CN105509749A - Mobile robot path planning method and system based on genetic ant colony algorithm - Google Patents

Abstract

The invention relates to a mobile robot path planning method and system based on a genetic ant colony algorithm. The mobile robot path planning method includes the steps that 1, modeling is conducted on the environment by establishing a coordinate system; 2, part of optimal solutions obtained through the genetic algorithm are converted to pheromones initial values of the ant colony algorithm; 3, optimum path search is conducted again through the ant colony algorithm, after optimum path search is ended, interlace operation is conducted on paths meeting the requirements of conditions, and the optimum path is finally obtained. The mobile robot path planning method and system overcome inevitable defects existing in a single ant colony algorithm, in other words, the ant colony algorithm is greater in blindness at the initial stage of search, the ant colony algorithm and the genetic algorithm are complementary in advantages, the search range of path search is shortened, and search efficiency of the optimum path is improved.

Description

Based on method for planning path for mobile robot and the system of GACA algorithm

Technical field

The present invention relates to intelligent robot algorithmic technique field, be specifically related to a kind of method for planning path for mobile robot based on GACA algorithm.

Background technology

Mobile robot is a key areas in intelligent control technology, except for except the fields such as interplanetary probe, ocean development and atomic energy, in factory automation, building, digs up mine, gets rid of the danger, military, service, also to have wide practical use in agricultural etc.The method of path planning has a lot, such as method of steepest descent, Artificial Potential Field Method, fuzzy reasoning method etc., and use method of steepest descent convergence slow, efficiency is not high, does not sometimes reach optimum solution; Use Artificial Potential Field Method to be convenient to the real-time control of bottom, but lack global information, there is local optimum problem; The advantage using fuzzy reasoning method maximum is that real-time is very good, but the formulation of the design of fuzzy membership functions, fuzzy control rule is mainly by the experience of people.

Genetic algorithm proves the strong algorithm of a kind of ability of searching optimum, has very strong robustness, concurrency.From the angle of macroscopic view, genetic algorithm has certain directivity, and therefore it is different from general random algorithm, a kind of instrument of the Stochastic choice that it uses just in directive search procedure, just because of its directivity, make it higher than general random algorithm efficiency.

Ant group algorithm be people be subject to real world ant seek carry out for impact and a kind of novel heuritic approach based on population optimizing that proposes, it is a kind of self-organization, walk abreast, the bionics algorithm of positive feedback, there is stronger robustness, can affect to external world in Actual path search and make dynamic response, it is widely used in path planning.

Summary of the invention

The object of this invention is to provide a kind of method for planning path for mobile robot and system, to solve the technical matters that blindness appears in ant group algorithm initial stage.

In order to solve the problems of the technologies described above, the invention provides a kind of method for planning path for mobile robot, comprising the steps:

Step S1, carries out modeling by setting up coordinate system to environment;

Step S2, a part of optimization solution genetic algorithm obtained is converted into the pheromones initial value of ant group algorithm;

Step S3, carries out optimum path search again by ant group algorithm, and optimizing terminates to carry out interlace operation to qualified path afterwards, finally obtains optimal path.

Further, by setting up coordinate system, the method that environment carries out modeling is comprised in described step 1:

The environmental detection device utilizing mobile robot to carry carries out modeling to generate a random initial path to environment.

Further, the method that a part of optimization solution genetic algorithm obtained in described step S2 is converted into the pheromones initial value of ant group algorithm comprises:

Step S21, initialization genetic parameter, to produce initial population;

Step S22, arranges fitness function, calculates the fitness of each population;

Step S23, before fitness is higher, 50% group of solution is converted to the pheromones initial value of ant group algorithm

Further, carry out optimum path search again in described step S3 by ant group algorithm, optimizing terminates to carry out interlace operation to qualified path afterwards, and the method finally obtaining optimal path comprises the steps:

Step S31, arranges ant group scale m ₂, maximum iteration time N _cand iterations initial value u is 0, and press following formula initialization remaining information element:

in formula: d _ijrepresent the distance between node i and node j;

Step S32, every ant selects next node, until it is then defeated to arrive impact point according to state transfer formula

Outbound path, if shorter than current iteration optimal path, then upgrades it, and state transfer formula is as follows:

$P_{ij}^k=\left\{\begin{array}{cc}\frac{{\[{\mathrm{\τ}}_{ij}\left(t\right)\]}^{\mathrm{\α}}\·{\[{\mathrm{\η}}_{ij}\left(t\right)\]}^{\mathrm{\β}}}{\underset{s\∈{\mathrm{allow}}_k}{\Σ}{\[{\mathrm{\τ}}_{is}\left(t\right)\]}^{\mathrm{\α}}\·{\[{\mathrm{\η}}_{is}\left(t\right)\]}^{\mathrm{\β}}},& s\∈{\mathrm{allow}}_k\\ 0,& s\∉{\mathrm{allow}}_k\end{array}\right.;$

In formula: allow _k(k=1,2...m ₂) be the set of ant k node to be visited, during beginning, allow _kin have (m ₂-1) individual element, namely comprises except ant k sets out other all nodes of node, along with the propelling of time, and allow _kin element reduce gradually, until be empty, namely represent that all nodes are all accessed complete;

τ _ijt () represents the pheromone concentration between t node i, node j on path; α is the pheromones significance level factor; β is the heuristic function significance level factor, and namely ant can be transferred to apart from short node with larger probability;

η _ijt () is heuristic function, represent that t ant transfers to the expected degree of node j from node i, computing formula is as follows: d _ijrepresent the distance between node i and node j;

Step S33, carries out global information element to the path of optimizing by following formula and upgrades, export the optimal path of current iteration simultaneously, judge whether current iteration optimal path and current global optimum path have the identical point except starting point and impact point simultaneously;

If have, then with this node for point of crossing, interlace operation is carried out to two paths, produces new route and compare with global optimum path, if new route is short, then renewal global optimum path;

The formula path of optimizing being carried out to the renewal of global information element is as follows:

τ _ij(t+1)＝(1-ρ)τ _ij(t)+ρΔτ _ij(t)+Δ ₁τ _ij-Δ ₂τ _ij；

${\mathrm{\Δ\τ}}_{ij}=\underset{k=1}{\overset{m_2}{\Σ}}{\mathrm{\Δ\τ}}_{ij}^k;$

In formula, ρ represents global information element volatility coefficient, represent the pheromone concentration that a kth ant discharges in node i and node j access path, Δ ₁τ _ijrepresent the pheromones increment through epicycle optimal path, Δ ₂τ _ijrepresent the pheromones increment through epicycle worst path; L ₁, L ₂represent the local optimum length in this circulation, the poorest length in local respectively; k ₁, k ₂represent this circulation local optimum, the poorest ant number in local respectively;

Step S34, cycle index u=u+1;

Step S35, places back in starting point ant, carries out next round iteration;

Step S36, if iterations u>N _c, then optimizing terminates, and exports optimal path; Otherwise proceed to step S35.

Another aspect, present invention also offers a kind of mobile robot path planning system, comprising:

Environmental modeling module, carries out modeling by setting up coordinate system to environment;

Pheromones for obtaining the pheromones initial value of ant group algorithm obtains module, and

Obtain with pheromones the optimal path that module is connected and obtain module.

Further, described environmental modeling module carries out modeling to generate a random initial path by the environmental detection device utilizing mobile robot to carry to environment;

Described environmental detection device comprises: the camera that mobile robot carries, sonar ring, infrared sensor.

Further, a part of optimization solution that described pheromones acquisition module is suitable for genetic algorithm being obtained is converted into the pheromones initial value of ant group algorithm; Namely

Initialization genetic parameter, to produce initial population; Fitness function is set, calculates the fitness of each population; And fitness is higher before 50% group of solution be converted to the pheromones initial value of ant group algorithm

Further, described optimal path obtains module and is suitable for carrying out optimum path search again by ant group algorithm, and optimizing terminates to carry out interlace operation to qualified path afterwards, finally obtains optimal path; Namely

Ant group scale m is set ₂, maximum iteration time N _cand iterations initial value u is 0, and press following formula initialization remaining information element:

in formula: d _ijrepresent the distance between node i and node j;

Every ant selects next node according to state transfer formula, until arrive impact point then outgoing route, if shorter than current iteration optimal path, then upgrade it, state transfer formula is as follows:

$P_{ij}^k=\left\{\begin{array}{cc}\frac{{\[{\mathrm{\τ}}_{ij}\left(t\right)\]}^{\mathrm{\α}}\·{\[{\mathrm{\η}}_{ij}\left(t\right)\]}^{\mathrm{\β}}}{\underset{s\∈{\mathrm{allow}}_k}{\Σ}{\[{\mathrm{\τ}}_{is}\left(t\right)\]}^{\mathrm{\α}}\·{\[{\mathrm{\η}}_{is}\left(t\right)\]}^{\mathrm{\β}}},& s\∈{\mathrm{allow}}_k\\ 0,& s\∉{\mathrm{allow}}_k\end{array}\right.;$

In formula: allow _k(k=1,2...m ₂) be the set of ant k node to be visited, during beginning, allow _kin have (m ₂-1) individual element, namely comprises except ant k sets out other all nodes of node, along with the propelling of time, and allow _kin element reduce gradually, until be empty, namely represent that all nodes are all accessed complete;

τ _ijt () represents the pheromone concentration between t node i, node j on path; α is the pheromones significance level factor; β is the heuristic function significance level factor, and namely ant can be transferred to apart from short node with larger probability;

η _ijt () is heuristic function, represent that t ant transfers to the expected degree of node j from node i, computing formula is as follows: d _ijrepresent the distance between node i and node j;

Carry out global information element to the path of optimizing by following formula to upgrade, export the optimal path of current iteration simultaneously, judge whether current iteration optimal path and current global optimum path have the identical point except starting point and impact point simultaneously;

If have, then with this node for point of crossing, interlace operation is carried out to two paths, produces new route and compare with global optimum path, if new route is short, then renewal global optimum path;

The formula path of optimizing being carried out to the renewal of global information element is as follows:

τ _ij(t+1)＝(1-ρ)τ _ij(t)+ρΔτ _ij(t)+Δ ₁τ _ij-Δ ₂τ _ij；

${\mathrm{\Δ\τ}}_{ij}=\underset{k=1}{\overset{m_2}{\Σ}}{\mathrm{\Δ\τ}}_{ij}^k;$

In formula, ρ represents global information element volatility coefficient, represent the pheromone concentration that a kth ant discharges in node i and node j access path, Δ ₁τ _ijrepresent the pheromones increment through epicycle optimal path, Δ ₂τ _ijrepresent the pheromones increment through epicycle worst path; L ₁, L ₂represent the local optimum length in this circulation, the poorest length in local respectively; k ₁, k ₂represent this circulation local optimum, the poorest ant number in local respectively;

Cycle index u=u+1;

Ant is placed back in starting point, carries out next round iteration;

If iterations u>N _c, then optimizing terminates, and exports optimal path; Otherwise continuation iteration.

The invention has the beneficial effects as follows, method for planning path for mobile robot of the present invention and system overcome the inevitable drawback that single ant group algorithm exists, namely ant group algorithm is too large in search starting stage blindness, achieve the mutual supplement with each other's advantages of ant group and genetic algorithm, reduce the seek scope of route searching, improve the search efficiency of optimal path.

Accompanying drawing explanation

Below in conjunction with drawings and Examples, the present invention is further described.

Fig. 1 is the FB(flow block) of method for planning path for mobile robot of the present invention;

Fig. 2 is the algorithm flow chart of method for planning path for mobile robot of the present invention.

Embodiment

In conjunction with the accompanying drawings, the present invention is further detailed explanation.These accompanying drawings are the schematic diagram of simplification, only basic structure of the present invention are described in a schematic way, and therefore it only shows the formation relevant with the present invention.

The principle of work of method for planning path for mobile robot of the present invention and system: because ant group algorithm is too large in search starting stage blindness, and genetic algorithm has good ability of searching optimum at the search initial stage; Therefore, the optimum solution that genetic algorithm produces is converted into the initial value of ant group algorithm pheromones, just effectively prevent the blindness at ant group algorithm initial stage.

Embodiment 1

As depicted in figs. 1 and 2, the present embodiment 1 provides a kind of method for planning path for mobile robot, comprises the steps:

Step S1, carries out modeling by setting up coordinate system to environment;

Step S2, a part of optimization solution genetic algorithm obtained is converted into the pheromones initial value of ant group algorithm;

Step S3, carries out optimum path search again by ant group algorithm, and optimizing terminates to carry out interlace operation to qualified path afterwards, finally obtains optimal path.

Carrying out the optional embodiment of one of modeling as environment, by setting up coordinate system, the method that environment carries out modeling being comprised in described step 1: the environmental detection device utilizing mobile robot to carry carries out modeling to generate a random initial path to environment.

Concrete, described environmental modeling, namely with mobile robot's starting point for true origin, the line of starting point and impact point is x-axis, is that y-axis sets up coordinate system at the vertical line of true origin place and x-axis; By the line segment n decile of starting point and impact point, Along ent is x _r(r=1,2...n-1), crosses x _rmake the vertical line l perpendicular to x-axis _r, randomly at l _rupper selection 1 p _r, successively these path point are coupled together from the off, until impact point, to form a random initial path.

As the pheromones initial value of ant group algorithm the optional embodiment of one.

The method that a part of optimization solution genetic algorithm obtained in described step S2 is converted into the pheromones initial value of ant group algorithm comprises:

Step S21, initialization genetic parameter, to produce initial population;

Namely population scale m is set ₁, crossover probability P _c, mutation probability P _m, and maximum iteration time N _maxloop initialization number of times s is 0, and generation initial population, producing initial population can have two kinds of methods to generate, first method is at starting point amplitude different from the intercropping of impact point and the sinusoidal curve in cycle in x-axis, obtains one group of direction sequence to generate initial individuals along sinusoidal track; Another kind method is according to stochastic generation initial path method, and the number of individuals that these two kinds of methods generate is all m ₁/ 2.

Step S22, arranges fitness function, calculates the fitness of each population;

By following formula determination fitness function f, calculate the fitness of each population:

$f=\frac{1}{(n_1-n_2)A+n_2\sqrt{2}A};$

In formula: n ₁for the grid sum that path comprises, n ₂for the grid logarithm that position is to angular dependence, A is the distance between adjacent cells central point, is the grid length of side;

And perform selection, intersection, mutation operation, cycle index s=s+1.

Operation is selected to adopt the pattern of competing between two, namely in population, Stochastic choice two individualities carry out fitness value and compare, the individuality of wherein larger fitness value is selected to enter the next generation, the relatively low individuality of fitness value is then eliminated, and this method avoids the generation of Premature Convergence and stagnation to a certain extent.

Interlace operation adopts some intersection of total Nodes, namely finds two to wait to intersect individual total node, exchanges chromosome before their total nodes or below, thus the continuous path individuality that formation two is new.

Mutation operation is in conjunction with Gene Mutation Principle, to the chromosome needing variation, the node of Stochastic choice except head and the tail node makes a variation, introduce the dynamic definitive variation rule of probability, the aberration rate less to the employing that adaptive value is large, to prevent excellent genes to be destroyed, also new gene can be introduced when being absorbed in locally optimal solution; The aberration rate that the employing that adaptive value is little is larger, effectively can improve individual adaptive value.And when making a variation, involved less aberration rate and comparatively Big mutation rate rate adopt 0.01 and 0.4 respectively.

If cycle index s>N _max, then genetic algorithm stops; Otherwise repeat above-mentioned iterative step;

Step S23, before fitness is higher, 50% group of solution is converted to the pheromones initial value of ant group algorithm using the initial value of solution higher for fitness value in genetic algorithm as ant group algorithm pheromones, efficiently solve the slow-footed problem of ant group algorithm preconvergence, in conjunction with both advantage substantially increase efficiency and the precision of search.

Concrete, carry out optimum path search again by ant group algorithm in described step S3, optimizing terminates to carry out interlace operation to qualified path afterwards, and the method finally obtaining optimal path comprises the steps:

Step S31, initialization ant swarm parameter, namely arranges ant group scale m ₂, maximum iteration time N _cand iterations initial value u is 0, and press following formula initialization remaining information element:

in formula: d _ijrepresent the distance between node i and node j;

Step S32, state transfer formula selects next node, until arrive impact point, namely every ant according to

State transfer formula selects next node, until arrive impact point then outgoing route;

Gained path is compared with current iteration optimal path, and it is upgraded, even optimum than current iteration

Path is short, then upgrade it, and state transfer formula is as follows:

$P_{ij}^k=\left\{\begin{array}{cc}\frac{{\[{\mathrm{\τ}}_{ij}\left(t\right)\]}^{\mathrm{\α}}\·{\[{\mathrm{\η}}_{ij}\left(t\right)\]}^{\mathrm{\β}}}{\underset{s\∈{\mathrm{allow}}_k}{\Σ}{\[{\mathrm{\τ}}_{is}\left(t\right)\]}^{\mathrm{\α}}\·{\[{\mathrm{\η}}_{is}\left(t\right)\]}^{\mathrm{\β}}},& s\∈{\mathrm{allow}}_k\\ 0,& s\∉{\mathrm{allow}}_k\end{array}\right.;$

In formula: allow _k(k=1,2...m ₂) be the set of ant k node to be visited, during beginning, allow _kin have (m ₂-1) individual element, namely comprises except ant k sets out other all nodes of node, along with the propelling of time, and allow _kin element reduce gradually, until be empty, namely represent that all nodes are all accessed complete;

τ _ijt () represents the pheromone concentration between t node i, node j on path; α is the pheromones significance level factor; β is the heuristic function significance level factor, and namely ant can be transferred to apart from short node with larger probability;

η _ijt () is heuristic function, represent that t ant transfers to the expected degree of node j from node i, computing formula is as follows: d _ijrepresent the distance between node i and node j;

Step S33, carry out global information element to the path of optimizing by following formula to upgrade, export the optimal path of current iteration simultaneously, judge whether current iteration optimal path and current global optimum path have the identical point except starting point and impact point, whether namely iteration is optimum has point of crossing with global optimum simultaneously;

If have, then with this node for point of crossing, interlace operation is carried out to two paths, produces new route and compare with global optimum path, if new route is short, then renewal global optimum path;

The formula path of optimizing being carried out to the renewal of global information element is as follows:

τ _ij(t+1)=(1-ρ) τ _ij(t)+ρ Δ τ _ij(t)+Δ ₁τ _ij-Δ ₂τ _ijthe next moment of t+1 indication (in this formula);

${\mathrm{\Δ\τ}}_{ij}=\underset{k=1}{\overset{m_2}{\Σ}}{\mathrm{\Δ\τ}}_{ij}^k;$

In formula, ρ represents global information element volatility coefficient, represent the pheromone concentration that a kth ant discharges in node i and node j access path, Δ ₁τ _ijrepresent the pheromones increment through epicycle optimal path, Δ ₂τ _ijrepresent the pheromones increment through epicycle worst path; L ₁, L ₂represent the local optimum length in this circulation, the poorest length in local respectively; k ₁, k ₂represent this circulation local optimum, the poorest ant number in local respectively;

Step S34, cycle index u=u+1;

Step S35, places back in starting point ant, carries out next round iteration;

Step S36, if iterations u>N _c, then optimizing terminates, and exports optimal path; Otherwise proceed to step S35.

Embodiment 2

As depicted in figs. 1 and 2, on embodiment 1 basis, the present embodiment 2 provides a kind of mobile robot path planning system, comprising:

Environmental modeling module, carries out modeling by setting up coordinate system to environment;

Pheromones for obtaining the pheromones initial value of ant group algorithm obtains module, and

Obtain with pheromones the optimal path that module is connected and obtain module.

Wherein, described environmental modeling module carries out modeling to generate a random initial path by the environmental detection device utilizing mobile robot to carry to environment; Described environmental detection device comprises: the camera that mobile robot carries, sonar ring, infrared sensor.

Further, a part of optimization solution that described pheromones acquisition module is suitable for genetic algorithm being obtained is converted into the pheromones initial value of ant group algorithm; I.e. initialization genetic parameter, to produce initial population; Fitness function is set, calculates the fitness of each population; And fitness is higher before 50% group of solution be converted to the pheromones initial value of ant group algorithm the concrete grammar obtained involved by pheromones initial value can see the relevant discussion of embodiment 1.

Described optimal path obtains module and is suitable for carrying out optimum path search again by ant group algorithm, and optimizing terminates to carry out interlace operation to qualified path afterwards, finally obtains optimal path; Namely

Ant group scale m is set ₂, maximum iteration time N _cand iterations initial value u is 0, and press following formula initialization remaining information element:

in formula: d _ijrepresent the distance between node i and node j;

Every ant selects next node according to state transfer formula, until arrive impact point then outgoing route, if shorter than current iteration optimal path, then upgrade it, state transfer formula is as follows:

$P_{ij}^k=\left\{\begin{array}{cc}\frac{{\[{\mathrm{\τ}}_{ij}\left(t\right)\]}^{\mathrm{\α}}\·{\[{\mathrm{\η}}_{ij}\left(t\right)\]}^{\mathrm{\β}}}{\underset{s\∈{\mathrm{allow}}_k}{\Σ}{\[{\mathrm{\τ}}_{is}\left(t\right)\]}^{\mathrm{\α}}\·{\[{\mathrm{\η}}_{is}\left(t\right)\]}^{\mathrm{\β}}},& s\∈{\mathrm{allow}}_k\\ 0,& s\∉{\mathrm{allow}}_k\end{array}\right.;$

In formula: allow _k(k=1,2...m ₂) be the set of ant k node to be visited, during beginning, allow _kin have (m ₂-1) individual element, namely comprises except ant k sets out other all nodes of node, along with the propelling of time, and allow _kin element reduce gradually, until be empty, namely represent that all nodes are all accessed complete;

τ _ijt () represents the pheromone concentration between t node i, node j on path; α is the pheromones significance level factor; β is the heuristic function significance level factor, and namely ant can be transferred to apart from short node with larger probability;

η _ijt () is heuristic function, represent that t ant transfers to the expected degree of node j from node i, computing formula is as follows: d _ijrepresent the distance between node i and node j;

Carry out global information element to the path of optimizing by following formula to upgrade, export the optimal path of current iteration simultaneously, judge whether current iteration optimal path and current global optimum path have the identical point except starting point and impact point simultaneously;

If have, then with this node for point of crossing, interlace operation is carried out to two paths, produces new route and compare with global optimum path, if new route is short, then renewal global optimum path;

The formula path of optimizing being carried out to the renewal of global information element is as follows:

τ _ij(t+1)=(1-ρ) τ _ij(t)+ρ Δ τ _ij(t)+Δ ₁τ _ij-Δ ₂τ _ijthe next moment of t+1 indication (in this formula);

${\mathrm{\Δ\τ}}_{ij}=\underset{k=1}{\overset{m_2}{\Σ}}{\mathrm{\Δ\τ}}_{ij}^k;$

In formula, ρ represents global information element volatility coefficient, represent the pheromone concentration that a kth ant discharges in node i and node j access path, Δ ₁τ _ijrepresent the pheromones increment through epicycle optimal path, Δ ₂τ _ijrepresent the pheromones increment through epicycle worst path; L ₁, L ₂represent the local optimum length in this circulation, the poorest length in local respectively; k ₁, k ₂represent this circulation local optimum, the poorest ant number in local respectively;

Cycle index u=u+1;

Ant is placed back in starting point, carries out next round iteration;

If iterations u>N _c, then optimizing terminates, and exports optimal path; Otherwise continuation iteration.

With above-mentioned according to desirable embodiment of the present invention for enlightenment, by above-mentioned description, relevant staff in the scope not departing from this invention technological thought, can carry out various change and amendment completely.The technical scope of this invention is not limited to the content on instructions, must determine its technical scope according to right.

Claims (8)

1. a method for planning path for mobile robot, is characterized in that, comprises the steps:

Step S1, carries out modeling by setting up coordinate system to environment;

Step S2, a part of optimization solution genetic algorithm obtained is converted into the pheromones initial value of ant group algorithm;

Step S3, carries out optimum path search again by ant group algorithm, and optimizing terminates to carry out interlace operation to qualified path afterwards, finally obtains optimal path.

2. method for planning path for mobile robot according to claim 1, is characterized in that, comprises in described step 1 by setting up coordinate system to the method that environment carries out modeling:

The environmental detection device utilizing mobile robot to carry carries out modeling to generate a random initial path to environment.

3. method for planning path for mobile robot according to claim 1 and 2, is characterized in that,

The method that a part of optimization solution genetic algorithm obtained in described step S2 is converted into the pheromones initial value of ant group algorithm comprises:

Step S21, initialization genetic parameter, to produce initial population;

Step S22, arranges fitness function, calculates the fitness of each population;

Step S23, before fitness is higher, 50% group of solution is converted to the pheromones initial value of ant group algorithm

4. method for planning path for mobile robot according to claim 3, is characterized in that,

Carry out optimum path search again by ant group algorithm in described step S3, optimizing terminates to carry out interlace operation to qualified path afterwards, and the method finally obtaining optimal path comprises the steps:

Step S31, arranges ant group scale m ₂, maximum iteration time N _cand iterations initial value u is 0, and press following formula initialization remaining information element:

in formula: d _ijrepresent the distance between node i and node j;

Step S32, every ant selects next node, until it is then defeated to arrive impact point according to state transfer formula

Outbound path, if shorter than current iteration optimal path, then upgrades it, and state transfer formula is as follows:

$P_{ij}^k=\left\{\begin{array}{c}\frac{{\[{\mathrm{\τ}}_{ij}\left(t\right)\]}^{\mathrm{\α}}\·{\[{\mathrm{\η}}_{ij}\left(t\right)\]}^{\mathrm{\β}}}{\underset{s\∈{\mathrm{allow}}_k}{\Σ}{\[{\mathrm{\τ}}_{is}\left(t\right))}^{\mathrm{\α}}\·{\[{\mathrm{\η}}_{is}\left(t\right)\]}^{\mathrm{\β}}},s\∈{\mathrm{allow}}_k\\ 0,s\∉{\mathrm{allow}}_k\end{array}\right.;$

In formula: allow _k(k=1,2...m ₂) be the set of ant k node to be visited, during beginning, allow _kin have (m ₂-1) individual element, namely comprises except ant k sets out other all nodes of node, along with the propelling of time, and allow _kin element reduce gradually, until be empty, namely represent that all nodes are all accessed complete;

τ _ijt () represents the pheromone concentration between t node i, node j on path; α is the pheromones significance level factor; β is the heuristic function significance level factor;

η _ijt () is heuristic function, represent that t ant transfers to the expected degree of node j from node i, computing formula is as follows: d _ijrepresent the distance between node i and node j;

Step S33, carries out global information element to the path of optimizing by following formula and upgrades, export the optimal path of current iteration simultaneously, judge whether current iteration optimal path and current global optimum path have the identical point except starting point and impact point simultaneously;

If have, then with this node for point of crossing, interlace operation is carried out to two paths, produces new route and compare with global optimum path, if new route is short, then renewal global optimum path;

The formula path of optimizing being carried out to the renewal of global information element is as follows:

τ _ij(t+1)＝(1-ρ)τ _ij(t)+ρΔτ _ij(t)+Δ ₁τ _ij-Δ ₂τ _ij；

${\mathrm{\Δ\τ}}_{ij}=\underset{k=1}{\overset{m_2}{\Σ}}{\mathrm{\Δ\τ}}_{ij}^k;$

In formula, ρ represents global information element volatility coefficient, represent the pheromone concentration that a kth ant discharges in node i and node j access path, Δ ₁τ _ijrepresent the pheromones increment through epicycle optimal path, Δ ₂τ _ijrepresent the pheromones increment through epicycle worst path; L ₁, L ₂represent the local optimum length in this circulation, the poorest length in local respectively; k ₁, k ₂represent this circulation local optimum, the poorest ant number in local respectively;

Step S34, cycle index u=u+1;

Step S35, places back in starting point ant, carries out next round iteration;

Step S36, if iterations u>N _c, then optimizing terminates, and exports optimal path; Otherwise proceed to step S35.

5. a mobile robot path planning system, is characterized in that, comprising:

Environmental modeling module, carries out modeling by setting up coordinate system to environment;

Pheromones for obtaining the pheromones initial value of ant group algorithm obtains module, and

Obtain with pheromones the optimal path that module is connected and obtain module.

6. mobile robot path planning system according to claim 5, is characterized in that,

Described environmental modeling module carries out modeling to generate a random initial path by the environmental detection device utilizing mobile robot to carry to environment;

Described environmental detection device comprises: the camera that mobile robot carries, sonar ring, infrared sensor.

7. mobile robot path planning system according to claim 6, is characterized in that, a part of optimization solution that described pheromones acquisition module is suitable for genetic algorithm being obtained is converted into the pheromones initial value of ant group algorithm; Namely

Initialization genetic parameter, to produce initial population; Fitness function is set, calculates the fitness of each population; And fitness is higher before 50% group of solution be converted to the pheromones initial value of ant group algorithm

8. mobile robot path planning system according to claim 7, is characterized in that,

Described optimal path obtains module and is suitable for carrying out optimum path search again by ant group algorithm, and optimizing terminates to carry out interlace operation to qualified path afterwards, finally obtains optimal path; Namely

Ant group scale m is set ₂, maximum iteration time N _cand iterations initial value u is 0, and press following formula initialization remaining information element:

in formula: d _ijrepresent the distance between node i and node j;

Every ant selects next node according to state transfer formula, until arrive impact point then outgoing route, if shorter than current iteration optimal path, then upgrade it, state transfer formula is as follows:

$P_{ij}^k=\left\{\begin{array}{c}\frac{{\[{\mathrm{\τ}}_{ij}\left(t\right)\]}^{\mathrm{\α}}\·{\[{\mathrm{\η}}_{ij}\left(t\right)\]}^{\mathrm{\β}}}{\underset{s\∈{\mathrm{allow}}_k}{\Σ}{\[{\mathrm{\τ}}_{is}\left(t\right))}^{\mathrm{\α}}\·{\[{\mathrm{\η}}_{is}\left(t\right)\]}^{\mathrm{\β}}},s\∈{\mathrm{allow}}_k\\ 0,s\∉{\mathrm{allow}}_k\end{array}\right.;$

In formula: allow _k(k=1,2...m ₂) be the set of ant k node to be visited, during beginning, allow _kin have (m ₂-1) individual element, namely comprises except ant k sets out other all nodes of node, along with the propelling of time, and allow _kin element reduce gradually, until be empty, namely represent that all nodes are all accessed complete;

τ _ijt () represents the pheromone concentration between t node i, node j on path; α is the pheromones significance level factor; β is the heuristic function significance level factor;

η _ijt () is heuristic function, represent that t ant transfers to the expected degree of node j from node i, computing formula is as follows: d _ijrepresent the distance between node i and node j;

Carry out global information element to the path of optimizing by following formula to upgrade, export the optimal path of current iteration simultaneously, judge whether current iteration optimal path and current global optimum path have the identical point except starting point and impact point simultaneously;

If have, then with this node for point of crossing, interlace operation is carried out to two paths, produces new route and compare with global optimum path, if new route is short, then renewal global optimum path;

The formula path of optimizing being carried out to the renewal of global information element is as follows:

τ _ij(t+1)＝(1-ρ)τ _ij(t)+ρΔτ _ij(t)+Δ ₁τ _ij-Δ ₂τ _ij；

${\mathrm{\Δ\τ}}_{ij}=\underset{k=1}{\overset{m_2}{\Σ}}{\mathrm{\Δ\τ}}_{ij}^k;$

In formula, ρ represents global information element volatility coefficient, represent the pheromone concentration that a kth ant discharges in node i and node j access path, Δ ₁τ _ijrepresent the pheromones increment through epicycle optimal path, Δ ₂τ _ijrepresent the pheromones increment through epicycle worst path; L ₁, L ₂represent the local optimum length in this circulation, the poorest length in local respectively; k ₁, k ₂represent this circulation local optimum, the poorest ant number in local respectively;

Cycle index u=u+1;

Ant is placed back in starting point, carries out next round iteration;

If iterations u>N _c, then optimizing terminates, and exports optimal path; Otherwise continuation iteration.