[go: up one dir, main page]

NL2039176A - Emergency rescue drone rapid three-dimensional trajectory planning device and method - Google Patents

Emergency rescue drone rapid three-dimensional trajectory planning device and method Download PDF

Info

Publication number
NL2039176A
NL2039176A NL2039176A NL2039176A NL2039176A NL 2039176 A NL2039176 A NL 2039176A NL 2039176 A NL2039176 A NL 2039176A NL 2039176 A NL2039176 A NL 2039176A NL 2039176 A NL2039176 A NL 2039176A
Authority
NL
Netherlands
Prior art keywords
module
target
data
trajectory
aggregation
Prior art date
Application number
NL2039176A
Other languages
Dutch (nl)
Other versions
NL2039176B1 (en
Inventor
Li Zhang
Li Jing
Xu Ziyue
Original Assignee
Univ Hubei
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 Univ Hubei filed Critical Univ Hubei
Publication of NL2039176A publication Critical patent/NL2039176A/en
Application granted granted Critical
Publication of NL2039176B1 publication Critical patent/NL2039176B1/en

Links

Classifications

    • GPHYSICS
    • G05CONTROLLING; REGULATING
    • G05DSYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D1/00Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
    • G05D1/60Intended control result
    • G05D1/656Interaction with payloads or external entities
    • G05D1/689Pointing payloads towards fixed or moving targets
    • GPHYSICS
    • G05CONTROLLING; REGULATING
    • G05DSYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D1/00Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
    • G05D1/20Control system inputs
    • G05D1/24Arrangements for determining position or orientation
    • G05D1/243Means capturing signals occurring naturally from the environment, e.g. ambient optical, acoustic, gravitational or magnetic signals
    • GPHYSICS
    • G05CONTROLLING; REGULATING
    • G05DSYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D1/00Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
    • G05D1/20Control system inputs
    • G05D1/24Arrangements for determining position or orientation
    • G05D1/246Arrangements for determining position or orientation using environment maps, e.g. simultaneous localisation and mapping [SLAM]
    • BPERFORMING OPERATIONS; TRANSPORTING
    • B64AIRCRAFT; AVIATION; COSMONAUTICS
    • B64UUNMANNED AERIAL VEHICLES [UAV]; EQUIPMENT THEREFOR
    • B64U2101/00UAVs specially adapted for particular uses or applications
    • B64U2101/55UAVs specially adapted for particular uses or applications for life-saving or rescue operations; for medical use
    • GPHYSICS
    • G05CONTROLLING; REGULATING
    • G05DSYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D2105/00Specific applications of the controlled vehicles
    • G05D2105/55Specific applications of the controlled vehicles for emergency activities, e.g. search and rescue, traffic accidents or fire fighting
    • GPHYSICS
    • G05CONTROLLING; REGULATING
    • G05DSYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D2107/00Specific environments of the controlled vehicles
    • G05D2107/30Off-road
    • G05D2107/36Catastrophic areas
    • GPHYSICS
    • G05CONTROLLING; REGULATING
    • G05DSYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D2109/00Types of controlled vehicles
    • G05D2109/20Aircraft, e.g. drones
    • G05D2109/25Rotorcrafts
    • G05D2109/254Flying platforms, e.g. multicopters
    • GPHYSICS
    • G05CONTROLLING; REGULATING
    • G05DSYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D2111/00Details of signals used for control of position, course, altitude or attitude of land, water, air or space vehicles
    • G05D2111/10Optical signals

Landscapes

  • Engineering & Computer Science (AREA)
  • Aviation & Aerospace Engineering (AREA)
  • Radar, Positioning & Navigation (AREA)
  • Remote Sensing (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Automation & Control Theory (AREA)
  • Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)
  • Alarm Systems (AREA)

Abstract

The present invention discloses an emergency rescue drone rapid threedimensional trajectory planning device and method, comprising a target monitoring module, wherein the target monitoring module consists of an external data receiving module, an external target detection module, a target position conversion module, and a target position output module. In this invention, by removing the coordinate data state field, the packet header is reduced, improving transmission speed and decreasing the response time for drone flight control. The use of a dual-channel module for data transmission enhances communication reliability and data transmission stability. Through method optimization, the invention achieves the simultaneous fulfillment of reducing trajectory cost, dynamically planning paths, and avoiding small obstacles.

Description

EMERGENCY RESCUE DRONE RAPID THREE-DIMENSIONAL TRAJECTORY
PLANNING DEVICE AND METHOD
Technical Field
The present invention relates to the field of drone technology, specifically to an emergency rescue drone rapid three-dimensional trajectory planning device and method.
Background Technology
Significant earthquakes capable of causing extensive damage have drawn attention to the importance of enhancing the efficiency of post-disaster reconstruction and earthquake rescue operations. In this context, drones, due to their unigue advantages, have become essential tools in emergency rescue efforts. Drones are small, easy to operate, capable of flexible route adjustments, and can quickly reach designated locations. Consequently, drones have gradually transitioned from the commercial sector to the field of emergency rescue. However, in earthquake search and rescue environments, which often consist of rubble, broken branches, scattered debris, and potentially collapsing buildings, the post-disaster terrain becomes extremely complex. This necessitates that emergency rescue drones possess the ability to autonomously avoid small, medium, and large obstacles and meet the increased requirements for real-time data transmission. Existing drone trajectory planning methods, however, have certain shortcomings. First, most current drone trajectory planning methods overlook the issue of drone response speed, rendering the calculated path coordinates unsuitable for continuously moving drone positions.
Second, reliable data support is needed for drone trajectory planning, demanding high data transmission stability. Third, existing drone trajectory planning methods often fail to simultaneously fulfill the requirements for reducing trajectory cost, dynamically planning paths, and avoiding small obstacles. Therefore, designing an emergency rescue drone rapid three-dimensional trajectory planning device and method is necessary.
Summary of the Invention
The objective of the present invention is to provide an emergency rescue drone rapid three-dimensional trajectory planning device and method to solve the problems mentioned in the background technology.
To achieve the aforementioned objective, the present invention provides the following technical solution: an emergency rescue drone rapid three-dimensional trajectory planning device comprising a target monitoring module, wherein the target monitoring module comprises an external data receiving module, an external target detection module, a target position conversion module, and a target position output module. The external data receiving module is connected to the external target detection module, which is connected to the target position conversion module. The target position conversion module is connected to the target position output module.
The target position output module is connected to a target receiving module within a controller module, which is connected to the target processing module. The target processing module is connected to a dual-channel module, which is connected to a target sending module. The target sending module is connected to a target reading module within an attitude control module, which is connected to a data aggregation module. The target monitoring module is installed in an industrial control box, which houses a power supply module.
As a further technical solution of the present invention, the controller module comprises the target receiving module, target processing module, dual-channel module, and target sending module, and is installed in the industrial control box.
As a further technical solution of the present invention, the attitude control module comprises the target reading module, data aggregation module, internal sensor receiving module, aggregation output module, trajectory receiving module, and flight adjustment module. The data aggregation module is connected to an internal sensor receiving module, which is connected to an internal sensor output module within an internal sensor module.
As a further technical solution of the present invention, the internal sensor module comprises of the internal sensor output module, acceleration output module, angular velocity output module, geomagnetic direction output module, geographic location output module, accelerometer module, gyroscope module, magnetometer module, and GPS module. The internal sensor output module is connected to the acceleration output module, which is connected to the accelerometer module.
As a further technical solution of the present invention, the internal sensor output module is connected to the angular velocity output module, which is connected to the gyroscope module. The internal sensor output module is connected to the geomagnetic direction output module, which is connected to the magnetometer module.
As a further technical solution of the present invention, the internal sensor output module is connected to the geographic location output module, which is connected to the GPS module.
As a further technical solution of the present invention, the data aggregation module is connected to the aggregation output module, which is connected to an aggregation receiving module within an optimization module. The aggregation receiving module is connected to the aggregation processing module.
As a further technical solution of the present invention, the optimization module comprises the aggregation receiving module, aggregation processing module, trajectory generation module, and trajectory transmission module. The aggregation processing module is connected to the trajectory generation module, which is connected to the trajectory transmission module. The trajectory transmission module is connected to the trajectory receiving module, which is connected to the flight adjustment module.
As a further technical solution of the present invention, the external data receiving module is connected to the external sensor module. The external sensor module is installed in the industrial control box, which houses the attitude control module, internal sensor module, and optimization module.
A rapid three-dimensional trajectory planning method for emergency rescue drones, comprising the following steps: Step 1, data collection and transmission; Step 2, method optimization; Step 3, trajectory generation;
In Step 1, the power supply module powers the entire device, and the external data receiving module collects video data from the external sensor module. The external target detection module identifies and tracks dynamic or static targets in the video, transmitting the detected targets to the target position conversion module. The target position conversion module processes the target data and converts it into target position data, which is then transmitted to the target position output module. The target position output module sends the data to the target receiving module, which forwards it to the target processing module. The target processing module processes the data by removing all state fields from the coordinate data to reduce the packet header. The processed target data is then transmitted to the target sending module via the dual-channel module. The target sending module sends the target data to the target reading module, and the data aggregation module collects the target data.
Simultaneously, the accelerometer module collects acceleration data and transmits it to the acceleration output module, which sends it to the internal sensor output module. The gyroscope module collects angular velocity data and transmits it to the angular velocity output module, which forwards it to the internal sensor output module, The magnetometer module collects geomagnetic direction data and transmits it to the geomagnetic direction output module, which sends it to the internal sensor output module. The GPS module collects geographic location data and transmits it to the geographic location output module, which forwards it to the internal sensor output module. The internal sensor output module transmits the collected acceleration, angular velocity, geomagnetic direction, and geographic location data to the internal sensor receiving module, which forwards the data to the data aggregation module. The data aggregation module aggregates all collected data and transmits it to the aggregation output module, which sends the aggregated data to the aggregation receiving module.
In Step 2, the aggregation receiving module receives the aggregated data and transmits it to the aggregation processing module, where a three-dimensional drone flight map is created using the following function: z(x,y) =sin(y +a) +b ¥sin(x) +c *cos{(d ¥(v2 + x2) +e *cos(y) + f *sm(f *(y2+ x2) + g *cos(y) where z represents the ground elevation at the current position of the drone, and x, y are the X-axis and Y-axis coordinates of the terrain, respectively; parameters 5 a b, c, dy e, f, g are defined based on sensor data to model the terrain; the threat zone's center coordinates and radius are set based on data detected by the target tracking processor and overlaid on the map; considering the drone's acceleration and angular velocity changes, the starting and ending points of the drone's flight are determined; the drone's navigation cost is evaluated using the path length L, height evaluation H, and turning angle evaluation C° with the fitness function F':
F=Wi*L+W:*H+Ws:*C where Wi, Wa Ws are weight coefficients summing to 1; path length 7. is calculated as the sum of Euclidean distances between consecutive points on the path; height evaluation H isthe root mean square of the differences between the heights at each path point and the average height; turning angle evaluation (C assesses the sharpness of turns by calculating the cosine values between vectors of consecutive points; method parameters and individual positions are initialized, with each position representing a set of drone trajectory coordinates; adaptive differential evolution mutation updates individual positions Y (t):
Y(t)=Yu(t) + a* (F(t) —Yos(1)) where YY Yr: are three randomly selected individuals in the population, and a is the adaptive step coefficient, defined as follows: [a *1.1, diversity <0.1 “3 a:*0.9, else the adaptive step coefficient « is dynamically adjusted based on population diversity, calculated every 50 iterations; diversity is defined as the mean ‚4 of the standard deviations o; of the population positions across all dimensions dim: 1 dim diversity = dim 27 based on probability p , the individual positions are updated in segments; when p is less than a random number 7, the position is updated using a sine-based method:
Yi t+) =Y (tO) +n(l+sin(r))* Yi (0), p <r in other cases, the position is updated as follows:
Yi j(t+1)=Y. (6) +n(2r-1)*Y: (0), else an enhanced local search strategy (ELS) is incorporated to refine the search and optimize local individual positions; this is done through a perturbation factor
Le , which identifies parts of the population with significant position changes and performs a secondary update on the global position to generate a local optimal position:
Yig+1) = ra", Yi (0)
Tr where ra is a random number in the range (0,1), f is a tuning parameter, and 7 and { are the maximum and current iteration counts, respectively; the resulting
Yi. i(r) represents the optimal position for each individual, and the fitness value F is calculated; the positions of individuals in the population are converted into drone trajectory coordinates, and the trajectory generation module generates a three- dimensional trajectory, producing the planned drone path;
In Step 3, the three-dimensional trajectory generated by the trajectory generation module is transmitted to the trajectory transmission module, which sends it to the trajectory receiving module, The trajectory receiving module receives the three- dimensional trajectory and immediately forwards it to the flight adjustment module, which adjusts the drone's flight attitude and direction based on each coordinate point, enabling effective obstacle avoidance.
Compared with the existing technology, the beneficial effects of the present invention are as follows: This emergency rescue drone rapid three-dimensional trajectory planning device and method collects video data from the external sensor module through the external data receiving module. The external target detection module identifies and tracks dynamic or static targets in the video, transmitting the detected targets to the target position conversion module. The target position conversion module processes the target data and converts it into target position data, which is then transmitted to the target position output module. The target position output module forwards it to the target receiving module, which sends it to the target processing module. The target processing module processes the data by removing all state fields from the coordinate data, reducing the packet header, which improves the transmission speed of coordinate data and reduces the response time for drone flight control. The processed target data is transmitted to the target sending module via the dual-channel module. The dual-channel module includes an independent backup communication channel, enhancing communication reliability and ensuring stable data transmission even if the main channel fails. The aggregated data is received by the aggregation receiving module and transmitted to the aggregation processing module, where method improvements and optimizations are performed. The method incorporates two individual position update strategies to enhance global search capabilities and improve local search performance. By using an enhancement factor to identify groups with high change rates and updating local optimal solutions, the individual positions are refined, local trajectory coordinates are improved, and the method simultaneously meets the requirements for reducing trajectory cost, dynamically planning paths, and avoiding small obstacles.
Description of the Drawings
FIG.1 shows the overall architecture of the invention;
FIG.2 shows a schematic view of the industrial control box location in the invention;
FIG.3 shows the flowchart of the target monitoring module in the invention;
FIG.4 shows the flowchart of the controller module in the invention;
FIG.5 shows the flowchart of the attitude control module in the invention;
FIG.6 shows the flowchart of the internal sensor module in the invention;
FIG.7 shows the flowchart of the optimization module in the invention;
F1G.8 shows the flowchart of the improved optimization method in the invention;
FIG.9 shows the method flowchart of the invention.
In the figures: 1, industrial control box; 11, power supply module; 12, target monitoring module; 13, controller module; 14, attitude control module; 15, internal sensor module; 16, optimization module; 17, external sensor module; 121, external data receiving module; 122, external target detection module; 123, target position conversion module; 124, target position output module; 131, target receiving module; 132, target processing module; 133, dual-channel module; 134, target sending module; 141, target reading module; 142, data aggregation module; 143, internal sensor receiving module; 144, aggregation output module; 145, trajectory receiving module; 146, flight adjustment module; 151, internal sensor output module; 152, acceleration output module; 153, angular velocity output module; 154, geomagnetic direction output module; 155, geographic location output module; 156, accelerometer module; 157, gyroscope module; 158, magnetometer module; 159, GPS module; 161, aggregation receiving module; 162, aggregation processing module; 163, trajectory generation module; 164, trajectory transmission module.
Detailed Embodiments
The technical solutions of the embodiments of the present invention will be clearly and comprehensively described below in conjunction with the accompanying drawings in the embodiments of the present invention. It is evident that the described embodiments are only part of the embodiments of the present invention, rather than all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without any creative efforts fall within the scope of protection of the present invention.
Please refer to Figures 1 to 8. An embodiment provided by the present invention is as follows: an emergency rescue drone rapid three-dimensional trajectory planning device, which comprises a target monitoring module 12. The target monitoring module 12 consists of an external data receiving module 121, an external target detection module 122, a target position conversion module 123, and a target position output module 124. The external data receiving module 121 is connected to the external target detection module 122, which is connected to the target position conversion module 123. The target position conversion module 123 is connected to the target position output module 124, which is connected to the target receiving module 131 within the controller module 13. The target receiving module 131 is connected to the target processing module 132, which is connected to the dual-channel module 133.
The dual-channel module 133 is connected to the target sending module 134, which is connected to the target reading module 141 within the attitude control module 14.
The target reading module 141 is connected to the data aggregation module 142, The target monitoring module 12 is installed in the industrial control box 1, which also houses the power supply module 11. The controller module 13 consists of the target receiving module 131, the target processing module 132, the dual-channel module 133, and the target sending module 134, and is installed in the industrial control box 1. The attitude control module 14 consists of the target reading module 141, the data aggregation module 142, the internal sensor receiving module 143, the aggregation output module 144, the trajectory receiving module 145, and the flight adjustment module 146. The data aggregation module 142 is connected to the internal sensor receiving module 143, which is connected to the internal sensor output module 151 within the internal sensor module 15. The internal sensor module 15 consists of the internal sensor output module 151, the acceleration output module 152, the angular velocity output module 153, the geomagnetic direction output module 154, the geographic location output module 155, the accelerometer module 156, the gyroscope module 157, the magnetometer module 158, and the GPS module 159. The internal sensor output module 151 is connected to the acceleration output module 152, which is connected to the accelerometer module 156. The internal sensor output module 151 is connected to the angular velocity output module 153, which is connected to the gyroscope module 157. The internal sensor output module 151 is connected to the geomagnetic direction output module 154, which is connected to the magnetometer module 158. The internal sensor output module 151 is connected to the geographic location output module 155, which is connected to the GPS module 159. The data aggregation module 142 is connected to the aggregation output module 144, which is connected to the aggregation receiving module 161 within the optimization module 16. The aggregation receiving module 161 is connected to the aggregation processing module 162. The optimization module 16 consists of the aggregation receiving module 161, the aggregation processing module 162, the trajectory generation module 163, and the trajectory transmission module 164. The aggregation processing module 162 ís connected to the trajectory generation module 163, which is connected to the trajectory transmission module 164. The trajectory transmission module 164 is connected to the trajectory receiving module 145, which is connected to the flight adjustment module 146, The external data receiving module 121 is connected to the external sensor module 17, which is installed in the industrial control box 1. The industrial control box 1 houses the attitude control module 14, the internal sensor module 15, and the optimization module 16.
Refer to FIG.9. An embodiment provided by the present invention is a rapid three- dimensional trajectory planning method for an emergency rescue drone, which comprises the following steps: Step 1, data collection and transmission; Step 2, method optimization; Step 3, trajectory generation.
In Step 1, the power supply module 11 powers the entire device, and the external data receiving module 121 collects video data from the external sensor module 17.
The external target detection module 122 identifies and tracks dynamic or static targets in the video, transmitting the detected targets to the target position conversion module 123. The target position conversion module 123 processes the target data and converts it into target position data, which is then transmitted to the target position output module 124. The target position output module 124 sends the data to the target receiving module 131, which forwards it to the target processing module 132.
The target processing module 132 processes the data by removing all state fields from the coordinate data to reduce the packet header. The processed target data is then transmitted to the target sending module 134 via the dual-channel module 133. The target sending module 134 sends the target data to the target reading module 141, and the data aggregation module 142 collects the target data. Simultaneously, the accelerometer module 156 collects acceleration data and transmits it to the acceleration output module 152, which sends it to the internal sensor output module 151. The gyroscope module 157 collects angular velocity data and transmits it to the angular velocity output module 153, which forwards it to the internal sensor output module 151. The magnetometer module 158 collects geomagnetic direction data and transmits it to the geomagnetic direction output module 154, which sends it to the internal sensor output module 151. The GPS module 159 collects geographic location data and transmits it to the geographic location output module 155, which forwards it to the internal sensor output module 151. The internal sensor output module 151 transmits the collected acceleration, angular velocity, geomagnetic direction, and geographic location data to the internal sensor receiving module 143, which forwards the data to the data aggregation module 142. The data aggregation module 142 aggregates all collected data and transmits it to the aggregation output module 144.
The aggregation output module 144 then sends the aggregated data to the aggregation receiving module 161;
In Step 2, the aggregation receiving module 161 receives the aggregated data and transmits it to the aggregation processing module 162, where a three-dimensional drone flight map is created using the following function: z(x,v)=sin(y +a) +b ¥sin(x) +c *cos{(d ¥(v2 + x2) +e*cos(y}+ f *sm(f *(y2+ x2) + g *cos(y) where z represents the ground elevation at the current position of the drone, and x, y are the X-axis and Y-axis coordinates of the terrain, respectively; parameters a b, c, d, e, f, g are defined based on sensor data to model the terrain; the threat zone's center coordinates and radius are set based on data detected by the target tracking processor and overlaid on the map; considering the drone's acceleration and angular velocity changes, the starting and ending points of the drone's flight are determined; the drone's navigation cost is evaluated using the path length L, height evaluation H, and turning angle evaluation C with the fitness function F:
E=WiL+Wo H+WaC where WLW: Ws are weight coefficients summing to 1; path length L is calculated as the sum of Euclidean distances between consecutive points on the path; height evaluation H is the root mean square of the differences between the heights at each path point and the average height; turning angle evaluation ( assesses the sharpness of turns by calculating the cosine values between vectors of consecutive points; method parameters and individual positions are initialized, with each position representing a set of drone trajectory coordinates; adaptive differential evolution mutation updates individual positions Y (t):
Y(t)=Flt)+a* (Tt) -Ts(2)) where Yn, Yo, Vr: are three randomly selected individuals in the population, and « is the adaptive step coefficient, defined as follows: ce *1.1, diversity <0.1 a= a: *0.9, else the adaptive step coefficient « is dynamically adjusted based on population diversity, calculated every 50 iterations; diversity is defined as the mean ‚4 of the standard deviations o; of the population positions across all dimensions dim: 1 dim diversity = dim 247 based on probability p , the individual positions are updated in segments; when p is less than a random number +, the position is updated using a sine-based method:
Yiot+)=Y. (f)+nll+sm(r})) ri (A), p<r in other cases, the position is updated as follows:
Yi (t+ 1)=Yi (6)+n(2r=1)*Y. (1),else an enhanced local search strategy (ELS) is incorporated to refine the search and optimize local individual positions; this is done through a perturbation factor
Io , which identifies parts of the population with significant position changes and performs a secondary update on the global position to generate a local optimal position: vr LI Dy)
Tr where ra is arandom number in the range (0,1), f is a tuning parameter, and 7 and { are the maximum and current iteration counts, respectively; the resulting
Yi j(t) represents the optimal position for each individual, and the fitness value F is calculated; the positions of individuals in the population are converted into drone trajectory coordinates, and the trajectory generation module 163 generates a three- dimensional trajectory, producing the planned drone path;
In Step 3, the three-dimensional trajectory generated by the trajectory generation module 163 is transmitted to the trajectory transmission module 164. The trajectory transmission module 164 then transmits the three-dimensional trajectory to the trajectory receiving module 145, The trajectory receiving module 145 receives the three-dimensional trajectory and immediately transmits it to the flight adjustment module 146. The flight adjustment module 146 adjusts the drone's flight attitude and direction based on the position of each coordinate point, enabling effective obstacle avoidance.
Operating Principle: During use, the external data receiving module 121 collects video data from the external sensor module 17, and the external target detection module 122 identifies and tracks dynamic or static targets in the video. The detected targets are transmitted to the target position conversion module 123, which processes the target data and converts it into target position data. This data is then transmitted to the target position output module 124, which forwards it to the target receiving module 131. The target receiving module 131 transmits the data to the target processing module 132, where it is processed, with all state fields removed from the coordinate data to reduce the packet header, thereby improving the transmission speed of coordinate data and reducing the response time for drone flight control. The processed target data is transmitted to the target sending module 134 via the dual- channel module 133. The dual-channel module 133 includes an independent backup communication channel that enhances communication reliability, ensuring stable data transmission even in the event of main channel failure. The aggregated data is received by the aggregation receiving module 161 and transmitted to the aggregation processing module 162, where method improvements and optimizations are carried out. The method incorporates two individual position update strategies to optimize global search capability and enhance local search performance. An enhancement factor identifies groups with high variation rates, updating local optimal solutions, refining individual positions, and improving local trajectory coordinates. This achieves the simultaneous fulfillment of reducing trajectory costs, dynamically planning paths, and avoiding small obstacles.
For those skilled in the art, it is apparent that the present invention is not limited to the details of the exemplary embodiments described above but can be implemented in other specific forms without departing from the spirit or essential characteristics of the invention. Therefore, the embodiments should be considered exemplary and non-limiting, with the scope of the invention defined by the appended claims rather than the foregoing description. It is intended that all changes falling within the meaning and range of equivalency of the claims are encompassed within the present invention. Any reference numerals in the claims should not be considered limiting to the claims they pertain to.

Claims (10)

CONCLUSIESCONCLUSIONS 1. Een snel driedimensionaal trajectplanningsapparaat voor noodreddingsdrones, omvattende een doelbewakingsmodule (12), waarbij: de doelbewakingsmodule (12) een externe gegevensontvangstmodule {121}, een externe doeldetectiemodule {122}, een doelpositietransformatiemodule (123) en een doelpositie-uitvoermodule (124) bevat; de externe gegevensontvangstmodule (121) is verbonden met de externe doeldetectiemodule {122}, die op zijn beurt verbonden is met de doelpositietransformatiemodule (123); de doelpositietransformatiemodule {123} is verbonden met de doelpositie-uitvoermodule {124}, die verbonden is met een doelontvangstmodule {131} binnen een controllermodule {13}; de doelontvangstmodule (131) is verbonden met een doelverwerkingsmodule {132}, die verbonden is met een dubbelkanaalsmodule (133); de dubbelkanaalsmodule (133) is verbonden met een doelzendmodule (134); de doelzendmodule (134) is verbonden met een doelleesmodule (141) binnen een houdingsbesturingsmodule (14); de doelleesmodule {141} is verbonden met een gegevensaggregatiemodule (142); de doelbewakingsmodule (12) is geplaatst in een industriële controlebox (1), die voorzien is van een voedingsmodule (11).1. A rapid three-dimensional trajectory planning device for emergency rescue drones, comprising a target monitoring module (12), wherein: the target monitoring module (12) includes an external data receiving module {121}, an external target detection module {122}, a target position transformation module (123) and a target position output module (124); the external data receiving module (121) is connected to the external target detection module {122}, which in turn is connected to the target position transformation module (123); the target position transformation module {123} is connected to the target position output module {124}, which is connected to a target receiving module {131} within a controller module {13}; the target receiving module (131) is connected to a target processing module {132}, which is connected to a dual-channel module (133); the dual channel module (133) is connected to a target transmitting module (134); the target transmitting module (134) is connected to a target reading module (141) within an attitude control module (14); the target reading module {141} is connected to a data aggregation module (142); the target monitoring module (12) is placed in an industrial control box (1), which is provided with a power supply module (11). 2. Het snelle driedimensionale trajectplanningsapparaat voor noodreddingsdrones volgens conclusie 1, waarbij: de controller-module (13) de doelontvangstmodule (131), de doelverwerkingsmodule (132), de dubbelkanaalsmodule {133) en de doelzendmodule (134) bevat; de controller-module (13) is geplaatst in de industriële controlebox {1}.2. The rapid three-dimensional trajectory planning device for emergency rescue drones according to claim 1, wherein: the controller module (13) includes the target receiving module (131), the target processing module (132), the dual-channel module (133) and the target transmitting module (134); the controller module (13) is placed in the industrial control box {1}. 3. Het snelle driedimensionale trajectplanningsapparaat voor noodreddingsdrones volgens conclusie 1, waarbij: de houdingsbesturingsmodule (14) de doelleesmodule (141), de gegevensaggregatiemodule (142), een interne sensorontvangstmodule {143}, een aggregatie-uitvoermodule (144), een trajectontvangstmodule (145) en een vluchtverstellingsmodule (146) bevat; de gegevensaggregatiemodule (142) is verbonden met de interne sensorontvangstmodule (143), die verbonden is met een interne sensoruitvoermodule (151) binnen een interne sensormodule (15). 3. The rapid three-dimensional trajectory planning device for emergency rescue drones according to claim 1, wherein: the attitude control module (14) includes the target reading module (141), the data aggregation module (142), an internal sensor receiving module {143}, an aggregation output module (144), a trajectory receiving module (145) and a flight adjustment module (146); the data aggregation module (142) is connected to the internal sensor receiving module (143), which is connected to an internal sensor output module (151) within an internal sensor module (15). 4, Het snelle driedimensionale trajectplanningsapparaat voor noodreddingsdrones volgens conclusie 3, waarbij: de interne sensormodule (15) de interne sensoruitvoermodule (151), een versnellingsuitvoermodule {152}, een hoeksnelheidsuitvoermodule (153), een geomagnetische richtinguitvoermodule (154), een geografische locatie-uitvoermodule (155), een versnellingsmeter-module (156), een gyroscoopmodule (157), een magnetometermodule (158) en een GPS-module (159) bevat; de interne sensoruitvoermodule (151) is verbonden met de versnellingsuitvoermodule (152), die verbonden is met de versnellingsmeter-module (156).4. The rapid three-dimensional trajectory planning device for emergency rescue drones according to claim 3, wherein: the internal sensor module (15) includes the internal sensor output module (151), an acceleration output module {152}, an angular velocity output module (153), a geomagnetic direction output module (154), a geographic location output module (155), an accelerometer module (156), a gyroscope module (157), a magnetometer module (158), and a GPS module (159); the internal sensor output module (151) is connected to the acceleration output module (152), which is connected to the accelerometer module (156). 5. Het snelle driedimensionale trajectplanningsapparaat voor noodreddingsdrones volgens conclusie 4, waarbij: de interne sensoruitvoermodule (151) is verbonden met de hoeksnelheidsuitvoermodule (153), die verbonden is met de gyroscoopmodule (157); de interne sensoruitvoermodule {151} is tevens verbonden met de geomagnetische richtinguitvoermodule (154), die verbonden is met de magnetometermodule (158).5. The rapid three-dimensional trajectory planning device for emergency rescue drones according to claim 4, wherein: the internal sensor output module (151) is connected to the angular velocity output module (153), which is connected to the gyroscope module (157); the internal sensor output module {151} is also connected to the geomagnetic direction output module (154), which is connected to the magnetometer module (158). 6. Het snelle driedimensionale trajectplanningsapparaat voor noodreddingsdrones volgens conclusie 5, waarbij: de interne sensoruitvoermodule (151) is verbonden met de geografische locatie-uitvoermodule (155), die verbonden is met de GPS-module (159).6. The rapid three-dimensional trajectory planning device for emergency rescue drones according to claim 5, wherein: the internal sensor output module (151) is connected to the geographic location output module (155), which is connected to the GPS module (159). 7. Het snelle driedimensionale trajectplanningsapparaat voor noodreddingsdrones volgens conclusie 3, waarbij: de gegevensaggregatiemodule (142) is verbonden met de aggregatie-uitvoermodule {144}, die verbonden is met een aggregatie-ontvangstmodule (161) binnen een optimalisatiemodule (16); de aggregatie-ontvangstmodule (161) is verbonden met de aggregatieverwerkingsmodule (162).7. The fast three-dimensional trajectory planning device for emergency rescue drones according to claim 3, wherein: the data aggregation module (142) is connected to the aggregation output module {144}, which is connected to an aggregation receiving module (161) within an optimization module (16); the aggregation receiving module (161) is connected to the aggregation processing module (162). 8. Het snelle driedimensionale trajectplanningsapparaat voor noodreddingsdrones volgens conclusie 7, waarbij: de optimalisatiemodule (16) de aggregatie-ontvangstmodule (161), de aggregatieverwerkingsmodule (162), een trajectgeneratiemodule (163) en een trajectzendmodule (164) bevat; de aggregatieverwerkingsmodule {162} is verbonden met de trajectgeneratiemodule (163), die verbonden is met de trajectzendmodule (164); de trajectzendmodule (164) is verbonden met de trajectontvangstmodule {145}, die verbonden is met de viuchtverstellingsmodule (146).8. The fast three-dimensional trajectory planning device for emergency rescue drones according to claim 7, wherein: the optimization module (16) includes the aggregation receiving module (161), the aggregation processing module (162), a trajectory generation module (163) and a trajectory transmitting module (164); the aggregation processing module {162} is connected to the trajectory generation module (163), which is connected to the trajectory transmitting module (164); the trajectory transmitting module (164) is connected to the trajectory receiving module {145}, which is connected to the flight adjustment module (146). 9. Het snelle driedimensionale trajectplanningsapparaat voor noodreddingsdrones volgens conclusie 1, waarbij: de externe gegevensontvangstmodule (121) verbonden is met een externe sensormodule (17); de externe sensormodule (17) is geplaatst in de industriële controlebox {1}, die tevens de houdingsbesturingsmodule (14), de interne sensormodule (15) en de optimalisatiemodule (16) bevat.9. The rapid three-dimensional trajectory planning device for emergency rescue drones according to claim 1, wherein: the external data receiving module (121) is connected to an external sensor module (17); the external sensor module (17) is placed in the industrial control box {1}, which also includes the attitude control module (14), the internal sensor module (15) and the optimization module (16). 10. Een snelle driedimensionale trajectplanningswerkwijze voor noodreddingsdrones, omvattende de volgende stappen: Stap 1: Gegevensverzameling en -transmissie; Stap 2: Werkwijzeoptimalisatie; Stap 3: Trajectgeneratie; waarbij: In Stap 1 voorziet de voedingsmodule (11) het gehele apparaat van stroom, en verzamelt de externe gegevensontvangstmodule (121) videogegevens van de externe sensormodule (17); de externe doeldetectiemodule {122} identificeert en volgt dynamische of statische doelen in de video en zendt het gedetecteerde doel naar de doelpositietransformatiemodule (123); de doelpositietransformatiemodule (123) verwerkt de doelgegevens en zet deze om in doelpositiegegevens, die vervolgens naar de doelpositie-uitvoermodule (124) worden verzonden; de doelpositie-uitvoermodule10. A rapid three-dimensional trajectory planning method for emergency rescue drones, comprising the following steps: Step 1: Data collection and transmission; Step 2: Process optimization; Step 3: Trajectory generation; wherein: In Step 1, the power module (11) supplies power to the entire device, and the external data receiving module (121) collects video data from the external sensor module (17); the external target detection module {122} identifies and tracks dynamic or static targets in the video and sends the detected target to the target position transformation module (123); the target position transformation module (123) processes the target data and converts it into target position data, which is then sent to the target position output module (124); the target position output module (124) stuurt de gegevens naar de doelontvangstmodule (131), die deze doorstuurt naar de doelverwerkingsmodule (132); de doelverwerkingsmodule {132) verwerkt de gegevens, waarbij alle toestandsvelden uit de coördinaatgegevens worden verwijderd om de pakketkop te verkleinen; de verwerkte doelgegevens worden vervolgens via de dubbelkanaalsmodule (133) naar de doelzendmodule (134) verzonden; de doelzendmodule {134) stuurt de doelgegevens naar de doelleesmodule (141), en de gegevensaggregatiemodule (142) verzamelt de doelgegevens; gelijktijdig verzamelt de versnellingsmetermodule (156) versnellingsgegevens en stuurt deze naar de versnellingsuitvoermodule (152), die de gegevens doorstuurt naar de interne sensoruitvoermodule (151); de gyroscoopmodule (157) verzamelt hoeksnelheidsgegevens en stuurt deze naar de hoeksnelheidsuitvoermodule (153), die de gegevens doorstuurt naar de interne sensoruitvoermodule (151); de magnetometermodule (158) verzamelt geomagnetische richtingsgegevens en stuurt deze naar de geomagnetische richtinguitvoermodule (154), die de gegevens doorstuurt naar de interne sensoruitvoermodule (151); de GPS-module (159) verzamelt geografische locatiegegevens en stuurt deze naar de geografische locatie- uitvoermodule (155), die de gegevens doorstuurt naar de interne sensoruitvoermodule {151}; de interne sensoruitvoermodule (151} stuurt de verzamelde versnellings-, hoeksnelheids-, geomagnetische richting- en geografische locatiegegevens naar de interne sensorontvangstmodule (143), die de gegevens doorstuurt naar de gegevensaggregatiemodule (142); de gegevensaggregatiemodule (142) aggregeert alle verzamelde gegevens en stuurt deze door naar de aggregatie- uitvoermodule {144}, die de geaggregeerde gegevens doorstuurt naar de aggregatie- ontvangstmodule (161).(124) sends the data to the target receiving module (131), which forwards it to the target processing module (132); the target processing module {132) processes the data, removing all state fields from the coordinate data to reduce the packet header; the processed target data is then sent to the target transmitting module (134) via the dual-channel module (133); the target transmitting module {134) sends the target data to the target reading module (141), and the data aggregation module (142) collects the target data; simultaneously, the accelerometer module (156) collects acceleration data and sends it to the acceleration output module (152), which forwards the data to the internal sensor output module (151); the gyroscope module (157) collects angular velocity data and sends it to the angular velocity output module (153), which forwards the data to the internal sensor output module (151); the magnetometer module (158) collects geomagnetic direction data and sends it to the geomagnetic direction output module (154), which forwards the data to the internal sensor output module (151); the GPS module (159) collects geographic location data and sends it to the geographic location output module (155), which forwards the data to the internal sensor output module {151}; the internal sensor output module (151} sends the collected acceleration, angular velocity, geomagnetic direction and geographic location data to the internal sensor receiving module (143), which forwards the data to the data aggregation module (142); the data aggregation module (142) aggregates all collected data and forwards it to the aggregation output module {144}, which forwards the aggregated data to the aggregation receiving module (161). In Stap 2 ontvangt de aggregatie-ontvangstmodule (161) de geaggregeerde gegevens en stuurt deze naar de aggregatieverwerkingsmodule (162), waar een driedimensionale droneviuchtkaart wordt gecreëerd met de volgende functie:In Step 2, the aggregation receiving module (161) receives the aggregated data and sends it to the aggregation processing module (162), where a three-dimensional drone flight map is created with the following function: z(x,y) =sin(y +a) +6 sin(x) +c*cos(d*(y2+x2)) +e*cos(y) + f *sin( J *(y2+x2)) + g *cos(y) waarbij z de hoogte van de grond op de huidige positie van de drone vertegenwoordigt en x, y respectievelijk de X- en Y-ascoördinaten van het terrein zijn; de parameters a, b, c, d, e. f, g worden bepaald op basis van sensorgegevens om het terrein te modelleren.z(x,y) =sin(y +a) +6 sin(x) +c*cos(d*(y2+x2)) +e*cos(y) + f *sin( J *(y2+x2)) + g *cos(y) where z represents the ground height at the drone's current position and x, y are the X- and Y-axis coordinates of the terrain, respectively; the parameters a, b, c, d, e. f, g are determined based on sensor data to model the terrain. De coördinaten en straal van de bedreigingszone worden ingesteld op basis van gegevens die zijn gedetecteerd door de doelsequentietracker en over de kaart gelegd, rekening houdend met de versnelling en hoeksnelheidsveranderingen van de drone worden de start- en eindpunten van de vlucht bepaald, de navigatiekosten van de drone worden geëvalueerd met behulp van de padlengte L, hoogte-evaluatie H en draaievaluatie C volgens de fitnessfunctie F: F=W*L+W>2*H+W:*C waarbij, Wi W2,W: de gewichtscoéfficiénten zijn die samen 1 vormen; de padlengte L wordt berekend als de som van de Euclidische afstanden tussen opeenvolgende punten op het pad; de hoogte-evaluatie H wordt berekend als de wortel van de som van de kwadraten van de verschillen tussen de hoogten van de punten op het pad en degemiddelde hoogte; de draaievaluatie C beoordeelt de intensiteit van de bochten door de cosinuswaarden te berekenen van de hoeken tussen de vectoren van opeenvolgende punten op het pad, de werkwijze initialiseert de parameters en de posities van individuen, waarbij elke positie een reeks coördinaten van een drone- traject vertegenwoordigt, de individuele posities Y (t) worden geüpdatet met behulp van adaptieve differentiële evolutiemutatie als volgt: YO) = (0) +r * (Fut) = Yo) waarbij YY Yrs respectievelijk drie willekeurig geselecteerde individuen uit de populatie zijn, en a de adaptieve stapcoëfficiënt is die de mutatieamplitude controleert, de berekening van « is als volgt: or ai § *1.1, diversity <0.1 a: *0.9, else De adaptieve stapcoëfficiënt «a wordt dynamisch aangepast, waarbij de populatiediversiteit diversity elke 50 iteraties wordt berekend, de diversiteit wordt gedefinieerd als het gemiddelde 44 van de standaarddeviaties van de populatieposities over alle dimensies (dim), berekend als: 1 dim diversity = —— ©; ’ Gm > Op basis van een waarschijnlijkheid p worden de posities van individuen in segmenten geüpdatet, wanneer p< r wordt de positie bijgewerkt met een sinus- gebaseerde werkwijze: Yo (t+ )=Y. (0) +nl+snr)*Y (fp <r In andere gevallen wordt de positie bijgewerkt als volgt: Vi t+) =Yi (0) +n(2r—-1)*Y. ;(f) else Het toevoegen van een Enhanced Local Search-strategie (ELS), waarmee de zoekopdracht wordt verfijnd en de lokale posities van individuen worden i . Frat (I —t geoptimaliseerd, door middel van een verstoringsfactor Lp wordt de groep individuen geïdentificeerd met grotere positieveranderingen, vervolgens wordt een secundaire globale positie-update uitgevoerd om een lokaal optimale positie te genereren: *ra* (1 —t Yan LI De 7 waarbij ra een willekeurig getal is in het bereik {0, 1), f een aanpassingsparameter is, en 7’ en 1 respectievelijk het maximale en huidige aantal iteraties zijn, de resulterende Yi (tf) vertegenwoordigt de optimale positie van elk individu, de fitnesswaarde F wordt berekend, en de posities van de individuen in de populatie worden omgezet in coördinaten voor het drone-traject, vervolgens genereert de trajectgeneratiemodule (163) een driedimensionaal traject, dat wordt beschouwd als het geplande drone-traject voor deze iteratie,The coordinates and radius of the threat zone are set based on data detected by the target sequence tracker and overlaid on the map, taking into account the acceleration and angular velocity changes of the drone, the start and end points of the flight are determined, the navigation cost of the drone is evaluated using the path length L, height evaluation H and turn evaluation C according to the fitness function F: F=W*L+W>2*H+W:*C where, Wi W2,W: are the weight coefficients that add up to 1; the path length L is calculated as the sum of the Euclidean distances between successive points on the path; the height evaluation H is calculated as the root of the sum of the squares of the differences between the heights of the points on the path and the average height; the turn evaluation C evaluates the intensity of the turns by calculating the cosine values of the angles between the vectors of successive points on the path, the method initializes the parameters and the positions of individuals, where each position represents a set of coordinates of a drone trajectory, the individual positions Y(t) are updated using adaptive differential evolution mutation as follows: YO) = (0) +r * (Fut) = Yo) where YY Yrs are three randomly selected individuals from the population respectively, and a is the adaptive step coefficient controlling the mutation amplitude, the calculation of « is as follows: or ai § *1.1, diversity <0.1 a: *0.9, else The adaptive step coefficient «a is adjusted dynamically, calculating the population diversity diversity every 50 iterations, the diversity is defined as the average 44 of the standard deviations of the population positions over all dimensions (dim), calculated as: 1 dim diversity = —— ©; ’ Gm > Based on a probability p the positions of individuals in segments are updated, when p< r the position is updated using a sine-based method: Yo (t+ )=Y. (0) +nl+snr)*Y (fp < r Otherwise the position is updated as follows: Vi t+) =Yi (0) +n(2r—-1)*Y. ;(f) else Adding an Enhanced Local Search (ELS) strategy, which refines the search and updates the local positions of individuals i . Frat (I —t optimized, by means of a disturbance factor Lp the group of individuals with larger position changes is identified, then a secondary global position update is performed to generate a locally optimal position: *ra* (1 —t Yan LI De 7 where ra is a random number in the range {0, 1), f is an adjustment parameter, and 7’ and 1 are the maximum and current number of iterations respectively, the resulting Yi (tf) represents the optimal position of each individual, the fitness value F is calculated, and the positions of the individuals in the population are converted into coordinates for the drone trajectory, then the trajectory generation module (163) generates a three-dimensional trajectory, which is regarded as the planned drone trajectory for this iteration, in Stap 3 wordt het door de trajectgeneratiemodule (163) gegenereerde driedimensionale traject overgedragen aan de trajectzendmodule (164), die het traject verzendt naar de trajectontvangstmodule (145), de trajectontvangstmodule (145) ontvangt het driedimensionale traject en stuurt het onmiddellijk door naar de vluchtverstellingsmodule (146), deze module past de viucht- en houding van de drone aan op basis van elk coördinaatpunt, zodat obstakels effectief worden vermeden.In Step 3, the three-dimensional trajectory generated by the trajectory generation module (163) is transmitted to the trajectory transmitting module (164), which transmits the trajectory to the trajectory receiving module (145), the trajectory receiving module (145) receives the three-dimensional trajectory and immediately forwards it to the flight adjustment module (146), this module adjusts the flight and attitude of the drone based on each coordinate point, so as to effectively avoid obstacles.
NL2039176A 2024-11-15 2024-11-27 Emergency rescue drone rapid three-dimensional trajectory planning device and method NL2039176B1 (en)

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202411632314 2024-11-15

Publications (2)

Publication Number Publication Date
NL2039176A true NL2039176A (en) 2025-02-14
NL2039176B1 NL2039176B1 (en) 2025-08-26

Family

ID=94538105

Family Applications (1)

Application Number Title Priority Date Filing Date
NL2039176A NL2039176B1 (en) 2024-11-15 2024-11-27 Emergency rescue drone rapid three-dimensional trajectory planning device and method

Country Status (1)

Country Link
NL (1) NL2039176B1 (en)

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20210300553A1 (en) * 2020-03-27 2021-09-30 Sony Corporation Safeguarded delivery of goods by unmanned aerial vehicles
US20240038077A1 (en) * 2022-08-01 2024-02-01 Ideaforge Technology Pvt. Ltd. Advanced pilot assistance system (apas) for estimating coverage area and viewing area and method thereof
US20240044651A1 (en) * 2020-12-18 2024-02-08 Safe Ops Systems, Inc. Systems and methods for dispatching and navigating an unmanned aerial vehicle

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20210300553A1 (en) * 2020-03-27 2021-09-30 Sony Corporation Safeguarded delivery of goods by unmanned aerial vehicles
US20240044651A1 (en) * 2020-12-18 2024-02-08 Safe Ops Systems, Inc. Systems and methods for dispatching and navigating an unmanned aerial vehicle
US20240038077A1 (en) * 2022-08-01 2024-02-01 Ideaforge Technology Pvt. Ltd. Advanced pilot assistance system (apas) for estimating coverage area and viewing area and method thereof

Also Published As

Publication number Publication date
NL2039176B1 (en) 2025-08-26

Similar Documents

Publication Publication Date Title
Goudarzi et al. Data collection using unmanned aerial vehicles for Internet of Things platforms
Du et al. A real-time collision avoidance strategy in dynamic airspace based on dynamic artificial potential field algorithm
CN112015199B (en) Track planning method and device for intelligent inspection drones in coal mines
CN115145275B (en) Multi-robot formation obstacle avoidance control method based on improved artificial potential field method
CN112230634A (en) A kind of robot obstacle avoidance method and device
CN115167440A (en) A multi-robot formation control method based on virtual pilot-lead-follow
JP7044061B2 (en) Mobiles, mobile control systems, mobile control methods, interface devices, and programs
CN109669481A (en) Spacecraft Safe Approach Control Method Based on Equal Collision Probability Surface Method
CN118500414B (en) A global path planning method and system based on improved artificial potential field method
CN116627127A (en) A Cooperative Exploration Method for Multiple Mobile Robots in Unknown Environment
CN110632940B (en) Active anti-interference time-varying track tracking control method for multiple unmanned aerial vehicles with hybrid quantizers
NL2039176B1 (en) Emergency rescue drone rapid three-dimensional trajectory planning device and method
Wang et al. Vision-Based Deep Reinforcement Learning of Unmanned Aerial Vehicle (UAV) Autonomous Navigation Using Privileged Information
Han et al. Underwater multi-target node path planning in hybrid action space: a deep reinforcement learning approach
CN116820129A (en) Unmanned aerial vehicle cluster formation method and system based on autonomous collaboration mechanism
CN115021800B (en) Method and device for searching Ka frequency band satellite terminal by using unmanned aerial vehicle and electronic equipment
KR20150068742A (en) Method and System for Formation Control of Multiple Mobile Robots
TWI809727B (en) Method for searching a path by using a three-dimensional reconstructed map
CN116593962A (en) A Radiation Source Location Method Based on Online UAV Track Planning
Min et al. Online trajectory planning strategy for UAV in dynamic threats based on bug-rapidly-exploring random tree algorithm
CN119596953A (en) A method, device, equipment and medium for cooperative capture of unmanned vehicle formation based on piloting and following
CN107272736A (en) A kind of unmanned plane avoidance guidance method flown based on pigeon passive type avoidance
CN115346139A (en) A kind of unmanned aerial vehicle laser cleaning method, device and equipment for solar photovoltaic panels
CN115167426A (en) Path planning method and device for multiple self-moving devices and self-moving device
Kilic et al. Leader-follower control and distributed communication based UAV swarm navigation in GPS-denied environment