CN113156992A - Three-layer architecture collaborative optimization system and method for unmanned aerial vehicle in edge environment - Google Patents
Three-layer architecture collaborative optimization system and method for unmanned aerial vehicle in edge environment Download PDFInfo
- Publication number
- CN113156992A CN113156992A CN202110390787.7A CN202110390787A CN113156992A CN 113156992 A CN113156992 A CN 113156992A CN 202110390787 A CN202110390787 A CN 202110390787A CN 113156992 A CN113156992 A CN 113156992A
- Authority
- CN
- China
- Prior art keywords
- layer
- task
- module
- edge
- cloud server
- 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.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course or altitude of land, water, air, or space vehicles, e.g. automatic pilot
- G05D1/10—Simultaneous control of position or course in three dimensions
- G05D1/101—Simultaneous control of position or course in three dimensions specially adapted for aircraft
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D30/00—Reducing energy consumption in communication networks
- Y02D30/70—Reducing energy consumption in communication networks in wireless communication networks
Abstract
A three-layer architecture collaborative optimization system and method for unmanned aerial vehicles in edge environment comprises an edge node layer, a mobile edge server layer and a cloud server layer which are sequentially arranged from bottom layer to upper layer; the edge node layer is used for acquiring the computing time and energy consumption for executing the task of the specific application; under the condition that the position of an edge terminal in an edge node layer of a mobile edge server layer is determined, the bandwidth is obtained by calculation of Euclidean distance, path loss and Shannon's theorem in a three-dimensional space; determining the signal coverage range and the optimal hovering position of the unmanned aerial vehicle through bandwidth calculation, and obtaining corresponding data transmission time and task calculation time according to the bandwidth; and the cloud server layer calculates the transmission energy consumption of the edge terminal and the cloud server layer in the edge node layer, and obtains the cloud computing time and the transmission time in the corresponding network environment. The invention aims at the computation-intensive tasks sensitive to the response time in the mobile edge environment and solves the problems of larger task execution time delay and higher energy consumption.
Description
Technical Field
The invention relates to the technical field of unmanned aerial vehicles, in particular to a three-layer architecture collaborative optimization system and method for an unmanned aerial vehicle in an edge environment.
Background
In the world of everything interconnection, various video acquisition terminals and video services are widely applied, and video processing technologies in a mobile environment are widely applied to a plurality of researches. Aiming at the execution of mobile video data processing and machine learning inference tasks and scheduling strategies, efficient task scheduling strategies need to be designed aiming at specific computing resources so as to guarantee the real-time performance and the reliability of video services. For this reason, Edge Computing (Edge Computing) has been widely applied in mobile video processing environments, and near the source of computation-intensive tasks, Edge Computing utilizes local Computing resources to perform video processing tasks to meet the demands of these delay-sensitive applications in terms of computation, delay, and power consumption.
However, the problems of high wiring difficulty and low gateway computing power in the edge environment, difficulty in adapting the fixed edge server node to the requirements of resource-intensive tasks and time-sensitive tasks in the mobile environment, and the like exist; in addition, the uneven distribution of the computing tasks on the edge terminals causes that the local computing resources cannot adapt to the task processing requirements. The limited computing resources and power make it difficult for edge terminals to continuously perform real-time computer vision-like tasks. The deployment of edge computing nodes in drones, cell towers, roadside units (RSUs), and autonomous vehicles with network connectivity is a major concern in many studies of edge servers in mobile environments. Among them, unmanned aerial vehicles have gained widespread use in a large number of fields, such as agricultural production, natural or man-made disaster monitoring. By piggybacking a variety of computing or communication resources, drones are greatly expanded in their computing and communication capabilities in view of the advantages of ease of deployment and bird's eye view. Although offloading data to mobile edge server tier nodes can significantly improve latency performance, the wireless network spectrum resources are burdened when transmitting data over the wireless network, and the dramatic increase in communication requirements during data offloading presents a significant challenge to the communication resources of the on-board network. For example, if the complete autonomous flight of the unmanned aerial vehicle is researched, all bandwidths are used for transmitting video streams, and the adaptive computer vision pipeline realizes dynamic and specified task optimization on the unmanned aerial vehicle in edge calculation; research shows that compared with local computing, the game computing unloading of unloading the computing task to the edge or the ground base station has better overall performance of task execution, and simultaneously, the balance of task unloading time and energy consumption is also ensured. That is, a portion of the computing task is offloaded to a server, such as a drone, and a portion is processed locally. These studies fail to consider the comprehensive utilization of the computing power, mobility and coverage area communication capability based on wireless signal attenuation of the drone, and the resources of the multi-layer architecture cannot be efficiently utilized.
Research on the drone server now mainly includes two aspects of a small execution unit as data transmission relay and offloading of communication:
in one aspect, optimization for communication resources is provided. Existing research has primarily focused on data transmission and forwarding for communication-limited edge devices in areas where there is no communication infrastructure or limited infrastructure; the method comprises the steps of collecting effective data in the environment which is difficult for human beings to reach by means of the powerful movement characteristics of the unmanned aerial vehicle; the unmanned aerial vehicle cluster communicates and exchanges information with each other. Although these all have certain effect to the task calculation of edge device in specific field, still have the computational resource scheduling inadequately, the wireless communication connection unstable scheduling problem of unmanned aerial vehicle still exists.
UAV edge server based task offloading and path optimization, on the other hand. The unmanned aerial vehicle is sent to the task-intensive execution area, the processing of the computing task is directly carried out in the computing resources carried by the unmanned aerial vehicle, the communication requirements of the edge node layer and the cloud server layer are reduced, and meanwhile, the response time of the computing task can be effectively improved. The sum of target costs among all users in each time slot is minimized by jointly optimizing the unmanned aerial vehicle flight path, the proportion of the offloading tasks and the user scheduling variables in each time slot. However, there has been insufficient research in the selection of a calculation offload destination and multi-ground-end signal coverage for drones.
Disclosure of Invention
In order to solve the problems that in order to deal with intensive calculation application of a dynamic area, the existing research method does not comprehensively consider the effective coverage area of unmanned aerial vehicle signals and the full utilization of multilayer system calculation resources, and fails to give full play to the maximum performance of system resources, the invention provides a three-layer architecture collaborative optimization system and method for unmanned aerial vehicles in an edge environment, and the specific scheme is as follows:
a three-layer architecture collaborative optimization system for unmanned aerial vehicles in an edge environment comprises an edge node layer, a mobile edge server layer and a cloud server layer which are sequentially arranged from the bottom layer to the upper layer;
when the edge node layer executes local unloading, recording the calculation time and energy consumption of specific equipment in the edge node layer for executing the task of the specific application; when the task is unloaded, all computation in the local unloading is realized on an edge node layer, and the unloading of the mobile edge server layer and the cloud server layer is realized on edge terminals in the edge node layer, wherein the video is extracted and encoded and preprocessed, and then the video is transmitted to a corresponding server for computation;
under the condition that the position of an edge terminal in an edge node layer of a mobile edge server layer is determined, the bandwidth is obtained by calculation of Euclidean distance, path loss and Shannon's theorem in a three-dimensional space; determining the signal coverage range and the optimal hovering position of the unmanned aerial vehicle through bandwidth calculation, and obtaining corresponding data transmission time and task calculation time according to the bandwidth;
and the cloud server layer calculates the transmission energy consumption of the edge terminal and the cloud server layer in the edge node layer, and obtains the cloud computing time and the transmission time in the corresponding network environment.
A three-layer architecture collaborative optimization method for unmanned aerial vehicles in edge environment specifically comprises the following steps:
s1, designing a system composed of an edge node layer, a mobile edge server layer and a cloud server layer, wherein the system comprises any one of claims 1-4;
s2, designing a time, energy consumption, bandwidth and position model based on a mobile edge server layer according to a system architecture;
s3, designing a task unloading optimization target based on the comprehensive parameters of time, bandwidth, energy consumption and unmanned aerial vehicle position based on the model of the step S2;
s4, aiming at the position and unloading optimization target of the unmanned aerial vehicle, a multi-chromosome genetic algorithm for enhancing elite retention is used for formulating path planning and task unloading strategies of an edge node layer, a mobile edge server layer and a cloud server layer;
and S5, respectively using three mixed unloading algorithms of an exhaustion algorithm, a greedy algorithm and a multi-chromosome elite retention genetic algorithm, and detecting typical targets in the real world in a wireless network environment.
The invention has the beneficial effects that:
(1) the cloud server layer exerts strong computing capacity under the condition that sufficient margin is reserved in network bandwidth, and performs real-time video task analysis and model training and unified generation of scheduling strategies; the mobile edge server layer fully utilizes the mobile capability thereof to provide effective computing unloading and communication service for the area with insufficient computing resources; the edge node layer carries out reasonable preprocessing operation on original video data and transmits data transmission to specified upper computing resources according to scheduling requirements, and the edge node layer comprises a cloud server layer and a mobile edge server layer carried by an unmanned aerial vehicle.
The unmanned aerial vehicle with the edge server is used for executing tasks in a field full of monitoring terminals, the computing capacity and the moving capacity of the unmanned aerial vehicle and the coverage area communication capacity based on wireless signal attenuation are comprehensively considered, and resources of a multi-layer framework are efficiently utilized. By means of a task unloading strategy of a mobile edge server layer and an unmanned aerial vehicle path optimization method, the problems of large delay of task execution time and high energy consumption are solved for the computation-intensive tasks sensitive to response time under a mobile edge environment.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and those skilled in the art can also obtain other drawings according to the drawings without creative efforts.
FIG. 1 is a schematic structural diagram of an edge node layer, a mobile edge server layer, and a cloud server layer;
FIG. 2 is a block diagram of an edge node layer;
fig. 3 is a flowchart of the operation of the edge node layer.
FIG. 4 is a block diagram of a mobile edge server layer;
FIG. 5 is a flow chart of the operation of the Mobile edge Server layer;
FIG. 6 is a block diagram of a cloud server layer;
fig. 7 is a flowchart of the work of the cloud server layer.
The various symbols in the figure are illustrated as follows:
1. an edge node layer; 11. an edge termination; 101. a first state sending module; 102. a task offload policy receiving and executing module; 103. a video stream transmitting module; 104. a first task computing module;
2. moving an edge server layer; 201. a second state sending module; 202. a task unloading distribution and unmanned aerial vehicle position receiving and executing module; 203. a second video stream receiving module; 204. a third task computing module; 205. detection feedback module
3. A cloud server layer; 301. a receiving query module; 302. a task unloading distribution and unmanned aerial vehicle positioning algorithm module; 303. a first video stream receiving module; 304. a second task computing module; 305. and a result display module.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. 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, an unmanned aerial vehicle three-layer architecture collaborative optimization system based on an edge environment includes an edge node layer, a mobile edge server layer, and a cloud server layer, which are sequentially arranged from a bottom layer to an upper layer;
when the edge node layer executes local unloading (which means that data processed by the edge node layer is calculated), only the calculation time and energy consumption of a specific device in the edge node layer for executing a task of a specific application are required to be recorded; when the task is unloaded, all computation in the local unloading is realized on an edge node layer, and the unloading of the mobile edge server layer and the cloud server layer is realized on edge terminals in the edge node layer, wherein the video is extracted and encoded and preprocessed, and then the video is transmitted to a corresponding server for computation;
under the condition that the position of an edge terminal in an edge node layer of a mobile edge server layer is determined, the bandwidth is obtained by calculation of Euclidean distance, path loss and Shannon's theorem in a three-dimensional space; determining the signal coverage range and the optimal hovering position of the unmanned aerial vehicle through bandwidth calculation, and obtaining corresponding data transmission time and task calculation time according to the bandwidth;
and the cloud server layer calculates the transmission energy consumption of the edge terminal and the cloud server layer in the edge node layer, and obtains the cloud computing time and the transmission time in the corresponding network environment.
The edge node layer, the mobile edge server layer, and the cloud server layer are described in detail below.
As shown in FIG. 2, the edge node layer comprises a plurality of edge nodes arranged in sequence
The first state sending module is used for sending a task scheduling request to the cloud server layer, sending the state and position information of the edge terminal of the edge node layer to the cloud server layer through a specified port when the cloud server layer is in the state of the query device, and judging whether the cloud server layer is ready to allocate an algorithm program or not;
the task unloading strategy receiving and executing module is connected with a preset port of the cloud server layer and used for acquiring a task allocation strategy and position information of the cloud server layer; the execution module analyzes the received task allocation strategy and the position information, so that the use of the next module code is facilitated. The position information refers to the position of the mobile edge server when executing the task, and is calculated by a task allocation strategy of the cloud server and an unmanned aerial vehicle positioning algorithm module.
The video stream sending module is used for selecting the IP address of the mobile edge server layer or the cloud server layer according to the task allocation strategy and then sending the video stream to a task processing process established by the mobile edge server layer or the cloud server layer according to the IP address selected by the allocated port connection;
the first task computing module processes the tasks that are not selected to be offloaded to the server using the computing resources of the local edge node layer.
As shown in fig. 3, the steps of the edge node layer are:
SA1, the first state sending module sends a scheduling request to a receiving and inquiring module of the cloud server layer; the first state sending module judges whether the cloud server layer is ready to distribute the algorithm program, if the cloud server layer is ready to distribute the algorithm program, the step SA2 is carried out, and if the cloud server layer is not ready to distribute the algorithm program, the step SA1 is carried out again;
SA2, a first state sending module sends real-time state information of all edge terminal devices of an edge node layer and task amount information of all the devices of the edge node layer to a receiving and querying module of a cloud server layer;
SA3, after receiving a cloud server layer task unloading allocation and an allocation strategy of an unmanned aerial vehicle positioning algorithm module, a task unloading strategy receiving and executing module judges whether to execute calculation at an edge node layer according to the task allocation strategy, when the calculation is executed at the edge node layer, the task unloading strategy receiving and executing module establishes a process for sending completion information with an IP address and a port number of the cloud server layer, calculates execution time and energy consumption data through a first task calculating module after the task is completed, sends the execution time and energy consumption data to the cloud server layer through a first state sending module, and records the execution time and energy consumption data in a local log of the edge node layer; when the computation is not executed in the edge node layer, the task unloading strategy receiving and executing module establishes a process for transmitting the video stream according to the allocated IP address and the allocated port, the video stream sending module of the edge node layer transmits the video to the first video stream receiving module of the cloud server layer, the execution time and energy consumption data are computed in the second task computing module of the cloud server layer, and the data are recorded in the local log of the cloud server layer;
SA4, judging whether the tasks in the first task computing module and the second task computing module are completely finished, if not, returning to step SA1, and if so, entering a finished state.
As shown in FIG. 4, the mobile edge server layer comprises a plurality of layers arranged in sequence
The second state sending module is used for sending a task scheduling request to the cloud server layer, sending the local and position information to a certain close-distance edge node layer through a specified port, and then forwarding the information to the cloud server layer through the edge node layer;
the task unloading distribution and unmanned aerial vehicle position receiving and executing module receives unloading distribution and position information from the cloud server layer through the edge node layer of which the signal index reaches a set value in a receiving range, controls the unmanned aerial vehicle to move through the OSDK provided by the Xinjiang and moves to a distributed hovering position.
The second video receiving module is used for establishing a process for receiving the video for the equipment task unloaded to the mobile edge server layer, and receiving a task video stream from the equipment after a port is allocated to the equipment;
the second task computing module is used for starting a video analysis process and processing transmitted video data while successfully receiving the tasks from the toilet edge node layer;
and the detection feedback module is used for transmitting the result processed by the data transmission channel back to the edge node layer so as to realize the real-time processing of the detection target.
As shown in fig. 5, the step of moving the edge server layer is:
SB1, the second state sending module sends a scheduling request to the cloud server layer receiving query module, the second state sending module judges whether the cloud server layer is ready to distribute the algorithm program, if the cloud server layer is ready to distribute the algorithm program, the step SB2 is entered, and if the cloud server layer is not ready to distribute the algorithm program, the step SB1 is re-operated;
the SB2 and the second state sending module send the real-time state information of the mobile edge server layer to the receiving and inquiring module of the cloud server layer; the real-time status information includes task execution status, location and power.
The method comprises the following steps that after an SB3 module and a task unloading distribution and unmanned aerial vehicle position receiving and executing module receive a distribution strategy and an unmanned aerial vehicle position of a cloud server layer task unloading distribution and unmanned aerial vehicle positioning algorithm module, whether calculation is executed in a mobile edge server layer is judged according to the task distribution strategy, when calculation is executed in the mobile edge server layer, the task unloading distribution and unmanned aerial vehicle position receiving and executing module establishes a process of a video analysis task according to a distributed port number, a second video stream receiving module receives a video stream from a video stream sending module in an edge node layer, and a result is returned to the edge node layer after processing; when the calculation is not carried out in the mobile edge server layer, no operation is needed, if the calculation is carried out locally, the calculation is processed locally by the task calculation module, and if the calculation is carried out in the cloud end, the calculation is sent to the cloud server by the local video stream sending module. After the current task is completed, sending execution time and energy consumption data to an edge node layer, and recording the execution time and energy consumption data in a log file of a mobile edge server layer;
SB4, determine whether all tasks are completed, if yes, enter the end state, if not, return to step SB 1.
As shown in fig. 6, the cloud server layer comprises a plurality of cloud servers arranged in sequence
The system comprises a receiving and inquiring module, a scheduling module and a processing module, wherein the receiving and inquiring module is used for inquiring the state information of an edge node layer, a mobile edge server layer and a cloud server layer and receiving a state information inquiring request which is provided when needed;
the task unloading distribution and unmanned aerial vehicle positioning algorithm module is used for taking state information of relevant equipment in a three-layer framework and task information needing unloading as a judgment basis after collecting a request needing unloading tasks, and returning to the receiving and inquiring module when a task distribution algorithm program does not need to be scheduled; when a task allocation algorithm program needs to be scheduled, a genetic algorithm with elite reservation is used for calculating a task allocation strategy which meets the requirements of execution time and energy consumption weighting and is approximately minimum for task unloading, the task allocation strategy and the unmanned aerial vehicle position are sent to a mobile edge server layer, and the task allocation strategy is sent to an edge terminal in an edge node layer; judging whether the mobile edge server layer and the edge node layer return confirmation information or not, if not, continuing confirmation, if so, establishing a processing process for the tasks unloaded to the cloud server layer according to the task allocation strategy, and establishing a completion confirmation process for the tasks on the edge node layer and the mobile edge server layer; the confirmation information is judged by an ip address and comprises an ip of the unmanned aerial vehicle and an ip of the edge node layer.
The first video stream receiving module is used for receiving the video stream of the edge node layer; establishing a process by using the distributed communication resources according to the port numbers distributed by the task distribution strategy to receive the video stream transmitted by the edge node, and performing decoding processing, so that the use of a calculation module is facilitated;
the second task computing module is used for processing and computing the video stream, receiving the execution time and energy consumption data recorded by the edge node layer and the edge server after the current task is finished, recording the execution time and energy consumption data in a log file, judging whether all tasks are finished, if the tasks are finished, finishing, and if the tasks are not finished, continuing returning to the task unloading distribution and unmanned aerial vehicle positioning algorithm module to judge whether a task distribution algorithm program needs to be scheduled;
and the result display module displays the state information of each device before task allocation, displays the device state and the execution completion state during task execution, and displays the processed video stream marked with the detection target by using an OpenCV display function.
As shown in fig. 7, the working process of the cloud server layer is as follows:
the SC1 receives the scheduling request sent by the first state sending module in the edge node layer and receives whether the query module needs to schedule the task allocation algorithm program, if the query module needs the task allocation algorithm program, the step SC2 is entered, and if the query module does not need the task allocation algorithm program, the step SC1 is returned;
the SC2 receives the real-time state information of the edge node layer, the mobile edge server layer and the cloud server layer and the corresponding task amount information on all the devices of the edge node layer through the receiving and inquiring module;
the SC3, the task unloading distribution and unmanned aerial vehicle positioning algorithm module execute work, send task distribution strategies and unmanned aerial vehicle positions to the mobile edge server layer, and send task distribution strategies to edge terminals in the edge node layer; judging whether the mobile edge server layer and the edge node layer return confirmation information, if so, entering a step SC4, and if not, entering a judgment program again;
the SC4, the task unloading distribution and unmanned aerial vehicle positioning algorithm module establishes a processing process for the tasks unloaded to the cloud server layer according to the distribution strategy and establishes a complete confirmation process for the tasks executed by the edge node layer and the edge server;
the SC5 and the first video stream receiving module receive the video stream of the edge node layer, and the processed video stream is displayed on the result display module by using an OpenCV display function;
after the current task is completed, SC6 receives execution time and energy consumption data recorded by the edge node layer equipment and the mobile edge server, and records the execution time and energy consumption data in a log file of a cloud server layer;
SC7 judges whether all tasks are completed, if yes, the state is ended, if not, the process returns to step SC 1.
The invention discloses a three-layer architecture collaborative optimization method for an unmanned aerial vehicle in an edge environment, wherein the three-layer architecture comprises a cloud server layer, a mobile edge server layer and an edge node layer, and the method comprises the following specific steps:
s1, designing a system architecture consisting of three layers of computing resources, namely an edge node layer, a mobile edge server layer and a cloud server layer;
s2, designing a time, energy consumption, bandwidth and position model based on a mobile edge server layer according to a system architecture; the method specifically comprises the following steps:
s21, under the condition that the position of the mobile edge server layer at the edge node layer is determined, the bandwidth is obtained through calculation of Euclidean distance, path loss and Shannon theorem in a three-dimensional space;
and S22, determining the signal coverage area and the optimal hovering position of the unmanned aerial vehicle through bandwidth calculation, and obtaining corresponding data transmission time and task calculation time according to the bandwidth.
S3, designing a task unloading optimization target based on the time, bandwidth, energy consumption and unmanned aerial vehicle position comprehensive parameters based on the model of the step S2, and formulating unmanned aerial vehicle path planning and task unloading strategies;
when the task is unloaded, all computation in the local unloading is realized locally, the unloading of the mobile edge server layer and the cloud server layer is performed on the edge terminal firstly, video extraction and coding preprocessing are performed, and then the video is transmitted to the server for computation, and the unloading step is as follows:
s31, taking the edge terminal position, the candidate task calculation amount and the data amount in the edge node layer as input variables, and extracting the calculation model for unloading strategy formulation and unmanned aerial vehicle path optimization into the normalized weighting of minimum time and energy consumption and target optimization through the model established in the step S1; the candidate task calculation amount and the data amount refer to the task amount on all the devices of the edge node layer and the video stream data amount to be processed.
S32, optimizing the output of the calculation for each task' S offload destination selection and drone location.
In the step S3, a multi-chromosome genetic algorithm for enhancing elite retention is used for optimizing the position and unloading target of the unmanned aerial vehicle to achieve the optimal state; in the scheme, two chromosome coding modes are used for coding the task unloading strategy and the unmanned aerial vehicle position of each device: the unload option on each plant is represented by discrete real numbers 0, 1, 2, and the three-dimensional geometric coordinates of the points are represented by continuous real numbers in the range of [0, 10 ]. And searching for an approximate global optimal solution by iteratively executing population individual fitness evaluation, individual selection, chromosome crossing and variation.
S4, aiming at the position and unloading optimization target of the unmanned aerial vehicle, a multi-chromosome genetic algorithm for enhancing elite retention is used for formulating path planning and task unloading strategies of an edge node layer, a mobile edge server layer and a cloud server layer;
and S5, respectively using three mixed unloading algorithms of an exhaustion algorithm, a greedy algorithm and a multi-chromosome elite retention genetic algorithm, and carrying out HAAR, DNN, MMOD and YOLOv3 scheduling on four typical target detection applications in the real world under three network environments of Wi-Fi, 4G and 5G.
Although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.
Claims (10)
1. A three-layer architecture collaborative optimization system for unmanned aerial vehicles in an edge environment is characterized by comprising an edge node layer, a mobile edge server layer and a cloud server layer which are sequentially arranged from the bottom layer to the upper layer;
when the edge node layer executes local unloading, recording the calculation time and energy consumption of specific equipment in the edge node layer for executing the task of the specific application; when the task is unloaded, all computation in the local unloading is realized on an edge node layer, and the unloading of the mobile edge server layer and the cloud server layer is realized on edge terminals in the edge node layer, wherein the video is extracted and encoded and preprocessed, and then the video is transmitted to a corresponding server for computation;
under the condition that the position of an edge terminal in an edge node layer of a mobile edge server layer is determined, the bandwidth is obtained by calculation of Euclidean distance, path loss and Shannon's theorem in a three-dimensional space; determining the signal coverage range and the optimal hovering position of the unmanned aerial vehicle through bandwidth calculation, and obtaining corresponding data transmission time and task calculation time according to the bandwidth;
and the cloud server layer calculates the transmission energy consumption of the edge terminal and the cloud server layer in the edge node layer, and obtains the cloud computing time and the transmission time in the corresponding network environment.
2. The cooperative optimization system for three-layer architecture of unmanned aerial vehicle in edge environment according to claim 1, wherein the edge node layer comprises sequential layers
The first state sending module is used for sending a task scheduling request to the cloud server layer, sending the state and position information of the edge terminal of the edge node layer to the cloud server layer through a specified port when the cloud server layer is in the state of the query device, and judging whether the cloud server layer is ready to allocate an algorithm program or not;
the task unloading strategy receiving and executing module is connected with a preset port of the cloud server layer and used for acquiring a task allocation strategy and position information of the cloud server layer; the execution module analyzes the received task allocation strategy and the received position information;
the video stream sending module is used for selecting the IP address of the mobile edge server layer or the cloud server layer according to the task allocation strategy and then sending the video stream to a task processing process established by the mobile edge server layer or the cloud server layer according to the IP address selected by the allocated port connection;
the first task computing module processes the tasks that are not selected to be offloaded to the server using the computing resources of the local edge node layer.
3. The edge-environment-oriented unmanned aerial vehicle three-tier architecture collaborative optimization system of claim 1, wherein the mobile edge server tier comprises sequentially arranged tiers
The second state sending module is used for sending a task scheduling request to the cloud server layer, sending the state and position information of the local machine to a certain close-distance edge node layer through a specified port, and forwarding the state and position information to the receiving and inquiring module of the cloud server by the edge node;
the task unloading distribution and unmanned aerial vehicle position receiving and executing module receives unloading distribution and position information from the cloud server layer through the edge node layer of which the signal index reaches a set value in a receiving range, controls the unmanned aerial vehicle to move through the OSDK provided by the unmanned aerial vehicle and moves to a distributed hovering position;
the second video receiving module is used for establishing a process for receiving the video for the equipment task unloaded to the mobile edge server layer, and receiving a task video stream from the equipment after a port is allocated to the equipment;
the second task computing module is used for starting a video analysis process and processing transmitted video data while successfully receiving the tasks from the toilet edge node layer;
and the detection feedback module is used for transmitting the result processed by the data transmission channel back to the edge node layer so as to realize the real-time processing of the detection target.
4. The cooperative optimization system of three-tier architecture for unmanned aerial vehicle in edge environment according to claim 1, wherein the cloud server comprises sequential arrangement
The cloud server layer comprises a plurality of layers which are arranged in sequence
The system comprises a receiving and inquiring module, a scheduling module and a processing module, wherein the receiving and inquiring module is used for inquiring the state information of an edge node layer, a mobile edge server layer and a cloud server layer and receiving a state information inquiring request which is provided when needed;
the task unloading distribution and unmanned aerial vehicle positioning algorithm module is used for taking state information of relevant equipment in a three-layer framework and task information needing unloading as a judgment basis after collecting a request needing unloading tasks, and returning to the receiving and inquiring module when a task distribution algorithm program does not need to be scheduled; when a task allocation algorithm program needs to be scheduled, a genetic algorithm with elite reservation is used for task unloading to calculate a task allocation strategy which meets the requirements of execution time and energy consumption normalization and is weighted and approximately minimum, the task allocation strategy and the unmanned aerial vehicle position are sent to a mobile edge server layer, and the task allocation strategy is sent to an edge terminal in an edge node layer; judging whether the mobile edge server layer and the edge node layer return confirmation information or not, if not, continuing confirmation, if so, establishing a processing process for the tasks unloaded to the cloud server layer according to the task allocation strategy, and establishing a completion confirmation process for the tasks on the edge node layer and the mobile edge server layer;
the first video stream receiving module is used for receiving the video stream of the edge node layer;
the second task computing module is used for processing and computing the video stream, receiving the edge node layer, the edge server and the recorded execution time and energy consumption data after the current task is completed, recording the execution time and energy consumption data in a log file, judging whether all tasks are completed or not, if the tasks are completed and finished, and if the tasks are not completed, continuing returning to the task unloading distribution and unmanned aerial vehicle positioning algorithm module to judge whether a task distribution algorithm program needs to be scheduled or not;
and the result display module displays the state information of each device before task allocation, displays the device state and the execution completion state during task execution, and displays the processed video stream marked with the detection target by using an OpenCV display function.
5. A three-layer architecture collaborative optimization method for unmanned aerial vehicles in edge environment is characterized by comprising the following steps:
s1, designing a system composed of an edge node layer, a mobile edge server layer and a cloud server layer, wherein the system comprises any one of claims 1-4;
s2, designing a time, energy consumption, bandwidth and position model based on a mobile edge server layer according to a system architecture;
s3, designing a task unloading optimization target based on the comprehensive parameters of time, bandwidth, energy consumption and unmanned aerial vehicle position based on the model of the step S2;
s4, aiming at the position and unloading optimization target of the unmanned aerial vehicle, a multi-chromosome genetic algorithm for enhancing elite retention is used for formulating path planning and task unloading strategies of an edge node layer, a mobile edge server layer and a cloud server layer;
and S5, respectively using three mixed unloading algorithms of an exhaustion algorithm, a greedy algorithm and a multi-chromosome elite retention genetic algorithm, and detecting typical targets in the real world in a wireless network environment.
6. The method of claim 5, wherein the task unloading optimization goal of step S3 is:
s31, taking the edge terminal position, the candidate task calculation amount and the data amount in the edge node layer as input variables, and extracting the calculation model of unloading strategy formulation and unmanned aerial vehicle path optimization as the target optimization of the normalized weighted sum of the minimized time and the energy consumption through the model established in the step S1;
s32, optimizing the output of the calculation for each task' S offload destination selection and drone location.
7. The method of claim 6, wherein the optimization calculation in step S21 uses a multiple-coding-operator genetic algorithm with enhanced elitistry.
8. The method of claim 6, wherein the step of the edge node layer is:
SA1, the first state sending module sends a scheduling request to a receiving and inquiring module of the cloud server layer; the first state sending module judges whether the cloud server layer is ready to distribute the algorithm program, if the cloud server layer is ready to distribute the algorithm program, the step SA2 is carried out, and if the cloud server layer is not ready to distribute the algorithm program, the step SA1 is carried out again;
SA2, a first state sending module sends real-time state information of all edge terminal devices of an edge node layer and task amount information needing to be processed on all the devices in the layer to a receiving and querying module of a cloud server layer;
SA3, after receiving a cloud server layer task unloading allocation and an allocation strategy of an unmanned aerial vehicle positioning algorithm module, a task unloading strategy receiving and executing module judges whether to execute calculation by an edge node layer according to the task allocation strategy, when the calculation is executed by the edge node layer, the task unloading strategy receiving and executing module establishes a process for sending completion information with an IP address and a port number of the cloud server layer, calculates execution time and energy consumption data by a first task calculating module after the task is completed, sends the execution time and energy consumption data to the cloud server layer by a first state sending module, and records the execution time and energy consumption data in a local log of the edge node layer; when the computation is not executed in the edge node layer, the task unloading strategy receiving and executing module establishes a process for transmitting the video stream according to the allocated IP address and the allocated port, the video stream sending module of the edge node layer transmits the video to the first video stream receiving module of the cloud server layer or the mobile edge server, the execution time and the energy consumption data are computed in the second task computing module in the corresponding computing layer, and the data are recorded in the local log of the cloud server layer;
SA4, judging whether the tasks in the first task computing module and the second task computing module are completely finished, if not, returning to step SA1, and if so, entering a finished state.
9. The method of claim 6, wherein the step of moving the edge server layer is:
SB1, the second state sending module sends a scheduling request to the cloud server layer receiving query module, the second state sending module judges whether the cloud server layer is ready to distribute the algorithm program, if the cloud server layer is ready to distribute the algorithm program, the step SB2 is entered, and if the cloud server layer is not ready to distribute the algorithm program, the step SB1 is re-operated;
the SB2 and the second state sending module send the real-time state information of the mobile edge server layer to the receiving and inquiring module of the cloud server layer;
the method comprises the following steps that after an SB3 module and a task unloading distribution and unmanned aerial vehicle position receiving and executing module receive a distribution strategy and an unmanned aerial vehicle position of a cloud server layer task unloading distribution and unmanned aerial vehicle positioning algorithm module, whether calculation is executed in a mobile edge server layer is judged according to the task distribution strategy, when calculation is executed in the mobile edge server layer, the task unloading distribution and unmanned aerial vehicle position receiving and executing module establishes a process of a video analysis task according to a distributed port number, a second video stream receiving module receives a video stream from a video stream sending module in an edge node layer, and a result is returned to the edge node layer after processing; when the calculation is not carried out in the mobile edge server layer, no operation is needed, if the calculation is carried out locally, the calculation is processed locally by the task calculation module, and if the calculation is carried out in the cloud end, the calculation is sent to the cloud server by the local video stream sending module; after the current task is completed, sending execution time and energy consumption data to an edge node layer, and recording the execution time and energy consumption data in a log file of a mobile edge server layer;
SB4, determine whether all tasks are completed, if yes, enter the end state, if not, return to step SB 1.
10. The method of claim 8, wherein the cloud server layer comprises:
the SC1 receives the scheduling request sent by the first state sending module in the edge node layer and the mobile edge server layer, receives whether the query module needs to schedule a task allocation algorithm program, if the query module needs the task allocation algorithm program, the step SC2 is carried out, and if the query module does not need the task allocation algorithm program, the step SC1 is carried out;
the SC2 receives the real-time state information of the edge node layer, the mobile edge server layer and the cloud server layer and the task information of the edge node layer through the receiving and inquiring module;
the SC3, the task unloading distribution and unmanned aerial vehicle positioning algorithm module execute work, send task distribution strategies and unmanned aerial vehicle positions to the mobile edge server layer, and send task distribution strategies to edge terminals in the edge node layer; judging whether the mobile edge server layer and the edge node layer return confirmation information, if so, entering a step SC4, and if not, entering a judgment program again;
the SC4, the task unloading distribution and unmanned aerial vehicle positioning algorithm module establishes a processing process for the tasks unloaded to the cloud server layer according to the distribution strategy and establishes a complete confirmation process for the tasks executed by the edge node layer and the edge server;
the SC5 and the first video stream receiving module receive the video stream of the edge node layer, and the processed video stream is displayed on the result display module by using an OpenCV display function;
after the current task is completed, the SC6 receives the execution time and energy consumption data recorded by the edge node layer and the mobile edge server layer, and records the execution time and energy consumption data in the log file of the cloud server layer;
SC7 judges whether all tasks are completed, if yes, the state is ended, if not, the process returns to step SC 1.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110390787.7A CN113156992B (en) | 2021-04-12 | 2021-04-12 | Three-layer architecture collaborative optimization method for unmanned aerial vehicle in edge environment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110390787.7A CN113156992B (en) | 2021-04-12 | 2021-04-12 | Three-layer architecture collaborative optimization method for unmanned aerial vehicle in edge environment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113156992A true CN113156992A (en) | 2021-07-23 |
| CN113156992B CN113156992B (en) | 2022-07-08 |
Family
ID=76890029
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110390787.7A Active CN113156992B (en) | 2021-04-12 | 2021-04-12 | Three-layer architecture collaborative optimization method for unmanned aerial vehicle in edge environment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113156992B (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113630422A (en) * | 2021-08-27 | 2021-11-09 | 北京航空航天大学杭州创新研究院 | Network security data processing method and system based on edge node |
| CN113778550A (en) * | 2021-08-17 | 2021-12-10 | 南京邮电大学 | Task unloading system and method based on mobile edge calculation |
| CN113794756A (en) * | 2021-08-26 | 2021-12-14 | 北京邮电大学 | Multi-video-stream unloading method and system supporting mobile equipment |
| CN114679492A (en) * | 2022-03-25 | 2022-06-28 | 中国联合网络通信集团有限公司 | Data unloading method, MEC server, unmanned aerial vehicle, device and system |
| CN114793339A (en) * | 2022-04-28 | 2022-07-26 | 西安交通大学 | Edge network node deployment method and system for automobile welding workshop |
| CN114895701A (en) * | 2022-04-18 | 2022-08-12 | 深圳织算科技有限公司 | Unmanned aerial vehicle inspection method and system |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20180183855A1 (en) * | 2016-12-28 | 2018-06-28 | Intel Corporation | Application computation offloading for mobile edge computing |
| US20190208007A1 (en) * | 2018-01-03 | 2019-07-04 | Verizon Patent And Licensing Inc. | Edge Compute Systems and Methods |
| CN110766159A (en) * | 2019-09-29 | 2020-02-07 | 南京理工大学 | Task allocation method for multi-UAV service edge calculation based on improved genetic algorithm |
| CN111240701A (en) * | 2019-12-31 | 2020-06-05 | 重庆大学 | Task unloading optimization method for end-edge-cloud collaborative computing |
| CN111756812A (en) * | 2020-05-29 | 2020-10-09 | 华南理工大学 | Energy consumption perception edge cloud cooperation dynamic unloading scheduling method |
| CN111901573A (en) * | 2020-08-17 | 2020-11-06 | 泽达易盛（天津）科技股份有限公司 | Fine granularity real-time supervision system based on edge calculation |
| CN112104494A (en) * | 2020-09-09 | 2020-12-18 | 南京信息工程大学 | Task security unloading strategy determination method based on air-ground cooperative edge computing network |
| CN112351503A (en) * | 2020-11-05 | 2021-02-09 | 大连理工大学 | Task prediction-based multi-unmanned-aerial-vehicle-assisted edge computing resource allocation method |
| CN112511197A (en) * | 2020-12-01 | 2021-03-16 | 南京工业大学 | Unmanned aerial vehicle auxiliary elastic video multicast method based on deep reinforcement learning |
-
2021
- 2021-04-12 CN CN202110390787.7A patent/CN113156992B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20180183855A1 (en) * | 2016-12-28 | 2018-06-28 | Intel Corporation | Application computation offloading for mobile edge computing |
| US20190208007A1 (en) * | 2018-01-03 | 2019-07-04 | Verizon Patent And Licensing Inc. | Edge Compute Systems and Methods |
| CN110766159A (en) * | 2019-09-29 | 2020-02-07 | 南京理工大学 | Task allocation method for multi-UAV service edge calculation based on improved genetic algorithm |
| CN111240701A (en) * | 2019-12-31 | 2020-06-05 | 重庆大学 | Task unloading optimization method for end-edge-cloud collaborative computing |
| CN111756812A (en) * | 2020-05-29 | 2020-10-09 | 华南理工大学 | Energy consumption perception edge cloud cooperation dynamic unloading scheduling method |
| CN111901573A (en) * | 2020-08-17 | 2020-11-06 | 泽达易盛（天津）科技股份有限公司 | Fine granularity real-time supervision system based on edge calculation |
| CN112104494A (en) * | 2020-09-09 | 2020-12-18 | 南京信息工程大学 | Task security unloading strategy determination method based on air-ground cooperative edge computing network |
| CN112351503A (en) * | 2020-11-05 | 2021-02-09 | 大连理工大学 | Task prediction-based multi-unmanned-aerial-vehicle-assisted edge computing resource allocation method |
| CN112511197A (en) * | 2020-12-01 | 2021-03-16 | 南京工业大学 | Unmanned aerial vehicle auxiliary elastic video multicast method based on deep reinforcement learning |
Non-Patent Citations (1)
| Title |
|---|
| Y. CHENG,ET AL.: "Energy-efficient Resource Allocation for UAV-empowered Mobile Edge Computing System", 《2020 IEEE/ACM 13TH INTERNATIONAL CONFERENCE ON UTILITY AND CLOUD COMPUTING (UCC)》, 30 December 2020 (2020-12-30), pages 408 - 413, XP033899183, DOI: 10.1109/UCC48980.2020.00064 * |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113778550A (en) * | 2021-08-17 | 2021-12-10 | 南京邮电大学 | Task unloading system and method based on mobile edge calculation |
| CN113778550B (en) * | 2021-08-17 | 2022-05-03 | 南京邮电大学 | Task unloading system and method based on mobile edge calculation |
| CN113794756A (en) * | 2021-08-26 | 2021-12-14 | 北京邮电大学 | Multi-video-stream unloading method and system supporting mobile equipment |
| CN113794756B (en) * | 2021-08-26 | 2022-09-06 | 北京邮电大学 | Multi-video-stream unloading method and system supporting mobile equipment |
| CN113630422A (en) * | 2021-08-27 | 2021-11-09 | 北京航空航天大学杭州创新研究院 | Network security data processing method and system based on edge node |
| CN113630422B (en) * | 2021-08-27 | 2023-04-18 | 北京航空航天大学杭州创新研究院 | Network security data processing method and system based on edge node |
| CN114679492A (en) * | 2022-03-25 | 2022-06-28 | 中国联合网络通信集团有限公司 | Data unloading method, MEC server, unmanned aerial vehicle, device and system |
| CN114679492B (en) * | 2022-03-25 | 2023-08-29 | 中国联合网络通信集团有限公司 | Data unloading method, MEC server, unmanned aerial vehicle, device and system |
| CN114895701A (en) * | 2022-04-18 | 2022-08-12 | 深圳织算科技有限公司 | Unmanned aerial vehicle inspection method and system |
| CN114793339A (en) * | 2022-04-28 | 2022-07-26 | 西安交通大学 | Edge network node deployment method and system for automobile welding workshop |
| CN114793339B (en) * | 2022-04-28 | 2023-06-30 | 西安交通大学 | Edge network node deployment method and system for automobile welding workshop |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113156992B (en) | 2022-07-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113156992B (en) | Three-layer architecture collaborative optimization method for unmanned aerial vehicle in edge environment | |
| CN113395679B (en) | Resource and task allocation optimization system of unmanned aerial vehicle edge server | |
| CN110336861B (en) | Unloading task allocation method of mobile edge computing system based on double-layer unmanned aerial vehicle | |
| Qiao et al. | Collaborative task offloading in vehicular edge multi-access networks | |
| Hu et al. | Ready player one: UAV-clustering-based multi-task offloading for vehicular VR/AR gaming | |
| Samie et al. | Computation offloading and resource allocation for low-power IoT edge devices | |
| CN110766159B (en) | Task allocation method for multi-UAV service edge calculation based on improved genetic algorithm | |
| CN109936619A (en) | A kind of Information Network framework, method and readable storage medium storing program for executing calculated based on mist | |
| CN113377125B (en) | Unmanned aerial vehicle system for air pollution detection | |
| CN111475274A (en) | Cloud collaborative multi-task scheduling method and device | |
| CN113286317B (en) | Task scheduling method based on wireless energy supply edge network | |
| CN109947574A (en) | A kind of vehicle big data calculating discharging method based on mist network | |
| CN110650487A (en) | Internet of things edge computing configuration method based on data privacy protection | |
| CN113507492A (en) | Internet of things three-dimensional space intelligent positioning and distributed edge calculation support method | |
| Yan et al. | Data offloading enabled by heterogeneous UAVs for IoT applications under uncertain environments | |
| Huda et al. | Deep reinforcement learning-based computation offloading in uav swarm-enabled edge computing for surveillance applications | |
| Henna et al. | Distributed and collaborative high-speed inference deep learning for mobile edge with topological dependencies | |
| Zheng et al. | Joint optimization of the deployment and resource allocation of UAVs in vehicular edge computing and networks | |
| CN115134370B (en) | Multi-unmanned aerial vehicle assisted mobile edge computing and unloading method | |
| Chen et al. | Multi-UAV Cooperation Based Edge Computing Offloading in Emergency Communication Networks | |
| CN115334540A (en) | Multi-unmanned aerial vehicle communication system based on heterogeneous unmanned aerial vehicles and energy consumption optimization method | |
| CN116017570A (en) | Edge computing system resource management method based on block chain | |
| Alkhawlani et al. | Hybrid approach for radio network selection in heterogeneous wireless networks | |
| Xu et al. | AI-SPACE: A Cloud-Edge Aggregated Artificial Intelligent Architecture for Tiansuan Constellation-Assisted Space-Terrestrial Integrated Networks | |
| CN116301045B (en) | Unmanned aerial vehicle data acquisition task allocation method oriented to space-time constraint network |
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 |