distance to all points of the path vector; thus, all enemy target-to-path vectors siThe sum of the risks of (a) is:
R(si)＝∑jR(tj,si)
thus, the slave path vector siInitially, the path with the least risk is selected as:
Figure FDA0003024884610000046
where m is the slave path vector siStarting to the total number of feasible paths of the destination node, wherein l is one path from {1,2, …, m }; r(s)kl) Selecting the kl path vector s of the first feasible path in the futureklThe resulting risk, summed up is the total risk of selecting the l-th feasible path,
Figure FDA0003024884610000047
i.e. the slave vector siThe first with the least risk of initial selection
Figure FDA0003024884610000051
The total path vector sum of the feasible paths;
path vector siThe path length of (a) is:
d(si)＝[(piex-pifx)2+(piey-pify)2]1/2
select the first
Figure FDA0003024884610000052
More than one feasible path is selected, and the path with the shortest path length is selected as the optimal path finally selected according to the shortest path criterion
Figure FDA0003024884610000053
6. The unmanned aerial vehicle adaptive navigation method based on intelligent perception according to claim 5, wherein the environment and target intelligent perception step comprises:
target identification and tracking based on multi-source heterogeneous information fusion: extracting effective characteristic information with different types from various sensing information, identifying a target and an environment by utilizing effective fusion of the characteristic information, and effectively tracking the target;
target intent inference: the intention is clarified according to the results of target recognition and tracking;
and (3) generating a battlefield situation perception map: on the basis of target identification and tracking and intention inference, generating a battlefield situation perception map by utilizing geographic information resources and perception information;
intelligent perception of deep learning and evidence reasoning: and establishing an application framework suitable for feature extraction by taking the actual problem as a background, driving a learning process by using actual data, and verifying the reasonability or correctness of the feature extraction result by using real data.
7. The unmanned aerial vehicle adaptive navigation method based on intelligent perception according to claim 6, wherein in the target identification and tracking step based on multi-source heterogeneous information fusion, the characteristic attributes of the sensing information are analyzed, the unified description of the sensing information is realized from a characteristic level and a decision level by utilizing a random set theory, and necessary preparation is provided for information fusion;
in the target intention deduction step, planning and predicting a combat mission, and making an auxiliary decision on accurate attack of an enemy time-sensitive target;
in the intelligent sensing step of deep learning and evidence reasoning, the system operation result is collected for a long time, the system learning result and the actual detection result are continuously collected in the long-term operation process of the system, and the system learning result is continuously verified by using the actual detection result.
8. The unmanned aerial vehicle adaptive navigation method based on intelligent perception according to claim 7, wherein in the target identification and tracking step based on multi-source heterogeneous information fusion, the perception information source comprises: radar information, optical imaging information, acoustic detection sensor information, electronic reconnaissance information, wide-range imaging satellite information and technical reconnaissance information; on the aspect of feature level information fusion, feature space evaluation is realized by using a random set theory and a generalized rough set theory, the status and the effect of the features with clearer class distribution are highlighted, the influence of the features with fuzzy class distribution is weakened, and a unified feature space is effectively constructed;
in the target intention inference step, functions of threat assessment and auxiliary decision are realized through situation analysis, threat estimation, tactical decision, striking effect assessment and man-machine interface management, on the basis of target identification and tracking of networked information fusion, an intention inference method for target intelligent information processing is established to form an intention inference system based on knowledge base trajectory classification, when target tracking is carried out, tracks obtained through tracking are processed, track segment estimation which is rough compared with actual tracks but has clearer concept is achieved, and the intention inference system is used for target intention inference;
in the battlefield situation awareness graph generating step, marking a vector set from all initial nodes to target nodes on the battlefield situation awareness graph, wherein each vector is a directional connecting line for linking one possible node to the next possible node;
in the intelligent sensing step of deep learning and evidence reasoning, a detection result in system operation is used as a basis for supervised learning to judge the accuracy and precision of a learning result; continuously correcting the characteristic relation table according to the verification result, and establishing an application framework suitable for characteristic extraction by taking the actual problem as the background; driving a learning process by using actual data, and verifying the reasonability or correctness of the feature extraction result by using real data;
in the step of selecting the self-adaptive navigation path, natural risks comprise natural barriers and weather conditions of the nodes, war risks comprise the degree of attack of an enemy and the possibility of monitoring of the enemy, the sum of the risks is subjected to quantitative processing, the length of each vector mainly comprises the traveling distance between the nodes, and the sum of the lengths of the vectors is also subjected to quantitative processing.
9. The unmanned aerial vehicle adaptive navigation method based on intelligent perception according to claim 6, wherein after selecting the optimal navigation path, the method further comprises:
optimizing, scheduling and fusing multi-navigation source information: and performing optimized scheduling and fusion on the satellite navigation information and navigation source information including the satellite navigation information, inertial navigation information and astronomical navigation information to obtain an optimal navigation instruction, and performing intelligent navigation according to the navigation instruction.
CN201911408053.6A 2019-12-31 2019-12-31 Unmanned aerial vehicle self-adaptive navigation system and method based on intelligent sensing Active CN111044052B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201911408053.6A CN111044052B (en) 2019-12-31 2019-12-31 Unmanned aerial vehicle self-adaptive navigation system and method based on intelligent sensing

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201911408053.6A CN111044052B (en) 2019-12-31 2019-12-31 Unmanned aerial vehicle self-adaptive navigation system and method based on intelligent sensing

Publications (2)

Publication Number Publication Date
CN111044052A CN111044052A (en) 2020-04-21
CN111044052B true CN111044052B (en) 2021-07-06

Family

ID=70242325

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201911408053.6A Active CN111044052B (en) 2019-12-31 2019-12-31 Unmanned aerial vehicle self-adaptive navigation system and method based on intelligent sensing

Country Status (1)

Country Link
CN (1) CN111044052B (en)

Families Citing this family (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP3901859A1 (en) * 2020-04-22 2021-10-27 Volocopter GmbH Method of and system for operating an aircraft for assessing operational risk
CN111829521B (en) * 2020-06-23 2022-05-03 浙江工业大学 Consistent target tracking method based on data driving
CN113568435B (en) * 2021-09-24 2021-12-24 深圳火眼智能有限公司 Unmanned aerial vehicle autonomous flight situation perception trend based analysis method and system
CN115014330A (en) * 2022-06-13 2022-09-06 武汉华中天易星惯科技有限公司 Unmanned platform photoelectric navigation intelligent perception and decision making system
CN115460392A (en) * 2022-08-22 2022-12-09 山东新一代信息产业技术研究院有限公司 Panorama 3D video concatenation system and system carrier based on SOC
CN115906673B (en) * 2023-01-10 2023-11-03 中国人民解放军陆军工程大学 Combat entity behavior model integrated modeling method and system
CN116310689B (en) * 2023-03-24 2023-09-08 中国科学院空天信息创新研究院 Unmanned intelligent multi-mode information fusion and target perception system and operation method
CN117647250B (en) * 2024-01-29 2024-04-30 深圳市爱保护科技有限公司 Navigation method and system based on intelligent bracelet

Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101964019A (en) * 2010-09-10 2011-02-02 北京航空航天大学 Against behavior modeling simulation platform and method based on Agent technology
CN104359473A (en) * 2014-10-24 2015-02-18 南京航空航天大学 Collaborative flight path intelligent planning method for formation flying of unmanned planes under dynamic environment
CN104834317A (en) * 2015-04-24 2015-08-12 华北计算技术研究所 Flying path planning method of unmanned plane capable of intelligently identifying threat types
CN108318032A (en) * 2017-12-21 2018-07-24 上海交通大学 A kind of unmanned aerial vehicle flight path Intelligent planning method considering Attack Defence
CN108427438A (en) * 2018-04-11 2018-08-21 北京木业邦科技有限公司 Flight environment of vehicle detection method, device, electronic equipment and storage medium
CN108759837A (en) * 2018-05-22 2018-11-06 北京和协导航科技有限公司 Unmanned plane multi computer communication combat system and method
CN109405650A (en) * 2018-11-07 2019-03-01 北京和协导航科技有限公司 A kind of unmanned plane integrated combat system based on Man-in-loop control
CN109724472A (en) * 2018-12-11 2019-05-07 四川九洲电器集团有限责任公司 A kind of vehicle-mounted strategic point regional complex prevents anti-system and method
CN110133573A (en) * 2019-04-23 2019-08-16 四川九洲电器集团有限责任公司 A kind of autonomous low latitude unmanned plane system of defense based on the fusion of multielement bar information

Patent Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101964019A (en) * 2010-09-10 2011-02-02 北京航空航天大学 Against behavior modeling simulation platform and method based on Agent technology
CN104359473A (en) * 2014-10-24 2015-02-18 南京航空航天大学 Collaborative flight path intelligent planning method for formation flying of unmanned planes under dynamic environment
CN104834317A (en) * 2015-04-24 2015-08-12 华北计算技术研究所 Flying path planning method of unmanned plane capable of intelligently identifying threat types
CN108318032A (en) * 2017-12-21 2018-07-24 上海交通大学 A kind of unmanned aerial vehicle flight path Intelligent planning method considering Attack Defence
CN108427438A (en) * 2018-04-11 2018-08-21 北京木业邦科技有限公司 Flight environment of vehicle detection method, device, electronic equipment and storage medium
CN108759837A (en) * 2018-05-22 2018-11-06 北京和协导航科技有限公司 Unmanned plane multi computer communication combat system and method
CN109405650A (en) * 2018-11-07 2019-03-01 北京和协导航科技有限公司 A kind of unmanned plane integrated combat system based on Man-in-loop control
CN109724472A (en) * 2018-12-11 2019-05-07 四川九洲电器集团有限责任公司 A kind of vehicle-mounted strategic point regional complex prevents anti-system and method
CN110133573A (en) * 2019-04-23 2019-08-16 四川九洲电器集团有限责任公司 A kind of autonomous low latitude unmanned plane system of defense based on the fusion of multielement bar information

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
A Risk-aware Path Planning Method for Unmanned Aerial Vehicles;Stefano Primatesta等;《2018 International Conference on Unmanned Aircraft Systems(ICUAS)》;20180615;905-913 *
The problem of minimum risk path planning for flying robots in dangerous environments;Andrey V.Savkin等;《Proceedings of the 35th Chinese Control Conference》;20160729;5404-5408 *

Also Published As

Publication number Publication date
CN111044052A (en) 2020-04-21

Similar Documents

Publication Publication Date Title
CN111044052B (en) Unmanned aerial vehicle self-adaptive navigation system and method based on intelligent sensing
US12072705B2 (en) Intelligent decision-making method and system for unmanned surface vehicle
US10249151B2 (en) Canine handler operations positioning system
CN110597264B (en) Unmanned aerial vehicle counter-braking system
Chun et al. Robot surveillance and security
CN104021292B (en) Dim target detection and tracking method based on formation active networking
CN104050368B (en) Error compensation-based group track fine association algorithm under system error
Peršić et al. Online multi-sensor calibration based on moving object tracking
Looney Exploring fusion architecture for a common operational picture
Hall et al. Multisensor data fusion
Fan et al. Path planning for a reconnaissance UAV in uncertain environment
Drake et al. Autonomous control of multiple UAVs for the passive location of radars
Gui Evaluation of unmanned equipment operational effectiveness based on ADC model
Imbiriba et al. Augmented physics-based machine learning for navigation and tracking
US10295662B2 (en) Producing data describing target measurements
CN115456090A (en) Knowledge inference engine-based multi-level composite intent prediction method for fighter
Sun et al. Convergence of S3C empowered distributed cooperative optimization for multi-unmanned surface vehicles
CN114545414A (en) Track management method for unmanned aerial vehicle anti-collision radar
Yang et al. Study on the trajectory positioning and prediction framework of unmanned aerial vehicle based on long short-term memory neural network
Davidovich et al. VISAS--Detecting GPS spoofing attacks against drones by analyzing camera's video stream
Memon et al. Tracking Multiple Autonomous Ground Vehicles using Motion Capture System operating in a Wireless Network
Liu et al. Research of UAV cooperative reconnaissance with self-organization path planning
CN117928559B (en) Unmanned aerial vehicle path planning method under threat avoidance based on reinforcement learning
EP2921878A1 (en) Producing data describing target states
Horovitz et al. UAVDome-Safe UAV Transportation

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant