CN115454067A - Path planning method based on fusion algorithm - Google Patents
Path planning method based on fusion algorithm Download PDFInfo
- Publication number
- CN115454067A CN115454067A CN202211105650.3A CN202211105650A CN115454067A CN 115454067 A CN115454067 A CN 115454067A CN 202211105650 A CN202211105650 A CN 202211105650A CN 115454067 A CN115454067 A CN 115454067A
- Authority
- CN
- China
- Prior art keywords
- algorithm
- path
- wolf
- pheromone
- heuristic
- 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.)
- Withdrawn
Links
- 238000004422 calculation algorithm Methods 0.000 title claims abstract description 107
- 238000000034 method Methods 0.000 title claims abstract description 39
- 230000004927 fusion Effects 0.000 title claims abstract description 16
- 241000282461 Canis lupus Species 0.000 claims abstract description 68
- 239000003016 pheromone Substances 0.000 claims abstract description 58
- 238000006243 chemical reaction Methods 0.000 claims abstract description 5
- 230000006870 function Effects 0.000 claims description 26
- 241000257303 Hymenoptera Species 0.000 claims description 16
- 230000000694 effects Effects 0.000 claims description 5
- 230000004888 barrier function Effects 0.000 claims description 4
- MYVIATVLJGTBFV-UHFFFAOYSA-M thiamine(1+) chloride Chemical compound [Cl-].CC1=C(CCO)SC=[N+]1CC1=CN=C(C)N=C1N MYVIATVLJGTBFV-UHFFFAOYSA-M 0.000 claims description 4
- 241000282421 Canidae Species 0.000 claims description 3
- 238000009826 distribution Methods 0.000 claims description 3
- 230000002028 premature Effects 0.000 claims description 3
- 230000007704 transition Effects 0.000 claims description 3
- 230000002787 reinforcement Effects 0.000 claims description 2
- UFHFLCQGNIYNRP-UHFFFAOYSA-N Hydrogen Chemical compound [H][H] UFHFLCQGNIYNRP-UHFFFAOYSA-N 0.000 claims 1
- 230000008713 feedback mechanism Effects 0.000 claims 1
- 229910052739 hydrogen Inorganic materials 0.000 claims 1
- 239000001257 hydrogen Substances 0.000 claims 1
- 230000007547 defect Effects 0.000 abstract 1
- 238000004519 manufacturing process Methods 0.000 description 5
- 238000004088 simulation Methods 0.000 description 4
- 230000009286 beneficial effect Effects 0.000 description 3
- 239000003795 chemical substances by application Substances 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 238000005457 optimization Methods 0.000 description 3
- 238000011160 research Methods 0.000 description 3
- 201000004569 Blindness Diseases 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000007613 environmental effect Effects 0.000 description 2
- 230000002349 favourable effect Effects 0.000 description 2
- 230000000977 initiatory effect Effects 0.000 description 2
- 230000007246 mechanism Effects 0.000 description 2
- 239000011664 nicotinic acid Substances 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 238000005728 strengthening Methods 0.000 description 2
- 238000009825 accumulation Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 239000012141 concentrate Substances 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 239000002245 particle Substances 0.000 description 1
- 230000008447 perception Effects 0.000 description 1
- 238000003860 storage Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
- G05D1/02—Control of position or course in two dimensions
- G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
- G05D1/0212—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory
- G05D1/0221—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory involving a learning process
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
- G05D1/02—Control of position or course in two dimensions
- G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
- G05D1/0212—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory
- G05D1/0223—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory involving speed control of the vehicle
Landscapes
- Engineering & Computer Science (AREA)
- Aviation & Aerospace Engineering (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides a path planning method based on a fusion algorithm. The gray wolf algorithm is fused with the improved ant colony algorithm, and the defects that the traditional method is high in randomness, prone to non-optimal planning paths and slow in convergence are overcome. The method comprises the following steps: 1. the route is pre-searched by utilizing the gray wolf algorithm, and the obtained optimal solution is introduced into an pheromone model of the ant colony algorithm so as to solve the problem of invalid route search caused by the lack of pheromones in the early stage of the ant colony algorithm. 2. And modifying the heuristic information of the ant colony algorithm, adding corner constraint in a heuristic function, reducing the redundancy of the path, adaptively updating the heuristic factor, and dynamically adjusting the importance of the heuristic factor and the heuristic factor. 3. And a pseudo-random strategy is adopted to select a path, so that the convergence of the algorithm is accelerated, and the balance between conversion rate adjustment determination and randomness is introduced. The fusion algorithm can enable the AGV to quickly search and obtain an optimal collision-free path from the starting node to the target node, so that the transportation efficiency is improved, and the transportation cost is reduced.
Description
Technical Field
The invention relates to the technical field of AGV path planning, in particular to a path planning method based on a fusion algorithm.
Background
With the progress of computer science and the rapid development of manufacturing level, various industries gradually change from automation to intelligence, and the artificial intelligence technology has become one of the most concerned research fields. The AGV is an intelligent mobile robot integrating multiple functions such as environmental perception, planning decision, intelligent control and the like, and is widely applied to the fields of logistics storage industry, manufacturing industry, wharf carrying and the like. The AGV has changed traditional manufacturing transportation structure, has reduced manpower and materials cost, has promoted manufacturing efficiency.
The path planning is one of important technical contents in the field of artificial intelligence such as an AGV, and the main task of the path planning is to enable the AGV to safely avoid an obstacle to reach a target point position according to certain optimal criteria (such as shortest path and shortest time). There are many researches on path planning algorithms, and the traditional algorithms include an a-star algorithm, a Dijkstra algorithm and the like. The device is concerned with due to the characteristics of simple structure, easy execution and the like. However, they are computationally intensive and the found path is not always optimal. In contrast, intelligent bionic algorithms such as particle swarm, bee colony algorithm and ant colony algorithm are more robust, have better adaptability based on simple individuals and rules, and more people concentrate on the research of the intelligent bionic algorithms on path planning.
The ant colony optimization algorithm is evolved according to the food search mechanism of ants and is mainly used for solving the problems of travelers and trolley path planning. In the algorithm, each ant is equivalent to an agent, the agent selects the next node through the pheromone and the distance, when an agent finds a complete path, the pheromone is updated according to the length of the path, and the higher the concentration of the pheromone is, the higher the probability that the path is selected is. After the iteration is completed, an optimal path is obtained by the algorithm. The ant colony algorithm is well known for its advantages of good environmental adaptability, strong robustness, convenient fusion with other algorithms, and the like, but has strong randomness, and is easy to have the problems of non-optimal planning path, slow convergence, and the like. This may cause the AGV to frequently turn, especially in a complex environment, which may cause a lot of redundant routes or even failure to complete the path planning.
Disclosure of Invention
The invention aims to provide a path planning method based on a fusion algorithm, and aims to solve the problems that the traditional ant colony algorithm is low in search efficiency and easy to fall into local optimization in the background technology.
In order to achieve the purpose, the technical scheme adopted by the invention is as follows:
a path planning method based on a fusion algorithm comprises the following steps:
the method comprises the following steps: and modeling the actual working environment of the AGV by adopting a grid method. The method comprises the following specific steps:
(1) And if the obstacle in the AGV working area is known, determining the position, carrying out two-dimensional gridding treatment on the whole working area, and placing the two-dimensional gridding treatment in a two-dimensional plane coordinate system.
(2) The grid graph is coordinated and digitalized from top to bottom and from left to right, and the corresponding relation is as follows:
wherein: n is a radical of x For number of map lines, N y For the number of map columns, mod is the remainder operation, ceil is the rounding operator to the positive infinity, i is the grid number, and a is the length of the grid.
(3) The AGV is regarded as a mass point, the AGV runs at a constant speed in the environment, and the mechanical influence of the AGV per se is ignored.
(4) Enough safety margin is reserved for the expansion treatment of the barrier.
Step two: the method comprises the following steps of generating an initial path for an ant colony algorithm by using a gray wolf algorithm based on obstacle information, wherein the method comprises the following specific steps:
(1) Randomly constructing a wolf population, calculating a population fitness value, and determining alpha, beta, delta and omega wolf.
(2) And (6) tracking. α, β and δ are responsible for finding prey and passing its location information to other wolf individuals:
wherein,
is the distance between the wolf individual and the prey, t is the number of iterations,
is the position of the prey,
the position of the gray wolf is the position of the gray wolf,
and
as a function of the number of the coefficients,
for convergence factor, in [0,2]The linear degressive effect is achieved in an inner line,
and
is [0,1]The random vector of (2).
(3) And (6) performing enclosure catching. After the tracking is finished, the information content of alpha, beta and delta wolf is more, the position of the prey is more known, and the wolf group updates the position along with the information:
wherein,
is a distance model of alpha, beta and delta wolf and omega wolf,
the direction and step length of the gray wolf movement,
the updated position of the wolf pack.
(4) And (6) hunting. To be received
Influence when
In time, the next position of the wolf pack is infinitely close to the prey, eventually initiating an attack.
(5) The wolf colony hunting route is a route obtained by the grey wolf algorithm, and the new initial pheromone distribution of the ant colony algorithm generated based on the grey wolf algorithm is as follows:
τ ij ＝τ 0 +τ ij GWO
wherein tau is ij Represents the new initial pheromone concentration, tau, obtained after the path pre-search between the ith node and the jth node 0 Is the environment initial pheromone between the two nodes i, j, τ ij GWO is pheromone increment between two nodes i and j, which is used for converting path information obtained by the gray wolf algorithm into the path information.
Step three: ant colony algorithm parameter initialization, m =50,N max ＝100，ρ＝0.2，Q＝10，α 0 ＝1.5，β 0 ＝7，ω＝0.6，λ＝0.5，u＝0.5。
Step four: the method for improving the heuristic function, the heuristic factor, the transition probability and the pheromone updating strategy of the ant colony algorithm specifically comprises the following steps:
(1) The new heuristic function not only comprehensively considers local and global path information, but also introduces a corner constraint function into the heuristic function, prevents redundant inflection points from appearing in the path, and enhances the guidance of the algorithm.
Wherein d is ij Is the Euclidean distance of grid i and grid j, d je Is the Euclidean distance of grid j and grid e, d ie Is the Euclidean distance of grid i and grid e, σ is a constant greater than 0, turn represents the direction.
(2) In the initial stage of iteration, ants are greatly influenced by the heuristic function, as the iteration progresses, pheromones gradually dominate the selection of the next position, and the heuristic factor of self-adaptive updating is as follows:
wherein alpha is 0 For the initial information heuristic factor, beta 0 To initially expect a heuristic factor, N max Is the maximum number of iterations, N is the current number of iterations, α max For maximum value of information heuristic factor, beta min Is the desired heuristic minimum.
(3) Selecting the next node by adopting a random transfer strategy combining deterministic selection development and stochastic selection exploration:
wherein q is 0 Is the conversion rate, representing the threshold of ant selecting the next node, q is uniformly distributed in [0,1]Is used as the random variable.
Is the probability that the kth ant goes from the ith grid to the jth grid in the t iteration, λ is a proportionality coefficient, N 0 For the iteration threshold, u is a constant.
(4) When the pheromone is updated, a reinforcement means of the optimal route of the wolf is added on the original basis by combining the elite strategy, and meanwhile, in order to prevent premature convergence caused by continuous increase of the pheromone of the elite strategy, a weight coefficient of self-adaptive attenuation is set:
κ(t)＝exp(1-ωt)
wherein, tau ij (t + 1) is the concentration of pheromone after updating, rho is the volatilization coefficient of pheromone, m is the number of ants, delta tau ij (t) pheromone increment, Q pheromone intensity factor, L k The walking of ant k in this search is Cheng Zhihe, kappa (t) is weight, L g Algorithm path T for gray wolf GWO ω is a coefficient.
Step five: and judging whether the maximum iteration times is reached, if so, outputting optimal information, and otherwise, returning to the fourth step to continue iteration.
Compared with the prior art, the invention has the beneficial effects that:
(1) The method utilizes the gray wolf algorithm to carry out quick pre-search, and adds the obtained optimal solution serving as the initial pheromone increment into the ant colony algorithm, so that the blindness of the ant colony algorithm in the initial search can be reduced.
(2) The invention improves the heuristic function, not only considers the distance between the current node and the optional node, but also adds the heuristic information of the distance between the current node and the target point. Meanwhile, a corner constraint function is introduced, so that the guidance of the algorithm is enhanced, and path redundant nodes are reduced.
(3) The invention adaptively updates the information elicitation factor and the expected elicitation factor, and the information elicitation factor and the expected elicitation factor respectively change to the maximum value and the minimum value in a linear form along with the increase of the iteration times. The relative importance of the two is continuously adjusted and dynamically changed, and the precision of path search is improved.
(4) In the invention, the conversion rate is introduced into a pseudo-random transfer strategy combining deterministic selection and stochastic selection, and in the initial stage, the certainty of path selection is higher, and the convergence rate of the algorithm is improved; in the later stage, ants tend to select paths by a roulette method, the global searching capability of the algorithm is improved, and the balance between ant colony development and exploration is effectively adjusted.
(5) The method combines the elite strategy, adds a strengthening means of the optimal route of the wolf on the original basis, is favorable for better guiding the bias of ant search, and enables the algorithm to be converged more quickly.
Drawings
FIG. 1 is a flow chart of a method of the present invention;
FIG. 2 is an AGV routing diagram applying a conventional ant colony algorithm.
FIG. 3 is an AGV iteration convergence diagram applying a conventional ant colony algorithm.
FIG. 4 is a diagram of an AGV routing using the improved algorithm of the present invention.
FIG. 5 is a graph of the AGV iteration convergence using the improved algorithm of the present invention.
Detailed Description
The technical solution in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention.
The path planning method based on the fusion algorithm as shown in fig. 1 comprises modeling an AGV working area, pre-searching a path by using a grey wolf algorithm, further improving an ant colony algorithm, and continuously iterating to screen out an optimal path. The method comprises the following specific steps:
executing the step one: and carrying out grid division and coding on the AGV working area.
Modeling the AGV working environment, and generating a 20-by-20 grid map. The white grid is a free grid, indicated by 0 in the simulation, and the AGV allows passage; the black grid is an obstacle grid, and is indicated by 1 in the simulation, and the AGV cannot pass through the black grid. The method specifically comprises the following steps:
(1) And (4) when the obstacle in the AGV working area is known, determining the position, and placing the generated grid map into a two-dimensional plane coordinate system.
(2) The grid graph is coordinated and digitalized from top to bottom and from left to right, and the corresponding relation is as follows:
wherein: n is a radical of x For number of map lines, N y For the number of map columns, mod is the remainder operation, ceil is the rounding operator to the positive infinity, i is the grid number, and a is the length of the grid.
(3) The AGV is regarded as a mass point, the AGV runs at a constant speed in the environment, and the mechanical influence of the AGV per se is ignored.
(4) Enough safety margin is reserved for the expansion treatment of the barrier.
Executing the step two: an initial path for the ant colony algorithm is generated with a gray wolf algorithm based on the obstacle information.
The initial pheromone concentration of the classical ant colony algorithm is constant, so that the prior information is lost in the path, the positive feedback of the ant colony algorithm is not obvious, and the calculation amount of path information identification is increased. The gray wolf algorithm is used for carrying out iterative search in advance to generate an initial path for the AGV to run, and the initial path is used as the initial population information increment of the ant colony algorithm. The method specifically comprises the following steps:
(1) Randomly constructing a wolf population, calculating a population fitness value, and determining alpha, beta, delta and omega wolf.
(2) And (6) tracking. α, β and δ are responsible for finding prey and passing its location information to other grayish wolf individuals:
wherein,
is the distance between the wolf individual and the prey, t is the number of iterations,
is the position of the prey,
the position of the gray wolf is the position of the gray wolf,
and
as a function of the number of the coefficients,
for convergence factor, in [0,2]The linear degressive effect is achieved in an inner line,
and
is [0,1]The random vector of (2).
(3) And (6) performing enclosure catching. After the tracking is finished, the information content of alpha, beta and delta wolf is more, the position of the prey is more known, and the wolf group updates the position along with the information:
wherein,
is a distance model of alpha, beta and delta wolf and omega wolf,
the direction and step length of the gray wolf movement,
the updated position of the wolf pack.
(4) And (6) hunting. To be received
Effects when
In time, the next position of the wolf pack is infinitely close to the prey, eventually initiating an attack.
(5) The wolves hunting route is a path obtained by the grey wolf algorithm, and the initial pheromone distribution based on the grey wolf algorithm is as follows:
τ ij ＝τ 0 +τ ij GWO
wherein τ is ij Represents the new initial pheromone concentration, tau, obtained after the path pre-search between the ith node and the jth node 0 Is the environment initial pheromone between the two nodes i, j, τ ij GWO is the pheromone increment between the two nodes i and j which converts the path information obtained by the wolf algorithm into.
And step three is executed: initializing ant colony algorithm parameters, wherein the number of ants is m =50, and the maximum iteration number is N max =100, volatility coefficient ρ =0.2, pheromone intensity coefficient Q =10, initial value α of information heuristic factor 0 =1.5, initial value β of the desired heuristic factor 0 =7, coefficient ω =0.6, proportionality coefficient λ =0.5, constant u =0.5.
And step four is executed: the taboo list is emptied, ants are placed at the starting point, and iteration is started.
And executing the step five: the method for planning the path of the AGV based on the improved ant colony algorithm specifically comprises the following steps:
(1) And calculating an improved heuristic function, adding the current node into the taboo table, and updating the taboo table.
The heuristic function functions to enable the ant colony algorithm to guide ants to find high quality mobile nodes. The traditional ant colony algorithm only takes the reciprocal of the current node to the next node as a unique heuristic factor, and the difference of nodes at adjacent positions is very small, so that the heuristic information of a target node is very weak. Particularly, in a large-scale environment map, the guiding function of the conventional heuristic function is poor, which is one of the main reasons that the convergence rate of the algorithm is slow and the searching efficiency is poor.
The new heuristic function not only considers the distance between the current node and the optional node in the construction process, but also adds heuristic information of the distance between the current node and the target point. Furthermore, considering the lack of guidance of the beneficial heuristic information by ants, a large number of redundant inflection points may occur in the path, resulting in a decrease in path smoothness. Therefore, a rotation angle constraint function is introduced into the heuristic function, and the guidance of the algorithm is enhanced.
Wherein d is ij Is the Euclidean distance of grid i and grid j, d je Is the Euclidean distance of grid j and grid e, d ie Is the Euclidean distance of grid i and grid e, σ is a constant greater than 0, and turn represents the direction.
(2) The elicitation factors are updated in a self-adaptive mode, ants are greatly influenced by the elicitation functions in the initial iteration stage, and as the iteration is carried out, pheromones gradually guide the selection of the next position, and the method specifically comprises the following steps:
wherein alpha is 0 For the initial information heuristic factor, beta 0 To initially expect a heuristic factor, N max Is the maximum number of iterations, N is the current number of iterations, α max Elicitation factors for informationMaximum value, beta min Is the desired heuristic minimum.
In the original ant colony algorithm, alpha and beta are taken by experience, and the population diversity is gradually reduced. The improvement is followed by an increase in the number of iterations, the information heuristic and the desired heuristic will each vary in a linear fashion to a maximum minimum. The relative importance of the two is continuously adjusted and dynamically changed, and the precision of path searching is improved.
(3) Selecting the next node by adopting a pseudo-random transfer strategy combining deterministic selection and stochastic selection:
wherein q is 0 Is the conversion rate, representing the threshold value of the ant for selecting the next node, and q is uniformly distributed in [0,1 ]]Is used as the random variable.
Is the probability that the kth ant goes from the ith grid to the jth grid in the t iteration, λ is a proportionality coefficient, N 0 For the iteration threshold, u is a constant.
In the initial stage, q 0 The value is large, ants can be added to select favorable paths according to the global path information, and the convergence speed is improved; at a later stage, q 0 The value is small, which is beneficial for ants to improve the global search capability through random search of transition probabilities.
(4) And combining the route information of the wolf, and updating the pheromone.
In the traditional ant colony algorithm, each ant carries out pheromone updating once according to the length of a self-constructed path after each iteration, and the shorter the path is, the more pheromones are released, and the more obvious the guiding effect is. When the path lengths are small, the difference of pheromones becomes very small, which is also one of the main reasons that the algorithm easily falls into a locally optimal solution in a complex and large-scale environment.
The invention combines the elite strategy, adds a strengthening means of the optimal path of the wolf on the original basis, and simultaneously sets a weight coefficient of self-adaptive attenuation for preventing premature convergence caused by continuous increase of pheromone of the elite strategy:
κ(t)＝exp(1-ωt)
wherein, tau ij (t + 1) is the concentration of pheromone after updating, rho is the volatilization coefficient of pheromone, m is the number of ants, delta tau ij (t) pheromone increment, Q pheromone intensity factor, L k The walking of ant k in this search is Cheng Zhihe, kappa (t) is weight, L g Algorithm path T for gray wolf GWO ω is a coefficient.
And a sixth step is executed: and D, judging whether all ants reach the end point, if so, executing the step seven, and otherwise, turning to the step five.
And a seventh step is executed: and judging whether the maximum iteration frequency is reached, if so, outputting optimal information, otherwise, adding 1 to the iteration frequency, and returning to the fifth step to continue the iteration.
And step eight is executed: and outputting the optimal path.
The invention fuses the gray wolf algorithm and the ant colony algorithm, so that the AGV can effectively avoid the barrier and then search an optimal path.
In order to verify the effectiveness and superiority of the improved method, the invention firstly utilizes the improved ant colony algorithm to carry out simulation in MATLAB.
The ant colony algorithm has certain randomness, and the performance of the algorithm can be judged according to the possibility that the optimal solution is obtained after the algorithm is operated once.
The experimental simulation result shows that the path found by the improved ant colony algorithm is shorter than the path found by the traditional ant colony algorithm, and the iteration times and the turning times of the optimal path are less.
Referring to fig. 2-5, in the 20 × 20 grid environment, the improved ant colony algorithm has a fast convergence speed and a short early-stage search path, mainly because the initial pheromone concentrations of the conventional ant colony algorithm are the same, which causes blindness of the iterative initial search, and the improved ant colony algorithm provides new initial pheromone concentrations through the path pre-search, which forms the difference of the pheromone concentrations and can find the optimal path more quickly. An elite mechanism based on a grey wolf path is added in the global pheromone updating process, ants can select a more optimal path, the weight of self-adaptive attenuation is set, the influence of an elite strategy in the initial stage is guaranteed, meanwhile, the excessive influence in the later stage is reduced, and the ant colony is prevented from falling into local optimum due to excessive pheromone accumulation. In the aspect of the optimal path turning times, the improved ant colony algorithm is less than that of the traditional ant colony algorithm, the main reason is that the heuristic function considers the end point position information and adds the turning angle constraint, and a pseudo-random transfer strategy is adopted when the next position is selected, so that the AGV tends to run straight on the end point and the path with small turning angle. In conclusion, the path planning method based on the fusion algorithm has better optimization effect, can quickly search the optimal path, reduces the turning times and amplitude of the path, improves the transportation efficiency and saves the production and transportation cost.
Claims (6)
1. A path planning method based on a fusion algorithm is characterized by comprising the following steps:
the method comprises the following steps: modeling the working environment of the AGV by adopting a grid method, and determining the starting point and the end point of the AGV;
step two: the method comprises the steps that an initial path for an ant colony algorithm is generated through a grey wolf algorithm based on obstacle information, and the initial path of the AGV is searched and generated by utilizing the characteristics of high efficiency and excellent feedback mechanism of the grey wolf algorithm and is used as an increment of ant colony algorithm pheromone distribution;
step three: initializing ant colony algorithm parameters including the number m of ants and the maximum iteration number N max Initial value of primitive pheromone τ 0 Pheromone volatilization coefficient rho, pheromone intensity coefficient Q, coefficient omega, information heuristic factor alpha 0 And alpha max Expectation heuristic factor beta 0 And beta min The proportionality coefficient λ, constant u;
step four: and (4) improving the ant colony algorithm to carry out path planning on the AGV to obtain an optimal path.
2. The fusion algorithm based path planning method according to claim 1, wherein in the first step, a grid method is used to model the actual working environment of the AGV, and the specific steps are as follows:
(1) The method comprises the following steps of (1) determining the position of a known obstacle in an AGV working area, carrying out two-dimensional gridding treatment on the whole working area, and placing the two-dimensional gridding treatment in a two-dimensional plane coordinate system;
(2) The grid graph is coordinated and digitalized from top to bottom and from left to right, and the corresponding relation is as follows:
wherein: n is a radical of hydrogen x For the number of map lines, N y Mod is the remainder operation,ceil is the rounding operator to the positive infinity, i is the grid number, a is the length of the grid;
(3) The AGV is regarded as a mass point, the AGV runs at a constant speed in the environment, and the mechanical influence of the AGV per se is ignored;
(4) Enough safety margin is reserved for the expansion treatment of the barrier.
3. The path planning method based on the fusion algorithm as claimed in claim 1, wherein the initial path for the ant colony algorithm is generated by the grayish wolf algorithm based on the obstacle information in the second step, and the specific steps are as follows:
(1) Randomly constructing a wolf population, calculating a population fitness value, and determining alpha, beta, delta and omega wolf;
(2) Tracking, α, β and δ are responsible for finding prey and passing its location information to other grayish wolf individuals:
wherein,
is the distance between the wolf individual and the prey, t is the number of iterations,
is the position of the prey,
the position of the gray wolf is the position of the gray wolf,
and
as a function of the number of the coefficients,
for convergence factor, in [0,2]The linear degressive effect is achieved in an inner line,
and
is [0,1]The random vector of (a);
(3) After the trapping and tracking are finished, the information content of alpha, beta and delta wolves is more, the positions of prey are more known, and the wolves follow the positions to update:
wherein,
is a distance model of alpha, beta and delta wolf and omega wolf,
the direction and step length of the gray wolf movement,
updated positions for the wolf pack;
(4) Hunting and receiving
Effects when
When the attack is started, the next position of the wolf pack is infinitely close to a prey, and finally the attack is initiated;
(5) The wolf colony hunting route is a route obtained by the grey wolf algorithm, and the new initial pheromone of the ant colony algorithm generated based on the grey wolf algorithm is as follows:
τ ij ＝τ 0 +τ ij GWO
wherein tau is ij Represents the new initial pheromone concentration, tau, obtained after the path pre-search between the ith node and the jth node 0 Is the environment initial pheromone between the two nodes i, j, τ ij GWO is the pheromone increment between the two nodes i and j which converts the path information obtained by the wolf algorithm into.
4. The method for path planning based on fusion algorithm as claimed in claim 1, wherein m =50,N in the third step max ＝100，ρ＝0.2，Q＝10，α 0 ＝1.5，β 0 ＝7，ω＝0.6，λ＝0.5，u＝0.5。
5. The path planning method based on the fusion algorithm according to claim 1, wherein heuristic functions, heuristic factors, transition probabilities, and pheromone update strategies of the ant colony algorithm are improved in step four:
(1) The new heuristic function not only comprehensively considers local and global path information, but also introduces a corner constraint function into the heuristic function, prevents redundant inflection points from appearing in the path, and enhances the guidance of the algorithm:
wherein d is ij Is the Euclidean distance of grid i and grid j, d je Is the Euclidean distance of grid j and grid e, d ie Is the Euclidean distance of grid i and grid e, σ is a constant greater than 0, turn represents the direction;
(2) In the initial stage of iteration, ants are greatly influenced by the heuristic function, as the iteration progresses, pheromones gradually dominate the selection of the next position, and the heuristic factor of self-adaptive updating is as follows:
wherein alpha is 0 For the initial information heuristic factor, beta 0 For the initial desired heuristic factor, N max Is the maximum number of iterations, N is the current number of iterations, α max For maximum value of information heuristic factor, beta min Is the desired heuristic minimum;
(3) Selecting the next node by adopting a pseudo-random transfer strategy combining deterministic selection development and stochastic selection exploration:
wherein q is 0 Is the conversion rate, representing the threshold of ant selecting the next node, q is uniformly distributed in [0,1]The random variable (b) in (c),
is the probability that the kth ant goes from the ith grid to the jth grid in the t iteration, λ is a proportionality coefficient, N 0 Is an iteration threshold, u is a constant;
(4) When the pheromone is updated, a reinforcement means of the optimal route of the wolf is added on the original basis by combining the elite strategy, and meanwhile, in order to prevent premature convergence caused by continuous increase of the pheromone of the elite strategy, a weight coefficient of self-adaptive attenuation is set:
κ(t)＝exp(1-ωt)
wherein, tau ij (t + 1) isThe new pheromone concentration, rho is pheromone volatilization coefficient, m is ant number, delta tau ij (t) pheromone increment, Q pheromone intensity factor, L k The walking of ant k in this search is Cheng Zhihe, kappa (t) is weight, L g Algorithm path T for gray wolf GWO ω is a coefficient.
6. The method for path planning based on fusion algorithm according to claim 1, wherein the path planning for AGVs is performed by improving the ant colony algorithm in step four to obtain the optimal path, which specifically includes:
a. entering iteration;
b. according to the improved pheromone model, the heuristic information and the transfer strategy, the AGV selects the next node and determines a path until reaching a target point;
c. updating the path pheromone according to the improved pheromone updating strategy;
d. and c, judging whether the maximum iteration times is reached, if so, outputting optimal information, and otherwise, returning to the step a to continue iteration.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211105650.3A CN115454067A (en) | 2022-09-08 | 2022-09-08 | Path planning method based on fusion algorithm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211105650.3A CN115454067A (en) | 2022-09-08 | 2022-09-08 | Path planning method based on fusion algorithm |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115454067A true CN115454067A (en) | 2022-12-09 |
Family
ID=84303401
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211105650.3A Withdrawn CN115454067A (en) | 2022-09-08 | 2022-09-08 | Path planning method based on fusion algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115454067A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116502784A (en) * | 2023-06-27 | 2023-07-28 | 长春工业大学 | Path planning method based on hybrid intelligent optimization algorithm |
| CN117151577A (en) * | 2023-10-31 | 2023-12-01 | 南京职豆豆智能科技有限公司 | Combined path optimization method and system considering group heterogeneity |
-
2022
- 2022-09-08 CN CN202211105650.3A patent/CN115454067A/en not_active Withdrawn
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116502784A (en) * | 2023-06-27 | 2023-07-28 | 长春工业大学 | Path planning method based on hybrid intelligent optimization algorithm |
| CN117151577A (en) * | 2023-10-31 | 2023-12-01 | 南京职豆豆智能科技有限公司 | Combined path optimization method and system considering group heterogeneity |
| CN117151577B (en) * | 2023-10-31 | 2024-02-06 | 南京职豆豆智能科技有限公司 | Combined path optimization method and system considering group heterogeneity |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109945881B (en) | Mobile robot path planning method based on ant colony algorithm | |
| CN115454067A (en) | Path planning method based on fusion algorithm | |
| CN109597425B (en) | Unmanned aerial vehicle navigation and obstacle avoidance method based on reinforcement learning | |
| CN112327923B (en) | Multi-unmanned aerial vehicle collaborative path planning method | |
| CN111982125A (en) | Path planning method based on improved ant colony algorithm | |
| CN110347151B (en) | Robot path planning method fused with Bezier optimization genetic algorithm | |
| CN111323016A (en) | Mobile robot path planning method based on self-adaptive ant colony algorithm | |
| CN110196061A (en) | Based on the mobile robot global path planning method for improving ant group algorithm | |
| CN114167865B (en) | Robot path planning method based on countermeasure generation network and ant colony algorithm | |
| CN108413963A (en) | Bar-type machine people's paths planning method based on self study ant group algorithm | |
| CN111880561A (en) | Unmanned aerial vehicle three-dimensional path planning method based on improved whale algorithm in urban environment | |
| Li et al. | Mobile robot path planning based on improved genetic algorithm with A-star heuristic method | |
| Hsu et al. | Path planning for mobile robots based on improved ant colony optimization | |
| CN110686695A (en) | Adaptive ant colony A-star hybrid algorithm based on target evaluation factor | |
| Martin et al. | Offline and online evolutionary bi-directional RRT algorithms for efficient re-planning in dynamic environments | |
| Arora et al. | A survey of comparison between various metaheuristic techniques for path planning problem | |
| CN113534819A (en) | Method and storage medium for pilot-follow multi-agent formation path planning | |
| CN114756027A (en) | Mobile robot path planning method based on improved ant colony algorithm and Bezier curve | |
| Yasear et al. | Fine-Tuning the Ant Colony System Algorithm Through Harris’s Hawk Optimizer for Travelling Salesman Problem. | |
| CN111080035A (en) | Global path planning method based on improved quantum particle swarm optimization algorithm | |
| CN117387619A (en) | Path planning method based on fusion improvement ant colony and genetic algorithm | |
| CN113790729A (en) | Unmanned overhead traveling crane path planning method and device based on reinforcement learning algorithm | |
| Santos et al. | Exploratory path planning using the Max-min ant system algorithm | |
| Garip et al. | Path planning for multiple mobile robots in static environment using hybrid algorithm | |
| CN117434950A (en) | Mobile robot dynamic path planning method based on Harris eagle heuristic hybrid algorithm |
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 | ||
| WW01 | Invention patent application withdrawn after publication | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20221209 |