CN105911992A - Automatic path programming method of mobile robot, and mobile robot - Google Patents

Abstract

The invention discloses an automatic path programming method of a mobile robot, and the mobile robot using the method. The method comprises the following steps of: collecting environment information; according to the collected environment information, carrying out modeling on an area in which path programming of the mobile robot is carried out so as to construct a two-dimensional plane coordinate map, and determining a start point, a stop point and coordinate positions of obstacles; based on the two-dimensional plane coordinate map, carrying out path optimization on the paths from the start point to the stop point by means of population initialization based on a Sobol sequence and a firefly algorithm of a dynamic adjustment disturbance coefficient update population, and programming the optimized path to for driving in the two-dimensional plane coordinate map; and according to the optimized path after programming, driving the mobile robot to move. According to the invention, the problem that an existing firefly algorithm is insufficient in convergence performance is overcome, the mobile robot is enabled to rapidly, accurately and automatically program the path, and the path programming capability of the mobile robot is improved.

Description

The automatic path planning method of a kind of mobile robot and mobile robot

Technical field

The present invention relates to electronic robot technical field, more specifically, relate to a kind of mobile machine The automatic path planning method of people and mobile robot.

Background technology

Mobile robot (Mobile robot) is that one by sensor, remote manipulator and automatically controls The robot system of mobile vehicle composition, be that an interdisciplinary study that development in recent years is got up is integrated should Product, it has concentrated machinery, electronics, computer, automatically control and artificial intelligence etc. learns Section's newest research results, represents the overachievement of electromechanical integration.Mobile robot has mobile merit Can, in terms of replacing people to be engaged under danger, adverse circumstances the environment work less than operation and people, than There are bigger mobility, motility in general robot.Mobile robot the most gradually applies raw with industry Produce the different industries such as agriculture, medical.

In the research of mobile robot correlation technique, airmanship is its core, and path planning is One important step of navigation research and problem.So-called path planning refers to that mobile robot is according to a certain Performance indications (such as distance, time, energy resource consumption etc.) search for one from initial state to dbjective state Optimum or sub-optimal path.The problem that path planning relates generally to includes: (1) utilizes the movement obtained Robot environment's information sets up relatively reasonable model, then finds one from initial state with certain algorithm Optimum or the collisionless path of near-optimization to dbjective state；(2) can be in processing environment model The error occurred in uncertain factor and path trace, makes external object drop to the impact of robot Little；(3) utilize known all information to carry out the action of guided robot, thus obtain the most excellent Behaviour decision making.Research currently for mobile robot path planning technology has been achieved for substantial amounts of Achievement, many scientists are studied from different aspect.Wherein, from robot to environment sensing Angle, the research of method for planning path for mobile robot is divided into three types: rule based on environmental model The method of drawing, the planing method of vision based and the paths planning method of Behavior-based control；From robot The degree grasping environmental information, can be divided into again global path planning based on environment priori Complete Information With local paths planning based on sensor information；From the perspective of whether planning environment changes over, Also can be divided into static path planning and active path planning；From the concrete calculation of mobile robot path planning On method and strategy, planning technology can be divided into following four classes: stencil matching Path Planning Technique, artificial gesture Field Path Planning Technique, map structuring Path Planning Technique and artificial intelligence's Path Planning Technique.Manually Intelligence Path Planning Technique is the path planning that modern artificial intelligence technology is applied to mobile robot In, such as artificial neural network, evolutionary computation, fuzzy logic and swarm intelligence algorithm etc..Wherein, based on The Path Planning Technique of artificial intelligence is study hotspot in recent years.

Glowworm swarm algorithm (Firefly Algorithm) is proposed in 2008 by Yang Xin-she A kind of new intelligent optimization algorithm, in this algorithm simulation nature, the biological characteristics of fire fly luminescence is sent out Exhibition and come, the same with ant group algorithm with particle cluster algorithm, be also a kind of based on colony naturally inspire Formula Stochastic Optimization Algorithms.This algorithm, once proposition, enjoys the concern of Chinese scholars, becomes calculating intelligence One new study hotspot of energy research field, this algorithm has been applied in function optimization, image at present Process, Combinatorial Optimization, feature selection, cluster analysis, Stock Price Forecasting, protein structure prediction And the research field such as dynamic markets price.The computational efficiency of existing Lampyridea group's algorithm is high, and internal memory is opened Selling little, the parameter of regulation is few, is simply easily achieved, but the disturbance system in existing glowworm swarm algorithm Number α is usually fixed constant, and this is defective for the search of algorithm, with other random searches Algorithm equally there is also Premature convergence.

Summary of the invention

It is an object of the invention to overcome drawbacks described above of the prior art, it is provided that a kind of mobile robot Automatic path planning method and mobile robot, it is based on Sobol sequence initialization population and dynamically Adjust the strategy of coefficient of disturbance, by the coefficient of disturbance in glowworm swarm algorithm is carried out self-adaptative adjustment Strengthen convergence of algorithm performance, thus improve mobile robot and carry out the ability of path planning.

For achieving the above object, first aspect present invention provides the automatic planning of a kind of mobile robot Path Method, comprises the following steps:

Gather environmental information；

The region mobile robot being ready for path planning by the environmental information collected is carried out Modeling is to build two dimensional surface coordinate diagram, and determines the coordinate position of starting point, terminal and barrier；

Based on two dimensional surface coordinate diagram, to mobile robot from the path of origin-to-destination by based on The glowworm swarm algorithm of the initialization of population of Sobol sequence and dynamically adjustment coefficient of disturbance Population Regeneration is carried out Optimum path search, thus in two dimensional surface coordinate diagram planning plan to implement into path optimizing；

According to the path optimizing planned, order about mobile robot and move.

Preferably, in the above-mentioned methods, described based on two dimensional surface coordinate diagram, to shifting Mobile robot and is dynamically adjusted by initialization of population based on Sobol sequence from the path of origin-to-destination The glowworm swarm algorithm of whole coefficient of disturbance Population Regeneration carries out optimum path search, thus in two dimensional surface coordinate diagram Middle planning plan to implement into the step of path optimizing specifically include:

Import the basic parameter of glowworm swarm algorithm, and initialize each basic parameter of glowworm swarm algorithm；

Use Sobol sequence initialization population, produce the position of popN Lampyridea, calculate every The object function of Lampyridea is to obtain corresponding brightness, and selects the conduct optimum position of brightness maximum Put；

Calculate the Attraction Degree of every Lampyridea, by there is the Lampyridea of high-high brightness to guide other Luciola vitticollis The movement of worm, updates the position of every Lampyridea, and recalculates the brightness of Lampyridea；

When reaching maximum search number of times, then export optimum individual and stop algorithm, otherwise, recalculating The Attraction Degree of every Lampyridea.

Preferably, in the above-mentioned methods, described Sobol sequence is with 2 as base, by One group of direction number V₁, V₂, V₃..., V_i..., V_nGenerate, wherein, V_i∈ (0,1), at Sobol In sequence, the value of i-th element jth dimension can be obtained by formula:

${x_i}^j=n_1{V_1}^j\⊕n_2{V_2}^j\⊕...\⊕n_s{V_s}^j.$

Preferably, in the above-mentioned methods, the computing formula of the Attraction Degree of described Lampyridea For:

${\mathrm{\β}}_{ij}={\mathrm{\β}}_0*e^{-{\mathrm{\γd}}_{ij}^2};$

In formula, β₀Be two Lampyridea distances be captivation when zero, γ is the absorption coefficient of light, d_ijIt is Distance between Lampyridea i and Lampyridea j.

Preferably, in the above-mentioned methods, the location updating formula of described Lampyridea is:

X_j(t+1)=X_j(t)+β_ij(X_i(t)-X_j(t))+α(t)*ε_j；

$\mathrm{\α}\left(t\right)={\mathrm{\α}}_{begin}+({\mathrm{\α}}_{begin}-{\mathrm{\α}}_{end})*\[\frac{2t}{T}-{\left(\frac{t}{T}\right)}^2\];$

In formula, X_i(t) and X_jT () is Lampyridea i and the Lampyridea j space bit when the t time iteration respectively Putting, α is coefficient of disturbance, ε_jBeing random number vector, T is iterations.

Second aspect present invention provides the mobile robot of a kind of automatic path planning, and its feature exists In, including:

Environment information acquisition module, is used for gathering environmental information；

Environmental information MBM, for by the environmental information that collects mobile robot prepared into The region of row path planning is modeled building two dimensional surface coordinate diagram, and determine starting point, terminal and The coordinate position of barrier；

Path planning module, for based on two dimensional surface coordinate diagram, to mobile robot from starting point to end The path of point is by initialization of population based on Sobol sequence and dynamically adjusts coefficient of disturbance Population Regeneration Glowworm swarm algorithm carry out optimum path search, thus in two dimensional surface coordinate diagram planning plan to implement into optimization Path；

Mobile driving module, for according to the path optimizing planned, orders about mobile robot and carries out Mobile.

Preferably, in the scheme of above-mentioned mobile robot, described path planning module Specifically include:

Basic parameter input block, for importing the basic parameter of glowworm swarm algorithm, and initializes Luciola vitticollis Each basic parameter of worm algorithm；

Sobol sequence initialization kind group unit, is used for using Sobol sequence initialization population, produces The position of popN Lampyridea；

Dynamic disturbances coefficient optimum path search unit, for calculating the object function of every Lampyridea to obtain Corresponding brightness, and select brightness maximum as optimal location；And calculate every Lampyridea Attraction Degree, by having the Lampyridea of high-high brightness to guide the movement of other Lampyrideas, updates every firefly The position of fireworm, and recalculate the brightness of Lampyridea；When reaching maximum search number of times, then export Excellent individuality also stops algorithm, otherwise, recalculates the Attraction Degree of every Lampyridea.

Preferably, in the scheme of above-mentioned mobile robot, described Sobol sequence is With 2 as base, by one group of direction number V₁, V₂, V₃..., V_i..., V_nGenerate, wherein, V_i∈ (0, 1), in Sobol sequence, the value of i-th element jth dimension can be obtained by formula:

${x_i}^j=n_1{V_1}^j\⊕n_2{V_2}^j\⊕...\⊕n_s{V_s}^j.$

Preferably, in the scheme of above-mentioned mobile robot, the attraction of described Lampyridea The computing formula of degree is:

${\mathrm{\β}}_{ij}={\mathrm{\β}}_0*e^{-{\mathrm{\γd}}_{ij}^2};$

In formula, β₀Be two Lampyridea distances be captivation when zero, γ is the absorption coefficient of light, d_ijIt is Distance between Lampyridea i and Lampyridea j.

Preferably, in the scheme of above-mentioned mobile robot, the position of described Lampyridea More new formula is:

X_j(t+1)=X_j(t)+β_ij(X_i(t)-X_j(t))+α(t)*ε_j；

$\mathrm{\α}\left(t\right)={\mathrm{\α}}_{begin}+({\mathrm{\α}}_{begin}-{\mathrm{\α}}_{end})*\[\frac{2t}{T}-{\left(\frac{t}{T}\right)}^2\];$

In formula, X_i(t) and X_jT () is Lampyridea i and the Lampyridea j space bit when the t time iteration respectively Putting, α is coefficient of disturbance, ε_jBeing random number vector, T is iterations.

Compared with prior art, the beneficial effects of the present invention is:

1, the present invention can build two dimensional surface coordinate diagram according to the environmental information collected, and calls Initialization component uses Sobol sequence initialization population, is then based on dynamic disturbances coefficient update population So that in two dimensional surface coordinate diagram planning plan to implement into path, finally combine two dimensional surface coordinate diagram and The path planned moves.The present invention disturbs based on Sobol sequence initialization population and dynamic adjustment The strategy of dynamic coefficient, by carrying out self-adaptative adjustment to the key parameter-coefficient of disturbance in glowworm swarm algorithm Strengthen convergence of algorithm performance, overcome the not enough problem of existing glowworm swarm algorithm constringency performance, Enable mobile robot automatic path planning quickly and accurately, improve mobile robot and carry out road The ability of footpath planning.

2, the present invention uses Sobol sequence to initialize Lampyridea population, it is possible to obtains and preferably adopts Sample coverage rate, to ensure the uniformity that initial population is distributed.

Accompanying drawing explanation

In order to be illustrated more clearly that the embodiment of the present invention or technical scheme of the prior art, below by right In embodiment or description of the prior art, the required accompanying drawing used is briefly described, it should be apparent that, Accompanying drawing in describing below is some embodiments of the present invention, comes for those of ordinary skill in the art Say, on the premise of not paying creative work, it is also possible to obtain other accompanying drawing according to these accompanying drawings.

Fig. 1 is the flow chart of the automatic path planning method of a kind of mobile robot that the present invention provides；

Fig. 2 is the schematic diagram of the two dimensional surface coordinate diagram that the present invention provides；

Fig. 3 is the structured flowchart of the mobile robot of a kind of automatic path planning that the present invention provides；

Fig. 4 is the structured flowchart of the path planning module that the present invention provides.

Detailed description of the invention

For making the purpose of the embodiment of the present invention, technical scheme and advantage clearer, below in conjunction with this Accompanying drawing in inventive embodiments, clearly and completely retouches the technical scheme in the embodiment of the present invention State, it is clear that described embodiment is a part of embodiment of the present invention rather than whole embodiments. Based on the embodiment in the present invention, those of ordinary skill in the art are not making creative work premise Lower obtained every other embodiment, broadly falls into the scope of protection of the invention.

Embodiment one

Embodiments of the invention one provide the automatic path planning method of a kind of mobile robot, below In conjunction with accompanying drawing, the present embodiment is described in detail.Fig. 1 is the method flow diagram of the embodiment of the present invention one, Refer to Fig. 1, the method for the embodiment of the present invention comprises the following steps:

Step S1, collection environmental information；

Wherein, move robot and can obtain external by infrared sensor or the scanning of other harvesters Environmental information.

Step S2, the environmental information passing through to collect are ready for the district of path planning to mobile robot Territory is modeled building two dimensional surface coordinate diagram, and determines the coordinate bit of starting point, terminal and barrier Put；

Path planning refers to find one in the working environment have barrier from origin-to-destination, nothing Walk around to collision all barriers motion path (that is: find out from A point to B point collisionless Short path).

As in figure 2 it is shown, when two dimensional surface coordinate diagram has built, the environmental information of mobile robot MBM can recognize that in the global area intending exploring, the position of barrier (such as: a, b, c) is sat Mark.Owing to having a plurality of from A point to the path of B point, it is therefore desirable to utilize the Lampyridea that the present invention improves Algorithm therefrom identifies shortest path.

Step S3, based on two dimensional surface coordinate diagram, logical from the path of origin-to-destination to mobile robot Cross initialization of population based on Sobol sequence and dynamically adjust the Lampyridea calculation of coefficient of disturbance Population Regeneration Method carries out optimum path search, thus in two dimensional surface coordinate diagram planning plan to implement into path optimizing；

Furthermore, step S3 specifically includes following steps:

Step S31, the basic parameter of importing glowworm swarm algorithm, and initialize each of glowworm swarm algorithm Basic parameter；

Wherein, described basic parameter can include population quantity popN, iterations T, initially attract Degree β₀, absorption coefficient of light γ, coefficient of disturbance α etc..Their initial value can be as shown in following table one:

Step S32, employing Sobol sequence initialization population, produce the position of popN Lampyridea, Calculate the object function of every Lampyridea to obtain corresponding brightness, and select the maximum conduct of brightness Optimal location；

Concrete, described Sobol sequence is with 2 as base, by one group of direction number V₁, V₂, V₃..., Vi ..., V_nGenerate, wherein, Vi ∈ (0,1).Assume that mono-group of sequence of Sobol is x₁, x₂, x₃..., x_i..., x_n,Represent the value of i-th element jth dimension in Sobol sequence, can obtain by formula:

${x_i}^j=n_1{V_1}^j\⊕n_2{V_2}^j\⊕...\⊕n_s{V_s}^j.$

There is sample distribution and be distributed inconsistent problem with true in the most common pseudo-random number sequence, Sobol random sequence is that ((low-discrepancy sequences) is one to a kind of low diversity sequence Stable random sequence, distributing homogeneity is good.The present invention uses Sobol sequence to initialize Luciola vitticollis Worm population, it is possible to obtain coverage rate of preferably sampling, to ensure the uniformity that initial population is distributed.

Glowworm swarm algorithm is a kind of heuritic approach, and the more weak Lampyridea of algorithm simulation brightness is to brightness The random search that stronger Lampyridea moves, the absolute brightness of usual Lampyridea in glowworm swarm algorithm Represent target function value, i.e. f (x^*)=max_x∈sF (x), this algorithm utilizes a quantity to be popN Lampyridea population solve this optimization problem, algorithm starting stage, all of Lampyridea by iteration It is probabilistically assigned in the s of search volume.x_iRepresent an i-th Lampyridea solution when the t time iteration, f(x_i) mean that the absolute brightness of the Lampyridea of its correspondence.

Step S33, calculate the Attraction Degree of every Lampyridea, the Lampyridea with high-high brightness draw Lead the movement of other Lampyrideas, update the position of every Lampyridea, and recalculate the brightness of Lampyridea；

Every Lampyridea has captivation β to other Lampyrideas, if the absolute brightness of Lampyridea i More than the absolute brightness of Lampyridea j, then Lampyridea j will be attracted to move to i by Lampyridea i.Lampyridea I captivation β to Lampyridea j_ijFormula be defined as:

${\mathrm{\β}}_{ij}={\mathrm{\β}}_0*e^{-{\mathrm{\γd}}_{ij}^2}---\left(1\right)$

Wherein, β₀Be two Lampyridea distances be captivation when zero, γ is the absorption coefficient of light (Light Absorption Coefficient), d_ijIt it is the distance between Lampyridea i and Lampyridea j.

If when the t time iteration, Lampyridea j moves to Lampyridea i, then the location updating of Lampyridea j is public Formula is:

X_j(t+1)=X_j(t)+β_ij(X_i(t)-X_j(t))+α(t)*ε_j (2)

In formula, X_i(t) and X_jT () is Lampyridea i and the Lampyridea j space bit when the t time iteration respectively Putting, α is coefficient of disturbance, ε_jIt it is random number vector.

The more new formula that α is used by the present invention is as follows:

$\mathrm{\α}\left(t\right)={\mathrm{\α}}_{begin}+({\mathrm{\α}}_{begin}-{\mathrm{\α}}_{end})*\[\frac{2t}{T}-{\left(\frac{t}{T}\right)}^2\]---\left(3\right)$

From the point of view of the operation of algorithm, a bigger α value beneficially global search, and a less α value Be conducive to Local Search, therefore improve convergence of algorithm performance by α is dynamically adjusted.

Step S34, when reaching maximum search number of times, then output optimum individual stop algorithm, otherwise, Return step S33 and recalculate the Attraction Degree of every Lampyridea.

Wherein, maximum search number of times refers to the optimizing number of times of glowworm swarm algorithm, i.e. iterations T.

The false code of this algorithm is as follows:

Step S4, according to the path optimizing planned, order about mobile robot and move.

Wherein, the path that optimum individual is passed by is exactly optimal path, when path planning module determines one After bar optimal path, mobile robot will move along this paths.

The method of the present invention is based on Sobol sequence initialization population and the dynamic plan adjusting coefficient of disturbance Slightly, convergence is strengthened by the coefficient of disturbance in glowworm swarm algorithm is carried out self-adaptative adjustment Can, overcome the not enough problem of existing glowworm swarm algorithm constringency performance, enable mobile robot fast Speed, exactly automatic path planning, improve mobile robot and carry out the ability of path planning.

Embodiment two

Embodiments of the invention two provide the mobile robot of a kind of automatic path planning, refer to figure 3, the mobile robot of the embodiment of the present invention includes environment sensing module 1, path planning module 2 and moves Dynamic driving module 3, wherein, environment sensing module 1 is provided with environment information acquisition module 11 and environmental information MBM 12, will be described in detail the function of above-mentioned module below.

Environment information acquisition module 11, is used for gathering environmental information.

Wherein, environment information acquisition module 11 could be arranged to infrared sensor or other harvesters, Mobile robot can obtain external environmental information by infrared sensor or the scanning of other harvesters.

Environmental information MBM 12, prepares mobile robot for the environmental information by collecting The region carrying out path planning is modeled building two dimensional surface coordinate diagram, and determines starting point, terminal Coordinate position with barrier.

Path planning refers to find one in the working environment have barrier from origin-to-destination, nothing Walk around to collision all barriers motion path (that is: find out from A point to B point collisionless Short path).

As in figure 2 it is shown, when two dimensional surface coordinate diagram has built, the environmental information of mobile robot MBM 12 can recognize that the position of barrier (such as: a, b, c) in the global area intending exploring Put coordinate.Owing to having a plurality of from A point to the path of B point, it is therefore desirable to utilize path planning module 2 Therefrom identify shortest path.

Path planning module 2, for based on two dimensional surface coordinate diagram, to mobile robot from starting point to The path of terminal is by initialization of population based on Sobol sequence and dynamically adjusts coefficient of disturbance more novel species The glowworm swarm algorithm of group carries out optimum path search, thus in two dimensional surface coordinate diagram planning plan to implement into excellent Change path.

As shown in Figure 4, furthermore, in the present embodiment, described path planning module 2 is concrete Including:

Basic parameter input block 21, for importing the basic parameter of glowworm swarm algorithm, and initializes firefly Each basic parameter of fireworm algorithm；

Wherein, described basic parameter can include population quantity popN, iterations T, initially attract Degree β₀, absorption coefficient of light γ, coefficient of disturbance α etc..Their initial value can be as shown in following table one:

Sobol sequence initialization kind group unit 22, is used for using Sobol sequence initialization population, produces The position of raw popN Lampyridea；

Concrete, described Sobol sequence is with 2 as base, by one group of direction number V₁, V₂, V₃..., Vi ..., V_nGenerate, wherein, Vi ∈ (0,1).Assume that mono-group of sequence of Sobol is x₁, x₂, x₃..., x_i..., x_n,Represent the value of i-th element jth dimension in Sobol sequence, can obtain by formula:

${x_i}^j=n_1{V_1}^j\⊕n_2{V_2}^j\⊕...\⊕n_s{V_s}^j.$

There is sample distribution and be distributed inconsistent problem with true in the most common pseudo-random number sequence, Sobol random sequence is that ((low-discrepancy sequences) is one to a kind of low diversity sequence Stable random sequence, distributing homogeneity is good.The present invention uses Sobol sequence to initialize Luciola vitticollis Worm population, it is possible to obtain coverage rate of preferably sampling, to ensure the uniformity that initial population is distributed.

Dynamic disturbances coefficient optimum path search unit 23, for calculating the object function of every Lampyridea to obtain To corresponding brightness, and select brightness maximum as optimal location；And calculate every Lampyridea Attraction Degree, by there is the Lampyridea of high-high brightness to guide the movement of other Lampyrideas, update every The position of Lampyridea, and recalculate the brightness of Lampyridea；When reaching maximum search number of times, then export Optimum individual also stops algorithm, otherwise, recalculates the Attraction Degree of every Lampyridea.

Glowworm swarm algorithm is a kind of heuritic approach, and the more weak Lampyridea of algorithm simulation brightness is to brightness The random search that stronger Lampyridea moves, the absolute brightness of usual Lampyridea in glowworm swarm algorithm Represent target function value, i.e. f (x^*)=max_x∈sF (x), this algorithm utilizes a quantity to be popN Lampyridea population solve this optimization problem, algorithm starting stage, all of Lampyridea by iteration It is probabilistically assigned in the s of search volume.x_iRepresent an i-th Lampyridea solution when the t time iteration, f(x_i) mean that the absolute brightness of the Lampyridea of its correspondence.

Every Lampyridea has captivation β to other Lampyrideas, if the absolute brightness of Lampyridea i More than the absolute brightness of Lampyridea j, then Lampyridea j will be attracted to move to i by Lampyridea i.Lampyridea I captivation β to Lampyridea j_ijFormula be defined as:

${\mathrm{\β}}_{ij}={\mathrm{\β}}_0*e^{-{\mathrm{\γd}}_{ij}^2}---\left(1\right)$

Wherein, β₀Be two Lampyridea distances be captivation when zero, γ is the absorption coefficient of light (Light Absorption Coefficient), d_ijIt it is the distance between Lampyridea i and Lampyridea j.

If when the t time iteration, Lampyridea j moves to Lampyridea i, then the location updating of Lampyridea j is public Formula is:

X_j(t+1)=X_j(t)+β_ij(X_i(t)-X_j(t))+α(t)*ε_j (2)

In formula, X_i(t) and X_jT () is Lampyridea i and the Lampyridea j space bit when the t time iteration respectively Putting, α is coefficient of disturbance, ε_jIt it is random number vector.

The more new formula that α is used by the present invention is as follows:

$\mathrm{\α}\left(t\right)={\mathrm{\α}}_{begin}+({\mathrm{\α}}_{begin}-{\mathrm{\α}}_{end})*\[\frac{2t}{T}-{\left(\frac{t}{T}\right)}^2\]---\left(3\right)$

From the point of view of the operation of algorithm, a bigger α value beneficially global search, and a less α value Be conducive to Local Search, therefore improve convergence of algorithm performance by α is dynamically adjusted.

Mobile driving module 3, for according to the path optimizing planned, orders about mobile robot and enters Row is mobile.

Wherein, the path that optimum individual is passed by is exactly optimal path, when path planning module 2 determines Article one, after optimal path, mobile driving module 3 will be ordered about mobile robot along this paths and be moved Dynamic.

The shifter people of the present invention is based on Sobol sequence initialization population and dynamically adjusts coefficient of disturbance Strategy, strengthens convergence of algorithm by the coefficient of disturbance in glowworm swarm algorithm is carried out self-adaptative adjustment Performance, overcomes the not enough problem of existing glowworm swarm algorithm constringency performance, and it can quickly, accurately Ground automatic path planning, improves path planning ability.

It should be noted that the mobile robot of a kind of automatic path planning of above-described embodiment offer, Only it is illustrated with the division of above-mentioned each functional module, in actual application, can be as desired Above-mentioned functions distribution is completed by different functional modules, the internal structure of system will be divided into difference Functional module, to complete all or part of function described above.

One of ordinary skill in the art will appreciate that realize in above-described embodiment method all or part of Step can be by program and completes to instruct relevant hardware, and described program can be stored in In one computer read/write memory medium, described storage medium, such as ROM/RAM, disk, light Dish etc..

Above-described embodiment is the present invention preferably embodiment, but embodiments of the present invention are not by upper Stating the restriction of embodiment, that is made under other any spirit without departing from the present invention and principle changes Become, modify, substitute, combine, simplify, all should be the substitute mode of equivalence, be included in the present invention Protection domain within.

Claims (10)

1. the automatic path planning method moving robot, it is characterised in that the method includes Following steps:

Gather environmental information；

The region mobile robot being ready for path planning by the environmental information collected is carried out Modeling is to build two dimensional surface coordinate diagram, and determines the coordinate position of starting point, terminal and barrier；

Based on two dimensional surface coordinate diagram, to mobile robot from the path of origin-to-destination by based on The glowworm swarm algorithm of the initialization of population of Sobol sequence and dynamically adjustment coefficient of disturbance Population Regeneration is carried out Optimum path search, thus in two dimensional surface coordinate diagram planning plan to implement into path optimizing；

According to the path optimizing planned, order about mobile robot and move.

The automatic path planning method of mobile robot the most according to claim 1, its feature It is, described based on two dimensional surface coordinate diagram, mobile robot is passed through from the path of origin-to-destination The glowworm swarm algorithm of initialization of population based on Sobol sequence and dynamically adjustment coefficient of disturbance Population Regeneration Carry out optimum path search, thus in two dimensional surface coordinate diagram planning plan to implement into the step tool of path optimizing Body includes:

Import the basic parameter of glowworm swarm algorithm, and initialize each basic parameter of glowworm swarm algorithm；

Use Sobol sequence initialization population, produce the position of popN Lampyridea, calculate every The object function of Lampyridea is to obtain corresponding brightness, and selects the conduct optimum position of brightness maximum Put；

Calculate the Attraction Degree of every Lampyridea, by there is the Lampyridea of high-high brightness to guide other Luciola vitticollis The movement of worm, updates the position of every Lampyridea, and recalculates the brightness of Lampyridea；

When reaching maximum search number of times, then export optimum individual and stop algorithm, otherwise, recalculating The Attraction Degree of every Lampyridea.

The automatic path planning method of mobile robot the most according to claim 2, its feature Being, described Sobol sequence is with 2 as base, by one group of direction number V₁, V₂, V₃..., V_i..., V_nGenerate, wherein, V_i∈ (0,1), in Sobol sequence, the value of i-th element jth dimension can Obtain by formula:

The automatic path planning method of mobile robot the most according to claim 2, its feature Being, the computing formula of the Attraction Degree of described Lampyridea is:

${\mathrm{\β}}_{ij}={\mathrm{\β}}_0*e^{-{\mathrm{\γd}}_{ij}^2};$

In formula, β₀Be two Lampyridea distances be captivation when zero, γ is the absorption coefficient of light, d_ijIt is Distance between Lampyridea i and Lampyridea j.

The automatic path planning method of mobile robot the most according to claim 4, its feature Being, the location updating formula of described Lampyridea is:

X_j(t+1)=X_j(t)+β_ij(X_i(t)-X_j(t))+α(t)*ε_j；

$\mathrm{\α}\left(t\right)={\mathrm{\α}}_{begin}+({\mathrm{\α}}_{begin}-{\mathrm{\α}}_{end})*\[\frac{2t}{T}-{\left(\frac{t}{T}\right)}^2\];$

In formula, X_i(t) and X_jT () is Lampyridea i and the Lampyridea j space bit when the t time iteration respectively Putting, α is coefficient of disturbance, ε_jBeing random number vector, T is iterations.

6. the mobile robot of an automatic path planning, it is characterised in that including:

Environment information acquisition module, is used for gathering environmental information；

Environmental information MBM, for by the environmental information that collects mobile robot prepared into The region of row path planning is modeled building two dimensional surface coordinate diagram, and determine starting point, terminal and The coordinate position of barrier；

Path planning module, for based on two dimensional surface coordinate diagram, to mobile robot from starting point to end The path of point is by initialization of population based on Sobol sequence and dynamically adjusts coefficient of disturbance Population Regeneration Glowworm swarm algorithm carry out optimum path search, thus in two dimensional surface coordinate diagram planning plan to implement into optimization Path；

Mobile driving module, for according to the path optimizing planned, orders about mobile robot and carries out Mobile.

The mobile robot of automatic path planning the most according to claim 6, it is characterised in that Described path planning module specifically includes:

Basic parameter input block, for importing the basic parameter of glowworm swarm algorithm, and initializes Luciola vitticollis Each basic parameter of worm algorithm；

Sobol sequence initialization kind group unit, is used for using Sobol sequence initialization population, produces The position of popN Lampyridea；

Dynamic disturbances coefficient optimum path search unit, for calculating the object function of every Lampyridea to obtain Corresponding brightness, and select brightness maximum as optimal location；And calculate every Lampyridea Attraction Degree, by having the Lampyridea of high-high brightness to guide the movement of other Lampyrideas, updates every firefly The position of fireworm, and recalculate the brightness of Lampyridea；When reaching maximum search number of times, then export Excellent individuality also stops algorithm, otherwise, recalculates the Attraction Degree of every Lampyridea.

The mobile robot of automatic path planning the most according to claim 7, it is characterised in that Described Sobol sequence is with 2 as base, by one group of direction number V₁, V₂, V₃..., V_i..., V_n Generate, wherein, V_i∈ (0,1), in Sobol sequence, the value of i-th element jth dimension can be pressed Formula obtains:

The mobile robot of automatic path planning the most according to claim 7, it is characterised in that The computing formula of the Attraction Degree of described Lampyridea is:

${\mathrm{\β}}_{ij}={\mathrm{\β}}_0*e^{-{\mathrm{\γd}}_{ij}^2};$

In formula, β₀Be two Lampyridea distances be captivation when zero, γ is the absorption coefficient of light, d_ijIt is Distance between Lampyridea i and Lampyridea j.

The mobile robot of automatic path planning the most according to claim 9, it is characterised in that The location updating formula of described Lampyridea is:

X_j(t+1)=X_j(t)+β_ij(X_i(t)-X_j(t))+α(t)*ε_j；

$\mathrm{\α}\left(t\right)={\mathrm{\α}}_{begin}+({\mathrm{\α}}_{begin}-{\mathrm{\α}}_{end})*\[\frac{2t}{T}-{\left(\frac{t}{T}\right)}^2\];$

In formula, X_i(t) and X_jT () is Lampyridea i and the Lampyridea j space bit when the t time iteration respectively Putting, α is coefficient of disturbance, ε_jBeing random number vector, T is iterations.