CN116400737B

CN116400737B - Safety path planning system based on ant colony algorithm

Info

Publication number: CN116400737B
Application number: CN202310645396.4A
Authority: CN
Inventors: 严明; 张羽瑄; 袁慧敏; 雷玲
Original assignee: Communication University of China
Current assignee: Communication University of China
Priority date: 2023-06-02
Filing date: 2023-06-02
Publication date: 2023-08-25
Anticipated expiration: 2043-06-02
Also published as: CN116400737A

Abstract

The invention belongs to the technical field of navigation, in particular relates to a security path planning system based on an ant colony algorithm, and aims to solve the problem that in the prior art, unmanned plane path planning is difficult to apply to complex scenes because paths are generated only through a single objective function. The invention comprises the following steps: acquiring a scene image of a target area through image acquisition equipment; presetting a risk target type and a corresponding risk range, and acquiring a risk target position and a risk target type in an area scene image through an image recognition model so as to construct an environment model; setting a course length cost, an environmental threat cost and a constraint cost, and setting corresponding weights to obtain a path objective function; and generating a safe path from a start point to an end point in the environment model through the path objective function and the improved ant colony algorithm. According to the invention, by setting the cost function, more influencing factors can be considered by the path optimization algorithm, and the adaptability of the path optimization algorithm in a complex environment is improved.

Description

Safety path planning system based on ant colony algorithm

Technical Field

The invention belongs to the technical field of navigation, and particularly relates to a security path planning system based on an ant colony algorithm.

Background

The unmanned aerial vehicle technology has the characteristics of low cost, high efficiency, good flexibility and the like, is favored in a plurality of fields and is widely applied. Unmanned aerial vehicles may assist in performing tasks in marine scenarios such as offshore operations, monitoring marine environmental changes, transporting marine materials, and the like. However, due to factors such as wide ocean area, complex and changeable environment and the like, the application of the unmanned aerial vehicle in the ocean scene faces serious challenges such as multiple material transportation resource waste, low task planning efficiency, poor path obstacle avoidance safety and the like. Most of the existing unmanned plane path planning methods are designed for specific scenes, and the optimized objective function is single, for example, the objective function aims at the minimum total time, so that it is difficult to consider a plurality of influencing factors of a complex scene, and therefore, path planning for the complex scene is a challenge.

In order to solve the path planning problem when the unmanned aerial vehicle executes tasks, modeling is needed to be carried out on the environment terrain and threat sources, and common modeling methods include a Voronoi diagram method, a grid method, a potential field method and the like. For example, to further improve the motion planning efficiency of the fast search random tree (RRT) algorithm, lightweight features are extracted from the generalized Voronoi diagram and a given environment is initialized to ensure that any state in free space can be connected to the feature diagram without any conflicts. In addition, in order to improve the performance index of the route decision, real numbers are inserted in the cell generation process when the model is built, so that the performance index of the decision is improved.

Common path planning methods are mainly divided into two major classes of traditional algorithms (mixed integer linear programming method, artificial potential field method (Artificial Potential Field, APF), dijkstra algorithm, voronoi diagram method, a-algorithm, D-algorithm and the like), intelligent optimization algorithms (genetic algorithm, ant colony algorithm, particle swarm algorithm, simulated annealing algorithm (SimulatedAnnealing, SA), artificial bee colony algorithm (Artificial Bee Colony Algorithm, ABC) and the like). For example, in order to solve the problem of path planning of unmanned aerial vehicle under multiple threats in a complex environment, a new PSO optimization algorithm based on spherical vectors is proposed. Firstly, a cost function is formulated to convert path planning into an optimization problem, then an effective configuration space is searched by using the algorithm to find an optimal path with minimized cost function, and then the unmanned aerial vehicle is controlled through constraint conditions. In order to accomplish the agricultural mission with the least unmanned aerial vehicle and without revisiting, the VegNet software is used for identifying the compressed areas, and then the optimal path connecting all the compressed areas and the injection points thereof is found out based on the improved Voronoi graph algorithm. In order to realize path planning of unmanned aerial vehicle tracking ground moving targets, a novel dynamic APF path planning technology is also provided, and the defects of hardware dependence, high calculation requirement, incapability of processing obstacles and the like of the multi-rotor unmanned aerial vehicle are overcome.

Disclosure of Invention

In order to solve the above-mentioned problems in the prior art, that is, the path generation system in the prior art generates a path only through a single objective function, and is difficult to apply to a complex scene, the present invention provides a security path planning system based on an ant colony algorithm, the system includes: the system comprises a scene acquisition module, a risk area and environment parameter acquisition module, an environment model construction module to be planned, an objective function construction module and a path planning module;

the scene acquisition module is configured to acquire a scene image of the target area through the image acquisition equipment;

the risk region and environment parameter acquisition module is configured to preset a risk target type and a corresponding risk range, and acquire a risk target position and a risk target type in a region scene image through the image recognition model;

the environment model construction module to be planned is configured to construct an environment model based on the target area scene image and the risk target position and the risk target type; the environment model comprises a plurality of risk target positions, risk target types, risk ranges and relations among the risk target ranges; the relationship between the risk target ranges includes: phase separation, phase tangency, coincidence and continuity;

The objective function construction module is configured to set a course length cost, an environmental threat cost and a constraint cost, and set corresponding weights to acquire a path objective function;

and the path planning module is configured to generate a safe path from a start point to an end point in the environment model through the path objective function and the improved ant colony algorithm.

In some preferred embodiments, the path planning system is applied to unmanned aerial vehicle path planning;

the environment model also comprises a three-dimensional environment model, and the three-dimensional environment model construction method comprises the following steps:

acquiring a depth image of a target area scene through a depth camera;

assigning a height value to the target area scene image through the depth image to obtain a target area scene image with height information;

setting a quasi-circular initial environment model based on the target area scene image with the height information:

wherein ,representing the height of the local topography +.>Represent the firstmCenter coordinates of the highest point of the local topography, < >> and />Represent the firstmLocal terrain gradient related factors-> and />Representing a horizontal plane coordinate value in the target area;

performing convolution interpolation for preset times on the initial environment model to obtain an interpolation environment model, performing smoothing treatment on the interpolation environment model, and setting a safe height interval to obtain a three-dimensional environment model;

Smoothing the interpolation environment model and setting a safe height interval, wherein the smoothing comprises the following steps: smoothing the interpolation environment model, and setting a first stepkFlight safety height section of personal unmanned aerial vehicle, wherein />Representing a minimum safe flight level, +.>Representing a maximum safe flight altitude; -providing said minimum safe flying height +.>Performing superposition processing on the smooth curved surface in the interpolation environment model after the smoothing processing to obtain a smooth unmanned plane flight safety curved surface +.>；

Obtaining smooth unmanned plane flight safety curved surfaceAnd adding a risk range and setting flight constraint conditions according to the risk target position by the post interpolation environment model to obtain the three-dimensional environment model.

In some preferred embodiments, the adding risk ranges and setting flight constraints specifically include:

on smooth unmanned aerial vehicle flight safety curved surfaceIn the post interpolation environment model, a threat zone model is established according to the risk target position and the risk range in the regional scene image:

wherein ,represent the firstkCoordinates of the unmanned aerial vehicle->Represent the firstsCenter point coordinates of the threat zones, +.>Represent the firstsThe radius of the individual threat zones,srepresenting threat zone sequence numbers;

the flight constraint conditions comprise maximum horizontal turning angle constraint, maximum pitch angle constraint, maximum course constraint and minimum inertial distance constraint;

The maximum horizontal turning angle constraint is specifically as follows:

wherein ,representing unmanned plane->At node-> and />Horizontal projection of voyage between +.>Indicating the horizontal turning angle->Indicating the maximum horizontal turning angle,irepresenting the sequence number of the node,krepresenting the serial number of the unmanned aerial vehicle;

the maximum pitch angle constraint is specifically as follows:

represents pitch angle, +.>Represents maximum pitch, +.>Represent the firstiThe coordinates of the individual nodes are used to determine,represent the firsti-Coordinates of 1 node;

the maximum voyage constraint is specifically as follows:

wherein ,representing unmanned plane->Is used for the purpose of the maximum range of (1),Nrepresenting the number of nodes in the overall path,N-1 is the number of segments of the path, ">Representing unmanned plane->Is the first of (2)iVoyage corresponding to each node->Representing unmanned plane->Is a general course of (1);

the minimum inertial distance constraint is specifically as follows:

wherein Representing unmanned plane->Is provided for the minimum inertial distance of (a).

In some preferred embodiments, the path objective function comprises:

the course length cost is:

wherein ,the cost of the path length is represented by,Rindicate path, & lt>Represent the firstiThe coordinates of the individual nodes are used to determine,coordinates +.>Coordinates to the target point->Is a straight line distance of (2);

the environmental threat costs are:

wherein ,representing unmanned plane- >Is>Indicating that the unmanned plane is at the firstiThe minimum inertia distance is considered on each node and then is the same as that of the first nodesVolume of area where the threat zones coincide, +.>Representing environmental threat cost which represents that unmanned plane is on pathRThe probability of collision to a threat zone after the minimum inertial distance is considered in the flight process of (2);

the constraint cost is:

wherein [ (S)]The value of the value is expressed,representing a safe height interval +.>Average value of>Representing the actual flying height deviation +.>Representing constraint costs;

cost the journey lengthEnvironmental threat cost->And constraint cost->Normalizing and scaling to obtain a path objective function:

wherein ,、/>、/>respectively are provided withRepresenting normalized voyage Length cost +.>Cost of environmental threatAnd constraint cost->Weight coefficient of (c) in the above-mentioned formula (c).

In some preferred embodiments, the improved ant colony algorithm specifically includes:

step A001, initializing ant colony algorithm parameters, and setting coordinates of a departure point and coordinates of a target point;

step A002, based on the environment model, making all ants start from the coordinates of the starting point;

step A003, searching the rest transferable node set for each ant, and according to the current pheromone concentration between different nodes Calculating transition probability->Distributing and moving all ants in the ant colony to respective next nodes according to the transition probability, and calculating the pheromone concentration of the next iteration round in a Logistic chaotic mapping modeThe method comprises the steps of carrying out a first treatment on the surface of the Transition probability->The method comprises the steps of calculating and obtaining a reference journey length cost, an environmental threat cost and a constraint cost;

step A004, repeating the method of step A003 for all ants which do not reach the coordinates of the target point;

step A005, calculating the path and path length of each ant reaching the coordinates of the target point；

Step A006, based on the path and path length of each antUpdating the optimal solution Gbest through an SA screening mechanism;

step A007, adjusting the pheromone concentration among all nodes left by the ant colony through a bipolar differentiation pheromone recording rule;

and A008, repeating the steps of the steps A003-A007, and iterating according to the current temperature in the SA screening mechanism and the length of the Metropolis chain until the temperature in the SA screening mechanism reaches the preset termination temperature to output the optimal solution.

In some preferred embodiments, the initializing ant colony algorithm parameters specifically includes:

initializing ant colony numbersIteration number->System initial temperature- >And termination temperature->The unmanned aerial vehicle performance parameters and the initial pheromone concentration.

In some preferred embodiments, the pheromone concentration of the next iteration round is calculated by means of Logistic chaotic mappingThe method specifically comprises the following steps:

by slave nodesiTo the point ofjChaotic sequence of current iteration turns of (a)Calculating the pheromone concentration of the next iteration round:

wherein ,representing slave nodesiTo the point ofjPheromone concentration of the next iteration round of (a),. About.>Representing slave nodesiTo the point ofjCurrent pheromone concentration,/->Is slave node in unit timeiTo the point ofjPheromone concentration,/->Representing the pheromone volatilization factor->Representing an adjustable parameter;

wherein ,the method comprises the following steps:

initial valueA random number between 0 and 1, and +.>，Representing slave nodesiTo the point ofjThe chaotic sequence of the previous iteration round of (a),μrepresents chaos factor, and->When->，/>And then the chaotic state is entered.

In some preferred embodiments, the information is based on the current pheromone concentrationCalculating transition probabilitiesThe method comprises the following steps:

wherein ,representing ant->Slave nodeiTo the nodejProbability of transition->The information elicitation factor is represented as a function of,representing the desired heuristic->Representing heuristic functions represents ant slave nodesiTransfer to nodejIs used as a reference to the desired value of (a), Representing nodesjBelonging to ant->Is provided for the remaining set of transferable nodes,trepresenting the current iteration round,representing nodesiTo the nodejStraight line distance between>Representing nodesiTo the nodejBetween not entering threat zonesAnd keep the minimum inertial distance +.>The minimum curve distance after that;

judgment nodeiTo the nodejThe number of threat zones passed;

if nodeiTo the nodejThrough a single threat zone, nodesiSum nodejAll add minimum inertial distance to threat zoneOutside the range of>Approximately as a nodeiAnd to the corresponding tangent point->Distance between->NodejAnd to the corresponding tangent point->Distance between->And tangential point->And (2) tangential point->Arc length therebetweenDegree->The minimum curve distance for the case of a single threat zone is calculated as follows:

wherein , and />The method comprises the following steps:

nodeiCorresponding tangent pointAnd nodejCorresponding tangential point->The arc length between +.>Through an included angleThe calculation method comprises the following steps:

wherein ,，/> and />Respectively nodesiAnd the center of a circle and a node of a threat zonejAnd the center of a circle and a node of a threat zoneiSum nodejThe distance between the two is obtained through coordinate conversion;

representing nodesiCorresponding tangential point->And nodejCorresponding tangential point->Included angle between->Representing nodesiAnd corresponding tangent point- >Included angle between->Representing nodesjAnd corresponding tangent point->An included angle between the two.

In some preferred embodiments, if node i to node j pass through a plurality of threat zones, determining whether there is an overlap region between the plurality of threat zones passed through;

if multiple threat zones pass through, adding minimum inertial distanceWith overlapping areas at the back->The calculation method of (1) is as follows:

judgment nodeiSum nodejWhether the connected straight line is positioned at the same side of the center of the first overlapping threat zone and the center of the second overlapping threat zone; if the two overlapping threat areas are on the same side, searching the outer common tangents of the first overlapping threat area and the second overlapping threat area, and selecting the midpoints of the two outer common tangentsmComputing nodeiTo the midpointmAvoiding threat zones therebetweensMinimum curve distance after 1And a midpointmTo the nodejAvoiding threat zones therebetweensMinimum curve distance after 2->The method comprises the steps of carrying out a first treatment on the surface of the If on different sides, at the nodeiTo the midpointmOn arcs or nodes betweenjTo the midpointmAuxiliary points are added on the circular arcs of the (E) to enable the nodes to beiOr nodejCarrying out multistage calculation by taking an included angle formed by the circle centers of the corresponding threat areas as an acute angle to the auxiliary point;

if multiple threat zones pass through, adding minimum inertial distanceRear non-overlapping area>The calculation method of (1) is as follows:

judgment node iSum nodejWhether the connected straight line is positioned at the same side of the center of the first non-overlapping threat zone and the center of the second non-overlapping threat zone; if at the same side, searching the outside of the first overlapping threat zone and the second overlapping threat zoneThe common tangent line is selected from the midpoints of the two external common tangent linesmComputing nodeiTo the midpointmAvoiding threat zones therebetweensMinimum curve distance after 1And a midpointmTo the nodejAvoiding threat zones therebetweensMinimum curve distance after 2->The method comprises the steps of carrying out a first treatment on the surface of the If the nodes are located at different sides, searching internal common tangent lines of the first non-overlapping threat area and the second non-overlapping threat area, selecting two midpoints of the internal common tangent lines, and calculating the nodesiTo the midpointmAvoiding threat zones therebetweensMinimum curve distance after 1->And the midpoint ofmTo the nodejAvoiding threat zones therebetweensMinimum curve distance after 2->And (3) summing.

In some preferred embodiments, the adjusting the pheromone concentration between the nodes left by the ant colony according to the bipolar differentiation pheromone recording rule specifically includes:

based on the path traversed and path lengthSorting ants in the ant colony according to the sequence from small to large, and dividing the ants into a first section, a second section and a third section according to the sorting;

setting that ants in the first interval can leave pheromones in the first interval, and setting the weight of the pheromones as the product of the inverse ratio of the path length of the ants finding food and the ranking factor; the ranking factor is that only the tail sequence number in the first section is reduced to a value from the initial sequence number in the first section in the direction of 1 to 0;

Resetting the pheromone left by the ant passing path in the second section;

the ants in the third section are set to mask the pheromone concentration information in the next iteration round, and the direction generation path is judged only by the heuristic information.

The invention has the beneficial effects that:

(1) According to the method for generating the path of the unmanned aerial vehicle in the complex scene based on the improved ant colony algorithm, provided by the invention, through setting the cost function, the path optimization algorithm can consider more influencing factors, and the adaptability of the path optimization algorithm in a complex environment is improved.

(2) The method for generating the complex scene unmanned aerial vehicle path based on the improved ant colony algorithm improves the algorithm by improving the traditional ACO algorithm and introducing methods such as chaotic mapping, bipolar differentiation pheromone recording rules, screening mechanisms and the like. The unmanned aerial vehicle path planning algorithm is more efficient and stable.

(3) The method and the system optimize path planning of the unmanned aerial vehicle in complex scenes such as ocean, consider the characteristics of the complex scenes, the requirements of the unmanned aerial vehicle for executing complex scene tasks and the characteristics of different algorithms, and select ant colony optimization algorithms (Ant Colony Optimization, ACO) in the intelligent swarm algorithm to improve and optimize so as to realize path planning of the unmanned aerial vehicle. In the simulation experiment, the improved algorithm has certain superiority by comparing the scheme results, objective function values, iteration conditions and other parameters of other traditional algorithms.

Drawings

Other features, objects and advantages of the present application will become more apparent upon reading of the detailed description of non-limiting embodiments, made with reference to the accompanying drawings in which:

fig. 1 is a block diagram of a security path planning system based on an ant colony algorithm according to an embodiment of the present disclosure;

fig. 2 is a schematic diagram of an improved ant colony algorithm according to an embodiment of the present disclosure;

FIG. 3 is a schematic view illustrating the effect of an environmental model according to an embodiment of the present disclosure;

fig. 4 is a schematic diagram of adjusting concentration of an ant colony remaining pheromone according to a bipolar differentiation pheromone recording rule according to an embodiment of the present disclosure;

fig. 5 is a schematic diagram of updating an optimal solution Gbest by an SA filtering mechanism according to an embodiment of the present disclosure;

fig. 6 is a perspective view of five algorithm comparison implementation results in single unmanned aerial vehicle path planning according to an embodiment of the present disclosure;

fig. 7 is a top view of five algorithm comparison implementation results in single unmanned aerial vehicle path planning according to an embodiment of the present disclosure;

fig. 8 is a left view of five algorithm comparison implementation results in single unmanned aerial vehicle path planning according to an embodiment of the present disclosure;

FIG. 9 is a schematic diagram of minimum curve path computation through a single threat zone in a single drone path plan provided in an embodiment of the present disclosure;

Fig. 10 is a schematic diagram of minimum curve path computation for a plurality of threat zones in unmanned aerial vehicle path planning, according to an embodiment of the present disclosure.

Detailed Description

The application is described in further detail below with reference to the drawings and examples. It is to be understood that the specific embodiments described herein are merely illustrative of the application and are not limiting of the application. It should be noted that, for convenience of description, only the portions related to the present application are shown in the drawings.

It should be noted that, without conflict, the embodiments of the present application and features of the embodiments may be combined with each other. The application will be described in detail below with reference to the drawings in connection with embodiments.

The application provides a security path planning system based on an ant colony algorithm, which comprises: the system comprises a scene acquisition module, a risk area and environment parameter acquisition module, an environment model construction module to be planned, an objective function construction module and a path planning module.

In order to more clearly describe a security path planning system based on the ant colony algorithm of the present application, each functional module in the embodiment of the present application is described in detail below with reference to fig. 1.

The invention relates to a security path planning system based on an ant colony algorithm, which comprises a scene acquisition module, a risk area and environmental parameter acquisition module, an environmental model construction module to be planned, an objective function construction module and a path planning module, wherein the detailed description of each functional module is as follows:

in this embodiment, the path planning system is applied to unmanned plane path planning;

Acquiring a depth image of a target area scene through a depth camera; information such as an meteorological image can be marked in the target area scene image according to the requirement.

The present embodiment assumes a 10km as the original environmental model that contains peaks, flat ground, waves at sea, reefs and air flow, the environmental model being generally circular-like due to the particularities of complex scenes.

wherein ,representing the height of the local topography +.>Represent the firstmCenter coordinates of the highest point of the local topography, < >> and />Represent the firstmLocal terrain gradient related factors-> and />Representing a horizontal plane coordinate value in the target area; the terrain gradient related factor is expressed by taking a terrain cross section at a certain height, and an ellipse can be obtained, and the length of the major axis of the ellipse is expressed as +.>The length of the minor axis is denoted->，/> and />The larger the terrain is, the flatter the terrain is, and the steeper the reverse is, and the simulation of various terrains in an offshore environment can be realized by adjusting the local terrain horizontal plane coordinate value, the height of the local terrain, the center coordinate of the highest point of the local terrain and the terrain gradient correlation factor.

and preferably carrying out 3 times of convolution interpolation on the convolution interpolation with preset times to obtain an interpolation environment model. The topography fluctuation in the interpolation environment model is large, and compared with the actual environment, the interpolation environment model is too sharp as shown in (a) of fig. 3, the actual environment cannot be closely simulated, and the topography gradient and curvature are limited to realize the curved surface smoothing treatment as shown in (b) of fig. 3; meanwhile, considering that the unmanned aerial vehicle has the maximum horizontal turning angle, the maximum climbing angle and the unmanned aerial vehicle performance limit in the flight process, in order to ensure the safety of the path obstacle avoidance and the high efficiency of completing the task in the unmanned aerial vehicle flight process, the unmanned aerial vehicle ground flight safety height interval is required to be set as shown in (c) in fig. 3.

Smoothing the interpolation environment model and setting a safe height interval, wherein the smoothing comprises the following steps: smoothing the interpolation environment model, and setting a first stepkFlight safety height section of personal unmanned aerial vehicle, wherein />Representing a minimum safe flight level, +. >Representing a maximum safe flight altitude; -providing said minimum safe flying height +.>Performing superposition processing on the smooth curved surface in the interpolation environment model after the smoothing processing to obtain a smooth unmanned plane flight safety curved surface +.>；

Obtaining smooth unmanned plane flight safety curved surfaceAdding a risk range and setting flight constraints according to the risk target position by the post interpolation environment modelThe condition is a three-dimensional environment model. All unmanned aerial vehicle path planning of this embodiment is all carried out based on safe curved surface to obtain safe efficient unmanned aerial vehicle path.

Common risk targets are threat bodies that clearly threat boundaries, such as peaks, reefs, ocean currents, etc. Once the unmanned aerial vehicle touches the boundaries of the threat bodies, collision is generated to damage the unmanned aerial vehicle, the planned route of the path needs to avoid the boundary coordinates of the threat bodies, and the unmanned aerial vehicle is ensured to fly safely. The risk target types mainly comprise mountain peaks, reefs, ocean currents, no-fly areas and the like, and threat models of the mountain peaks are built through the initial complex environment models. Several other threats appear spheroid. Once the unmanned aerial vehicle touches the areas, crash occurs, namely the damage probability of the unmanned aerial vehicle in the areas is 1. In the path planning process, discrete unmanned plane nodes are usually planned first, and then the nodes are connected in a straight line or a curve to form a complete flight path. Therefore, it is necessary to ensure that the complete path of the unmanned aerial vehicle does not involve the threat zone, rather than merely determining that the nodes of the unmanned aerial vehicle do not involve the threat zone. The number of nodes should be increased as much as possible to ensure that the unmanned aerial vehicle can successfully avoid the threat zone. The nodes described in this embodiment are the flight path points of the unmanned aerial vehicle.

In this embodiment, the adding a risk range and setting a flight constraint condition specifically includes:

the maximum horizontal turning angle constraint means that the unmanned aerial vehicle cannot complete a large-angle turning action in the horizontal direction and can only turn within a certain angle range.

The maximum horizontal turning angle constraint is specifically as follows:

the maximum pitch angle refers to the range of angular transformation that the unmanned aerial vehicle can perform in the vertical direction.

The maximum pitch angle constraint is specifically as follows:

the maximum voyage constraint is specifically as follows:

the unmanned aerial vehicle needs to avoid threat areas to adjust the gesture in actual flight, and the gesture adjustment needs a certain buffer time due to inertia effect, so that the flight state cannot be changed rapidly. Therefore, in order to ensure that the unmanned aerial vehicle can fly safely according to the planned path, it is necessary to fly in an as-is state for a distance, which is called the minimum inertial distance, before the unmanned aerial vehicle is ready to change the flight angle, adjust the attitude, and the like.

The minimum inertial distance constraint is specifically as follows:

In three-dimensional space, the unmanned aerial vehicle generally has a plurality of flyable paths between a departure point and a destination point, and the purpose of path planning is to select an optimal path from the plurality of possible paths. Therefore, a corresponding cost function is assigned to each path, an objective function of the path is formulated according to the cost function, and the objective function is used as a standard for evaluating the advantages and disadvantages of the unmanned aerial vehicle path. The cost function is generally divided into a path length cost, an environmental threat cost, and a constraint cost, and the objective function is generally a weighted sum of the cost functions.

in this embodiment, the path objective function specifically includes:

in order to avoid threat areas and bypass obstacles in the flight process of the unmanned aerial vehicle, the final flight path is not a straight line connecting a departure point and a target point, but a curve fitting the environment, and the cost of the length of the flight pathMeans that the curved path has a multiple-loss path length than the straight path;

the course length cost is:

wherein ,the cost of the path length is represented by,Rindicate path, & lt>Represent the firstiCoordinates of individual nodes>Coordinates +.>Coordinates to the target point->Is a straight line distance of (2); in this embodiment, the coordinates of the departure point and the coordinates of the target point may be expressed in the form of vectors;

the environmental threat cost and the voyage length cost are related to each other and the relationship between the two is eliminated. In order to avoid an environmental threat zone in the unmanned aerial vehicle flight process, the range length is often increased. But decreasing the voyage length increases the risk of collision with the environmental threat zone. Therefore, when planning the path, the influence proportion of the two needs to be comprehensively considered according to the actual situation. The environmental threat cost means that the unmanned aerial vehicle generates a certain potential risk when approaching the environmental terrain and the threat zone, and the risk is larger when the distance is closer.

The environmental threat costs are:

wherein ,representing unmanned plane->Is>Indicating that the unmanned plane is at the firstiThe minimum inertia distance is considered on each node and then is the same as that of the first nodesVolume of area where the threat zones coincide, +.>Representing environmental threat cost which represents that unmanned plane is on pathRThe probability of collision to a threat zone after the minimum inertial distance is considered in the flight process of (2);

the constraint condition must be satisfied when the path planning is performed, so that the absolute value of each parameter of the path deviating from the constraint condition on the premise that the constraint condition is satisfied is called constraint cost.

The constraint cost is:

wherein ,、/>、/>respectively represent normalized journey length cost +.>Cost of environmental threatAnd constraint cost->Weight coefficient of (c) in the above-mentioned formula (c). Finding +.>、/> and />The value of (2) is not an order of magnitude and will therefore +.>Reduced by 1000 times, will-> and />Amplified by a factor of 100 to reach a suitable level. The smaller the objective function value, the more efficient the planned path is represented, and the safer the unmanned aerial vehicle flies.

And the path planning module is configured to generate a safe path from a start point to an end point in the environment model through the path objective function and the improved ant colony algorithm. In this embodiment, the improved ant colony algorithm improves the ACO algorithm in the three stages of the early, middle and late stages of the solving process. The randomness of the initial state of the initial-stage enhancement pheromone is realized, the guiding function of the pheromone is enhanced in the iterative process, and the risk of sinking into a local optimal solution is reduced after each generation of optimal solution is obtained.

The ACO algorithm is an algorithm for simulating the cooperative foraging process of the ant colony in nature, which is proposed by Marco Dorigo doctor. Ants can leave pheromone on a walking path in the foraging process, and ant colony judges the advancing direction according to the pheromone to finally find food. Factors influencing the advancing direction of ants are mainly: pheromone concentration on the path and heuristic information of the current node. Pheromones may volatilize gradually over time, and the concentration of pheromones may affect the probability of other peers selecting the route. The higher the pheromone concentration is, the more ants selecting the path, the higher the pheromone concentration is left by the more ants, and the circulation forms a positive feedback mechanism, so that the shortest path for acquiring food is found, namely, the ants are used for simulating the unmanned aerial vehicle.

In this embodiment, the improved ant colony algorithm specifically includes:

in this embodiment, the initializing ant colony algorithm parameters specifically includes:

initializing ant colony numbersIteration number->System initial temperature->And termination temperature->The unmanned aerial vehicle performance parameters and the initial pheromone concentration.

The ACO algorithm is easy to generate the problems of high trial and error times, low search efficiency, easy sinking into a local optimal solution and the like due to the following reasons: the initialization concentration of the first pheromone and the initialization concentration of the pheromone are balanced, heuristic information is excessively depended on when a path is searched for in the first iteration, so that the types of solutions are single, and the subsequent iteration direction is influenced. Secondly, the error path passed by ants also leaves pheromone, so that the number of trial and error is increased, and the ant colony searching efficiency is reduced. Thirdly, the selection of the paths has randomness, and the paths are easy to fall into a local optimal solution under the limited iteration times.

For the three reasons above, this section is modified by the following strategies: and initializing and updating the pheromone by adopting the Logistic chaotic map when the prior initialization is prepared. In the iterative process, a bipolar differentiation rule is used to record the pheromone. After the local optimal solution is obtained, an SA mechanism is introduced to screen and update the local optimal solution.

step A003, searching the rest transferable node set for each ant, and according to the current pheromone concentration between different nodesCalculating transition probability->Distributing and moving all ants in the ant colony to respective next nodes according to the transition probability, and calculating the pheromone concentration of the next iteration round in a Logistic chaotic mapping modeThe method comprises the steps of carrying out a first treatment on the surface of the Transition probability->The method comprises the steps of calculating and obtaining a reference journey length cost, an environmental threat cost and a constraint cost;

chaos is a form of nonlinear dynamic motion that is widely found in nature, and is a random situation that exists in deterministic systems. The chaotic system has the characteristics of complex characteristics, long-term unpredictability, overall stability, local instability, pseudo-randomness and the like. In order to solve the problem of single type of solution caused by the concentration balance of the initialization of the pheromone in the ACO algorithm, a logic chaotic mapping improved pheromone concentration calculation equation is introduced. The Logistic chaotic map is a classical chaotic system and is a random sequence generated by a simple deterministic system.

In this embodiment, the ant colony searches for an optimal path in space through a plurality of iterative traversals, and leaves a certain amount of pheromones on the walked path. The pheromone concentration consists of two parts, wherein the first part is the residual amount of the pheromone which is not volatilized after the previous generation ant colony is traversed, and the second part is the increment of the pheromone which is left on each path by the ant colony after the current traversing.

In this embodiment, the method of Logistic chaotic mapping calculates the pheromone concentration of the next iteration roundThe method specifically comprises the following steps:

wherein ,the method comprises the following steps: />

In the present embodiment, the information element concentration is determined according to the current information element concentrationCalculating transition probability->The method comprises the following steps:

wherein ,representing ant->Slave nodeiTo the nodejProbability of transition->The information elicitation factor is represented as a function of,representing the desired heuristic->Representing heuristic functions represents ant slave nodesiTransfer to nodejIs used as a reference to the desired value of (a),representing nodesjBelonging to ant->Is provided for the remaining set of transferable nodes,trepresenting the current iteration round, Representing nodesiTo the nodejStraight line betweenDistance (L)>Representing nodesiTo the nodejBetween not entering threat zonesAnd keep the minimum inertial distance +.>The minimum curve distance after that;

judgment nodeiTo the nodejThe number of threat zones passed;

if nodeiTo the nodejPassing through a single threat zone, as shown in fig. 9 (a), the nodeiSum nodejAll add minimum inertial distance to threat zoneOutside the range of>Approximately as a nodeiAnd to the corresponding tangent point->Distance between->NodejAnd to the corresponding tangent point->Distance between->And tangential point->And (2) tangential point->The arc length between +.>As shown in fig. 9 (b), the minimum curve distance for the case of passing through a single threat zone is calculated as follows:

wherein , and />The method comprises the following steps: />

representing nodesiCorresponding tangential point->And nodejCorresponding tangential point->Included angle between->Representing nodesiAnd corresponding tangent point->Included angle between- >Representing nodesjAnd corresponding tangent point->An included angle between the two.

In this embodiment, if the node i to the node j pass through a plurality of threat areas, determining whether there is an overlapping area between the plurality of threat areas passing through;

if multiple threat zones pass through, adding minimum inertial distanceThere is an overlap region behind, as shown in figure 10 (a),the calculation method of (1) is as follows:

if multiple threat zones pass through, adding minimum inertial distanceThe latter non-overlapping region, as shown in figure 10 (b),the calculation method of (1) is as follows:

Judgment nodeiSum nodejWhether the connected straight line is positioned at the same side of the center of the first non-overlapping threat zone and the center of the second non-overlapping threat zone; if the two overlapping threat areas are on the same side, searching the outer common tangents of the first overlapping threat area and the second overlapping threat area, and selecting the midpoints of the two outer common tangentsmComputing nodeiTo the midpointmAvoiding threat zones therebetweensMinimum curve distance after 1And a midpointmTo the nodejAvoiding threat zones therebetweensMinimum curve distance after 2->The method comprises the steps of carrying out a first treatment on the surface of the If the nodes are located at different sides, searching internal common tangent lines of the first non-overlapping threat area and the second non-overlapping threat area, selecting two midpoints of the internal common tangent lines, and calculating the nodesiTo the midpointmAvoiding threat zones therebetweensMinimum curve distance after 1->And the midpoint ofmTo the nodejAvoiding threat zones therebetweensMinimum curve distance after 2->And (3) summing.

in this embodiment, the SA algorithm introduces random factors in the search process, and when iteratively updating the optimal solution, accepts other inferior solutions with a certain probability to prevent the solution from falling into a local optimal solution, as shown in fig. 5, assuming an initial solution AIterating from the leftmost to the right, when the local optimal solution is searchedBAt this point, the search is continued to the left with a certain probability without selecting the end. Continuing the iterative search according to the rule, the local optimal solution is searchedCAfter several rounds of search to jump out the local optimal solution, the global optimal solution can be found outD。

In this embodiment, updating the optimal solution Gbest through the SA filtering mechanism specifically includes:

computing probability of accepting new solution following Metropolis criterionThe method comprises the following steps:

wherein ,the probability of accepting a new solution is expressed as the temperatureTWhen the energy difference isdEIs used for controlling the cooling probability of the air conditioner,Trepresenting the initial temperature, the update mode of the solution is composed ofdEIf the system energy corresponding to the new solution is lower, the new solution is fully accepted, otherwise, the new solution is added with +.>Is accepted by the new solution.

conventional ACO algorithms provide that each ant follows a path that leaves a pheromone, and that the concentration of pheromone left by an ant is inversely proportional to the path length that it follows. The concentration distribution of the pheromones on the map is too much, the pheromones left by the ants with ranking can cause certain misleading, the pheromones left by the ants with ranking front have submerged probability, and the iterative search efficiency is finally affected.

The bipolar differentiation pheromone recording rule is mainly used for breaking the rule of equally recording each ant pheromone in the traditional algorithm and providing the idea of bipolar differentiation treatment.

In this embodiment, the adjusting the pheromone concentration between the nodes left by the ant colony according to the bipolar differentiation pheromone recording rule, as shown in fig. 4, specifically includes:

based on the path traversed and path lengthSorting ants in the ant colony according to the sequence from small to large, and dividing the ants into a first section, a second section and a third section according to the sorting; for example, the first 20% of ants are set as a first interval, the middle 60% of ants are set as a second interval, and the last 20% of ants are set as a third interval; />

resetting the pheromone left by the ant passing path in the second section;

By the method of the embodiment, verification is performed in a simulation experiment mode, multiple comparison experiments are performed under the same environment and problem background with the traditional path planning algorithm, and set simulation environment parameters are shown in table 1:

TABLE 1 simulation experiment environment parameter table

Index (I)	Parameters (parameters)
		Simulation platform	MATLAB R2016a
Operating system	Windows 10
		Notebook computer	HUAWEI Matebook14
Processor and method for controlling the same	AMD Ryzen 5 4600H with Radeon Graphics @3.00 GHz
		RAM	16GB
System type	64 bits

The comparison experiment establishes aThe initial complex scene model of the system is used for simulating the distribution situation of the terrain and the threat zone of the complex environment and displaying the path planning result. The unmanned aerial vehicle warehouse coordinates are set to (6.00,8.00,4.50). Table 2 is a path planning threat zone parameter table, table 3 is a drone performance parameter table, and table 4 is a modified ACO algorithm initial parameter table.

TABLE 2 Path planning threat zone parameter Table

Attributes of	Coordinates of
		Threat zone 1	(4.14,2.81,5.80,0.80)
Threat zone 2	(6.28,4.28,5.60,1.00)
		Threat zone 3	(2.00,5.00,4.40,0.80)
Threat zone 4	(1.80,1.60,4.20,1.00)
		Threat zone 5	(8.40,2.00,4.20,1.20)
Threat zone 6	(5.60,1.00,5.20,0.60)

Table 3 path planning unmanned aerial vehicle performance parameter table

Attributes of	Numerical value
		Number of unmanned aerial vehicle	5 frames
High flying speed	[0.5,5]km/min
		Maximum horizontal turning angle	60°
Maximum pitch angle	60°
		Maximum range	75km
Minimum inertial distance	0.2km
		Safe flight altitude interval	[0.4,0.8]km

Table 4 modified ACO algorithm initial parameter table

Attributes of	Numerical value
		Number of ants	100 pieces of
Number of iterations	200 times
		Information factor	3
Heuristic factor	5
		Total amount of pheromone	150
Initial temperature of system	90℃
		Termination temperature	10℃
Coefficient of objective function	λ ₁ =0.4，λ ₂ =0.2，λ ₃ =0.4
		Number of runs	50 times

The threat zone parameters are randomly set according to the simulated offshore environment, the performance parameters of the unmanned aerial vehicle refer to the existing related documents, and the improved ACO algorithm parameters are set with better algorithm performance in multiple operations.

And setting the coordinates of the task points to be (2.00,0.50,4.50) respectively, wherein the flight speeds of the unmanned aerial vehicle are 5km/min, and planning the flight path of the unmanned aerial vehicle from a warehouse to reach the task points. The differences between the improved ACO algorithm and the traditional ACO, APF, SA and ABC algorithms are verified in terms of path planning scheme, iterative process, objective function performance, etc., respectively. The two groups of experiments are independently operated for 50 times to obtain an optimal solution, the results of the experiments are shown in fig. 6, 7 and 8, and the path planning results are shown in table 5;

table 5 table unmanned aerial vehicle path planning two sets of experimental results summary tables

Algorithm	Curve path length	Straight line distance	Path break	Objective function value
					Improved ACO	9.24	8.50	8.71%	5.03
ACO	9.52	8.50	12.00%	5.38
					APF	9.38	8.50	10.35%	5.32
ABC	9.76	8.50	14.82%	5.82
					SA	9.94	8.50	16.94%	6.19

Fig. 6, 7 and 8 are perspective, top and left views, respectively, of path planning. Paths planned by the improved ACO algorithm, the APF algorithm, the ABC algorithm and the SA algorithm are respectively represented by different line types. As can be seen by comparison, the improved ACO path is relatively smooth, the degree of change in the horizontal turning and pitch angles of the overall path is small, and a moderate safe flight distance can be maintained from the threat zone. The ACO path is smooth as a whole, the situation that the pitch angle fluctuation is large once occurs at the junction of the terrain and the threat zone, and a certain safe flight distance can be kept between the ACO path and the threat zone. The APF path is relatively smooth, the degree of change in the horizontal turning angle and pitch angle of the overall path is small, and the safe flight distance from the threat zone is relatively small. The ABC path is relatively smooth when the turning amplitude is smaller, the situation that the degree of change of the horizontal turning angle and the pitch angle is larger easily occurs at the turning position, and the safe flight distance between the ABC path and a threat zone is smaller. The fluctuation times of the SA path are relatively more, the degree of change of the horizontal turning angle and the pitch angle of the whole path is smaller, and the safe flight distance between the SA path and a threat zone is relatively larger.

The technical scheme of the embodiment carries out optimal design on the path planning of the unmanned aerial vehicle in complex scenes such as ocean. Considering the characteristics of complex scenes, the requirements of unmanned aerial vehicle for executing complex scene tasks and the characteristics of different algorithms, the ant colony optimization algorithm (Ant Colony Optimization, ACO) in the intelligent group algorithm is selected to be improved and optimized so as to realize the path planning of the unmanned aerial vehicle. In the simulation experiment, the improved algorithm has certain superiority by comparing the scheme results, objective function values, iteration conditions and other parameters of other traditional algorithms.

Those of skill in the art will appreciate that the various illustrative modules, method steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, computer software, or combinations of both, and that the program(s) corresponding to the software modules, method steps, may be embodied in Random Access Memory (RAM), memory, read Only Memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, removable disk, CD-ROM, or any other form of storage medium known in the art. To clearly illustrate this interchangeability of electronic hardware and software, various illustrative components and steps have been described above generally in terms of their functionality. Whether such functionality is implemented as electronic hardware or software depends upon the particular application and design constraints imposed on the solution. Those skilled in the art may implement the described functionality using different approaches for each particular application, but such implementation is not intended to be limiting.

The terms "first," "second," and the like, are used for distinguishing between similar objects and not for describing a particular sequential or chronological order.

The terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus/apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus/apparatus.

Thus far, the technical solution of the present invention has been described in connection with the preferred embodiments shown in the drawings, but it is easily understood by those skilled in the art that the scope of protection of the present invention is not limited to these specific embodiments. Equivalent modifications and substitutions for related technical features may be made by those skilled in the art without departing from the principles of the present invention, and such modifications and substitutions will be within the scope of the present invention.

Claims

1. A security path planning system based on an ant colony algorithm, the system comprising: the system comprises a scene acquisition module, a risk area and environment parameter acquisition module, an environment model construction module to be planned, an objective function construction module and a path planning module;

the path planning system is applied to unmanned plane path planning;

acquiring a depth image of a target area scene through a depth camera;

；

smoothing the interpolation environment model and setting a safe height interval, wherein the smoothing comprises the following steps: smoothing the interpolation environment model, and setting a first stepkFlight safety height section of personal unmanned aerial vehicle, wherein />Representing a minimum safe flight level, +.>Representing a maximum safe flight altitude; -providing said minimum safe flying height +.>Performing superposition processing on the smooth curved surface in the interpolation environment model after the smoothing processing to obtain a smooth unmanned plane flight safety curved surface；

Obtaining smooth unmanned plane flight safety curved surfaceAdding a risk range and setting flight constraint conditions according to the risk target position by the post interpolation environment model to obtain a three-dimensional environment model;

the risk adding range and the flight constraint condition setting specifically comprise:

；

the maximum horizontal turning angle constraint is specifically as follows:

；

the maximum pitch angle constraint is specifically as follows:

；

the maximum voyage constraint is specifically as follows:

；

wherein ,representing unmanned plane->Maximum of (2)The navigation distance of the navigation device is set,Nrepresenting the number of nodes in the overall path,N-1 is the number of segments of the path, ">Representing unmanned plane->Is the first of (2)iVoyage corresponding to each node- >Representing unmanned plane->Is a general course of (1);

the minimum inertial distance constraint is specifically as follows:

；

wherein Representing unmanned plane->Is a minimum inertial distance of (2);

the path objective function includes:

the course length cost is:

；

the environmental threat costs are:

；

the constraint cost is:

；

wherein [ (S)]Representing the value of the fetch，Representing a safe height interval +.>Is used for the average value of (a),representing the actual flying height deviation +.>Representing constraint costs;

cost the journey lengthEnvironmental threat cost- >And constraint cost->Normalization and scaling are carried out, and a path objective function is obtained:

；

wherein ,、/>、/>respectively represent normalized journey length cost +.>Environmental threat cost->And constraint cost->Weight coefficient of (2);

2. The security path planning system based on the ant colony algorithm of claim 1, wherein the improved ant colony algorithm specifically comprises:

step A003, searching the rest transferable node set for each ant, and according to the current pheromone concentration between different nodesCalculating transition probability->Distributing and moving all ants in the ant colony to respective next nodes according to the transition probability, and calculating the pheromone concentration of the next iteration round by using a Logistic chaotic mapping mode>The method comprises the steps of carrying out a first treatment on the surface of the Transition probability->The method comprises the steps of calculating and obtaining a reference journey length cost, an environmental threat cost and a constraint cost;

3. The ant colony algorithm-based safety path planning system according to claim 2, wherein the initializing ant colony algorithm parameters specifically comprises:

initializing ant colony numbersIteration number->System initial temperature->Stop temperature->And unmanned aerial vehicle performance parameters and initial pheromone concentration.

4. A security path planning system based on ant colony algorithm according to claim 3, wherein the calculation is performed by means of Logistic chaotic mapping Pheromone concentration of next iteration roundThe method specifically comprises the following steps:

；

wherein ,the method comprises the following steps:

；

initial valueA random number between 0 and 1, and +.>，/>Representing slave nodesiTo the point ofjThe chaotic sequence of the previous iteration round of (a),μrepresents chaos factor, and->When->，And then the chaotic state is entered.

5. The ant colony algorithm-based safety path planning system according to claim 4, wherein the safety path planning system is characterized in that the safety path planning system is based on the current pheromone concentrationCalculating transition probability->The method comprises the following steps:

；

wherein ,representing ant->Slave nodeiTo the nodejProbability of transition->Representing information elicitation factors,/->Representing the desired heuristic->Representing heuristic functions represents ant slave nodesiTransfer to nodejIs>Representing nodesjBelonging to ant->Is provided for the remaining set of transferable nodes,trepresenting the current iteration round, ++ >Representing nodesiTo the nodejStraight line distance between>Representing nodesiTo the nodejBetween not entering threat zonesAnd keep the minimum inertial distance +.>The minimum curve distance after that;

judgment nodeiTo the nodejThe number of threat zones passed;

if nodeiTo the nodejThrough a single threat zone, nodesiSum nodejAll add minimum inertial distance to threat zoneOutside the range of>Approximately as a nodeiAnd to the corresponding tangent point->Distance between->NodejAnd to the corresponding tangent point->Distance between->And tangential point->And (2) tangential point->The arc length between +.>The minimum curve distance for the case of a single threat zone is calculated as follows:

；

wherein , and />The method comprises the following steps:

；

representing nodesiCorresponding tangential point->And nodejCorresponding tangential point->Included angle between->Representing nodesiAnd corresponding tangent point->Included angle between->Representing nodesjAnd corresponding tangent point->An included angle between the two.

6. The ant colony algorithm-based safety path planning system according to claim 5, wherein if the node i to the node j pass through the plurality of threat zones, determining whether there is an overlapping area between the plurality of threat zones passed through;

judgment nodeiSum nodejWhether the connected straight line is positioned at the same side of the center of the first overlapping threat zone and the center of the second overlapping threat zone; if the two overlapping threat areas are on the same side, searching the outer common tangents of the first overlapping threat area and the second overlapping threat area, and selecting the midpoints of the two outer common tangentsmComputing nodeiTo the midpointmAvoiding threat zones therebetweensMinimum curve distance after 1And a midpointmTo the nodejAvoiding threat zones therebetweensMinimum curve distance after 2->The method comprises the steps of carrying out a first treatment on the surface of the If on different sides, at the nodeiTo the midpointmOn arcs or nodes betweenjTo the midpointmAdding auxiliary points on the circular arcs of the two threat areas, enabling the included angle formed by the node i or the node j to the auxiliary points by the circle centers of the corresponding threat areas to be an acute angle, and performing multi-section calculation;

judgment nodeiSum nodejWhether the connected straight line is positioned at the same side of the center of the first non-overlapping threat zone and the center of the second non-overlapping threat zone; if the two overlapping threat areas are on the same side, searching the outer common tangents of the first overlapping threat area and the second overlapping threat area, and selecting the midpoints of the two outer common tangents mComputing nodeiTo the midpointmAvoiding threat zones therebetweensMinimum curve distance after 1And a midpointmTo the nodejAvoiding threat zones therebetweensMinimum curve distance after 2->The method comprises the steps of carrying out a first treatment on the surface of the If the nodes are located at different sides, searching internal common tangent lines of the first non-overlapping threat area and the second non-overlapping threat area, selecting two midpoints of the internal common tangent lines, and calculating the nodesiTo the midpointmAvoiding threat zones therebetweensMinimum curve distance after 1->And the midpoint ofmTo the nodejAvoiding threat zones therebetweensMinimum curve distance after 2->And (3) summing.

7. The ant colony algorithm-based safety path planning system according to claim 2, wherein the adjusting of the pheromone concentration between the nodes left by the ant colony by the bipolar differentiation pheromone recording rule specifically comprises:

setting the ants in the first interval to leave pheromones on the passing path, and setting the weight of the pheromones as the product of the inverse ratio of the path length of the ants finding food and the ranking factor; the ranking factor is a value which is sequentially decreased from the starting sequence number to the ending sequence number of the first section in the direction of 1 to 0;

Resetting the pheromone left by the ant passing path in the second section;