CN105025619A - Method for adjusting brightness of light source in response to dark environment on the basis of robot motion process - Google Patents

Abstract

The invention provides a method for adjusting the brightness of a light source in response to a dark environment on the basis of a robot motion process. The method is characterized by obtaining a mathematic model among three items including a light source brightness grade, image brightness quality, and robot observation distance by performing a calibration experiment on a robot vision detection system; adding the mathematic model into a robot motion locus planning optimization process, establishing an optimization model with optimal time and light intensity, solving the optimization model to obtain a robot motion locus path with optimal time and an optimal light intensity locus for guaranteeing image quality; and sending a light source brightness control instruction to a light source controller according to the obtained optimal light intensity locus while a robot motions along a planned locus. The method automatically adjust the brightness of the light source in the robot motion process so as to balance a change in the robot observation distance, thereby guaranteeing the imaging quality of a camera installed on the end actuator of the robot.

Description

Based on the light-source brightness control method of tackling dark surrounds in robot kinematics

Technical field

The present invention relates to robot, industry spot scene visual detection field, a kind of light-source brightness control method based on tackling dark surrounds in robot kinematics specifically, in the process that robot moves along the movement locus planned, light-source brightness regulates the image quality ensureing robot end's camera automatically, avoids occurring that image crosses dark or excessively bright situation.

Background technology

Extreme (the high temperature of manual work is not suitable at some, dark etc.) in operational environment, camera and LED light source is fixedly mounted usually through at robot arm end, remote operating manipulator motion is utilized to carry out the field data of test and monitoring extreme environment inside, for follow-up attended operation provides important information.The camera imaging quality of mechanical arm tail end is most important to whole vision detection system, and in the environment of dark, due to the motion of mechanical arm, the observed range on the surface of camera and light source distance observation thing is not fixing, but can change, when system has requirement to figure picking rate, camera exposure parameter is usually changeless, now, do not regulate if light-source brightness is constant, imageing sensor just there will be under-exposed or saturated, image frame is caused to be crossed dark or excessively bright, lose the detailed information on inspected object surface, therefore when the light durability of light source and uniformity are all guaranteed, the brightness regulation of light source becomes the key factor obtaining high-quality original image.Along with the raising of industrial automatization, manual setting and adjustment are no longer only confined to the control of light source, in order in robot kinematics, reach the brightness of automatic control and adjustment light source, ensure camera imaging quality, be necessary to study the light-source brightness control method that a kind of vision inspection process based on robot motion always tackles dark surrounds.

Summary of the invention

For defect of the prior art, the object of this invention is to provide a kind of light-source brightness control method based on tackling dark surrounds in robot kinematics.

When in the process that robot moves under dark surrounds, in order to ensure vision-based detection camera imaging quality and realize the automation adjustment of light-source brightness, the invention provides a kind of light-source brightness control method based on tackling dark surrounds in robot kinematics, the method is first by carrying out calibration experiment to robot vision detection system, obtain light-source brightness grade, Mathematical Modeling between image brightness quality and robot observed range three, then this Mathematical Modeling is used in the optimizing process of robot motion's trajectory planning, thus obtain the optimum light-source brightness value track corresponding with movement locus, thus in the process of robot along the orbiting motion optimized, light-source brightness value ensures camera imaging quality according to optimizing the light-source brightness value track adjustment obtained.

For realizing above-mentioned technical purpose, the technical solution adopted in the present invention is: based on the light-source brightness control method of tackling dark surrounds in robot kinematics.Specifically comprise the following steps:

Step 1: in order to quantize the subjective assessment of people to image brightness quality, set up image brightness quality evaluation index A.

Step 2: in the robot vision detection system formed primarily of robot, camera, light source, light source controller, camera and light source are installed on the end effector of robot, due to the material properties on examined object surface, the brightness degree scope of light source controller and the difference of observed range scope, need to carry out calibration experiment to robot vision detection system, for the Mathematical Modeling set up between light-source brightness grade E, image brightness quality A and observed range D three provides experimental data.

Step 3: the experimental data provided according to step 2 is carried out data fitting and set up Mathematical Modeling A=F (E between light-source brightness grade E, image brightness quality A and observed range D, D), wherein, F (E, D) represents the function about E, D.

Step 4: be used in the optimizing process of robot trajectory planning about the Mathematical Modeling between A, E, D by what obtain in step 3, sets up robot motion's track optimizing model of a time and light-source brightness optimum

Step 5: by the Optimized model in solution procedure 4, obtains robot motion's track of shortest time and the optimum light-source brightness value track corresponding with movement locus.In the motion process of robot when observed range change, this light-source brightness value track can ensure the image quality of the camera being arranged on robot end, there will not be image to cross dark or excessively bright situation.

Step 6: while robot motion's orbiting motion of the shortest time that robot obtains according to step 5, also the optimum light intensity brightness value track corresponding with movement locus obtained according to step 5 to light source controller issues brightness regulation control command with in accordance with track, thus observed range can be solved change the image brought and cross dark or excessively bright problem, ensure the image quality of camera in robot kinematics.

In technique scheme, step 1 specifically comprises: for realizing being changed by image brightness quality accurate evaluation light intensity, the HSV color space be close with human vision is selected to carry out graphical analysis, after original image is HSV color space by RGB color space conversion, the average gray value of computed image V component (luminance component) is as image brightness quality evaluation index A, and computing formula is wherein, the size of image is the distribution of gray value at the plane of delineation that M*N, I (u, v) represent V component, the lateral coordinates under u presentation video plane pixel coordinates system, the along slope coordinate under v presentation video plane coordinate system.

In technique scheme, step 2 specifically comprises: carry out calibration experiment to robot vision detection system, and detailed process is:

Step 2.1: the first observed range D of fixed camera distance examined object, then the brightness degree scope of light source is divided into K part (such as 250 parts), change light-source brightness grade and obtain corresponding image, computed image brightness quality evaluation index A.

Step 2.2: fix certain light-source brightness grade (such as 130), then changes observed range D, obtains correspondence image and computed image brightness quality evaluation index A.

In technique scheme, step 3 specifically comprises: set up image brightness quality A, light-source brightness grade E, the triangular Mathematical Modeling A=F (E, D) of observed range D, and detailed process is:

Step 3.1: because A=F (E, D) is a nonlinear complex relationship, be convenience and the accuracy of matching, taken the logarithm ln A=lnF (E, D) in A=F (E, D) formula both sides, be designated as:

a＝ln A,e＝lnE,d＝lnD，

Obtain: a=f (e, d).

Step 3.2: by the experimental data obtained in step 2 (1), to each variable E, after A takes the logarithm, carries out multinomial linear fit, obtains linear fit result a=f ₁(e)=P ₁e ³+ P ₂e ²+ P ₃e+P ₄; Wherein, f ₁e () represents the functional relation expression formula between the result that linear fit obtains and a and e, P ₁, P ₂, P ₃, P ₄polynomial constant coefficient in representative function expression formula.

Step 3.3: in like manner, by the experimental data obtained in step 2 (2), to each variables D, after A takes the logarithm, carries out multinomial linear fit and obtains wherein, f ₂d () represents the functional relation expression formula of linear fit result and a and d, polynomial constant coefficient in representative function relational expression

Step 3.4: comprehensive (2) and (3) fitting result is by following formula

The expression of a=f (e, d) can be obtained:

$a=f(e,d)=P_1{e}^3+P_2{e}^2+P_{3}e+p_1^1{d}^2+p_2^{1}d+C.$ Wherein, a ₀presentation video brightness quality evaluation index A initial value take the logarithm after value, e ₀represent light-source brightness E initial value take the logarithm after value, d ₀represent observed range D initial value take the logarithm after value, C represents the constant entry value in the concrete function expression of a=f (e, d).

Step 3.5: a=ln A, e=lnE, d=lnD substitute in the expression of the described a=f (e, d) obtained in step 3.4 can obtain A=F (E, D) expression formula.

In technique scheme, step 4 specifically comprises: robot motion's track optimizing model of settling time and light-source brightness optimum, and detailed process is:

Step 4.1: establish Q ₁, Q ₂..., Q _nbe the point sequence of trajectory planning process in operating space, provide the perfect light source brightness degree E of each point accordingly ₁, E ₂..., E _n, by the point sequence Q in operating space ₁~ Q _nafter the joint variable that inverse kinematics is converted in joint of mechanical arm space, at adjacent set point sequence Q _iwith Q _i+1joint variable time dependent cubic polynomial function expression s is obtained according to cubic spline interpolation between corresponding joint variable _i(t)=c _i4t ³+ c _i3t ²+ c _i2t+c _i1, wherein, i=1,2 ..., n; c _i4, c _i3, c _i2, c _i1the constant coefficient of representative polynomial function expression, t represents through adjacent set point sequence Q _iwith Q _i+1the time interval;

In like manner, at adjacent set point sequence Q _iwith Q _i+1corresponding E _iwith E _i+1between utilize cubic spline interpolation to obtain light-source brightness grade time dependent cubic polynomial function expression E _i(t)=λ _i4t ³+ λ _i3t ²+ λ _i2t+ λ _i1, wherein i=1,2 ..., n; λ _i4, λ _i3, λ _i2, λ _i1the constant coefficient of representative polynomial function expression, t represents through adjacent set point sequence Q _iwith Q _i+1the time interval.

Step 4.2: two target functions of setting Optimized model,

$\left\{\begin{array}{c}\min \underset{i=1}{\overset{n-1}{\mathrm{\Σ}}}{h}_i\\ {\min }_T{\mathrm{\ΔA}}^2={\left(F\right\{E(t+T),D(t+T)\}-F\{E\left(t\right),D\left(t\right)\left\}\right)}^2\end{array}\right.$

H _irepresent through the time interval of adjacent set point sequence, Δ A represents the knots modification of image brightness quality index A in sampling time T.Min _trepresent oeprator of minimizing, E (t+T) represents the light-source brightness value in t+T moment, D (t+T) represents the observed range in t+T moment, and E (t) represents the light-source brightness value of t, and D (t) represents the observed range of t;

Step 4.3: the constraints of given optimization object function:

$\left\{\begin{array}{c}\max \left(\right|{\stackrel{\&CenterDot;}{\mathrm{\&theta;}}}_j\left(t\right)\left|\right)\&le;\sup \left({\mathrm{Vc}}_j\right)\\ \max \left(\right|{\stackrel{\&CenterDot;\&CenterDot;}{\mathrm{\&theta;}}}_j\left(t\right)\left|\right)\&le;\sup \left({\mathrm{Ac}}_j\right)\\ D=D\left(t\right)\\ 0\&le;E\left(t\right)\&le;E_H\\ A_L<A\left(t\right)<A_H\\ \max \left(\right|\stackrel{\&CenterDot;}{E}\left(t\right)\left|\right)\&le;\mathrm{\&zeta;}\end{array}\right.$

Wherein, sup (Vc _j) and sup (Ac _j) be the kinematic constraint upper limit of a jth joint of mechanical arm in joint space, represent the higher limit of speed, the higher limit of acceleration respectively.A _l, A _hrepresent image brightness quality evaluation index higher limit, lower limit that artificial subjective judgement image brightness quality is suitable, E _hrepresent the higher limit of light-source brightness grade, ζ is then the higher limit of light-source brightness change of rank amount. represent the speed of a jth joint of mechanical arm in t joint space, represent the acceleration of a jth joint of mechanical arm in t joint space, A (t) represents t image brightness quality evaluation index, represent t light-source brightness rate of change; J is positive integer.

In technique scheme, step 5 specifically comprises: the Model for Multi-Objective Optimization in solution procedure 4, obtains robot motion's track of shortest time and the optimum light-source brightness value track of the guarantee image brightness quality corresponding with movement locus.

In technique scheme, step 6 specifically comprises: issue regulating and controlling instruction according to the movement locus and optimum light-source brightness value track that obtain shortest time in step 5 to Robot Motion Controller and light source controller, thus ensures the image quality of robot end's camera in robot kinematics.

Compared with prior art, the present invention has following beneficial effect:

The novel part of the inventive method is by setting up light-source brightness grade E, Mathematical Modeling between the observed range D of camera distance determinand and image brightness quality evaluation index A three, and this Mathematical Modeling is added in the process of robot motion's track optimizing, the kinematical constraint of joint space is only considered compared to traditional track optimizing model, Optimized model in the inventive method also contemplates vision-based detection demand in task space and high-quality original image, eliminate observed range in motion process by regulating light-source brightness value to change the image brought and cross bright or excessively dark situation, thus ensure camera imaging quality.

Accompanying drawing explanation

By reading the detailed description done non-limiting example with reference to the following drawings, other features, objects and advantages of the present invention will become more obvious:

Fig. 1 is the inventive method flow chart.

Fig. 2 is embodiment of the present invention robot vision detection system structural representation.

In figure: the cross section of the airtight device to be detected of 1-, 2-mechanical arm, 3-camera and LED light source

Embodiment

Below in conjunction with specific embodiment, the present invention is described in detail.Following examples will contribute to those skilled in the art and understand the present invention further, but not limit the present invention in any form.It should be pointed out that to those skilled in the art, without departing from the inventive concept of the premise, some changes and improvements can also be made.These all belong to protection scope of the present invention.

Present disclosure provides a kind of light-source brightness control method based on tackling dark surrounds in robot kinematics.Specifically, the method is by carrying out calibration experiment to robot vision detection system, obtain light-source brightness grade, Mathematical Modeling between image brightness quality and robot observed range three, then in robot motion's trajectory planning optimizing process, this Mathematical Modeling is added, set up the Optimized model of a time and light intensity optimum, also one is obtained for ensureing the optimum light intensity track of picture quality while obtaining time optimal robot motion's trajectory path by solving-optimizing model, finally while robot is according to the orbiting motion planned, light-source brightness control command is issued to light source controller according to the optimum light intensity track obtained.Being can be implemented in robot kinematics by the inventive method regulates light-source brightness to carry out the change of balanced robot's observed range, to ensure the image quality of the camera be arranged on end effector of robot automatically.

The present embodiment provides a kind of light-source brightness control method based on tackling dark surrounds in robot kinematics.Fig. 2 is embodiment of the present invention robot vision detection system structural representation, and our object is the inner surface utilizing the camera on remote operating mechanical arm tail end to remove to detect airtight cavity, is without illumination dark surrounds in chamber.As shown in Figure 1, the method includes the steps of for the inventive method flow chart:

Step 1: in order to quantize the subjective assessment of people to image brightness quality, set up image brightness quality evaluation index A.

Step 2: carry out calibration experiment to robot vision detection system, for the Mathematical Modeling set up between light-source brightness grade E, image brightness quality A and observed range D three provides experimental data to prepare.

Step 3: the experimental data provided according to step 2 is carried out data fitting and set up light-source brightness grade E, the Mathematical Modeling between image brightness quality A and observed range D, A=F (E, D).

Step 4: be used in the optimizing process of robot trajectory planning about the Mathematical Modeling between A, E, D by what obtain in step 3, sets up robot motion's track optimizing model of a time and light-source brightness optimum.

Step 5: by the Optimized model in solution procedure 4, obtains the track of time optimal robot motion's track and the optimum light-source brightness value corresponding with movement locus.

Step 6: while robot moves according to the time optimal movement locus that step 5 obtains, also the optimum light intensity brightness value track obtained according to step 5 to light source controller issues brightness regulation control command, thus observed range can be solved change the image brought and cross dark or excessively bright problem, ensure the image quality of camera in robot kinematics.

Be detailed implementation step below:

Step 1 is specifically implemented to comprise: for realizing being changed by image brightness quality accurate evaluation light intensity, the HSV color space be close with human vision is selected to carry out graphical analysis, after original image is HSV color space by RGB color space conversion, the average gray value of computed image V component (luminance component) is as image brightness quality evaluation index A, and computing formula is wherein, the size of image is the distribution of gray value at the plane of delineation that M*N, I (u, v) represent V component.Lateral coordinates under u presentation video plane pixel coordinates system, the along slope coordinate under v presentation video plane coordinate system.

Step 2 is specifically implemented to comprise:

Step 2.1: the first observed range D of fixed camera distance examined object, then K part (such as 250 parts) is divided into according to the brightness degree scope of light source controller, change light-source brightness grade and obtain corresponding image, computed image brightness quality evaluation index A.

Step 2.2: fix certain light-source brightness grade (such as 130 parts), then changes observed range D, obtains correspondence image and computed image brightness quality evaluation index A.

Step 3 is specifically implemented to comprise: set up the triangular Mathematical Modeling A=F (E, D) of image brightness quality A, light-source brightness grade E and observed range D, detailed process is:

Step 3.1: because A=F (E, D) is a nonlinear complex relationship, be convenience and the accuracy of matching, taken the logarithm ln A=lnF (E, D) in A=F (E, D) formula both sides, be designated as:

a＝ln A,e＝lnE,d＝lnD

Obtain: a=f (e, d), wherein, f (e, d) represents the function about e, d.

Step 3.2: by the experimental data obtained in step 2 (1), to each variable E, after A takes the logarithm, carries out multinomial linear fit, obtains linear fit result a=f ₁(e)=P ₁e ³+ P ₂e ²+ P ₃e+P ₄; Wherein, f ₁e () represents the functional relation expression formula between the result that linear fit obtains and a and e, P ₁, P ₂, P ₃, P ₄polynomial constant coefficient in representative function expression formula.

Step 3.3: in like manner, by the experimental data obtained in step 2 (2), to each variables D, after A takes the logarithm, carries out multinomial linear fit and obtains wherein, f ₂d () represents the functional relation expression formula of linear fit result and a and d, polynomial constant coefficient in representative function relational expression.

Step 3.4: combining step 3.2 and step 3.3: fitting result is by following formula

The expression of a=f (e, d) can be obtained $a=f(e,d)=P_1{e}^3+P_2{e}^2+P_{3}e+p_1^1{d}^2+p_2^{1}d+C;$ Wherein, a ₀value after presentation video brightness quality evaluation index initial value A takes the logarithm, e ₀value after expression light-source brightness initial value E takes the logarithm, d ₀value after expression observed range initial value D takes the logarithm, C represents the constant entry value in the concrete function expression of a=f (e, d).

Step 3.5: a=ln A, e=lnE, d=lnD substitute in the result a=f (e, d) obtained in (4) can obtain A=F (E, D) expression formula.

Step 4 is specifically implemented to comprise: robot motion's track optimizing model of settling time and light-source brightness optimum, and detailed process is:

(1) Q is established ₁, Q ₂..., Q _nbe the point sequence of trajectory planning process in operating space, provide each point perfect light source brightness degree E accordingly ₁, E ₂..., E _n, by the point sequence Q in operating space ₁~ Q _nafter the joint variable that inverse kinematics is converted in joint of mechanical arm space, at adjacent set point sequence Q _iwith Q _i+1between obtain joint variable time dependent cubic polynomial function expression s according to cubic spline interpolation _i(t)=c _i4t ²+ c _i3t ³+ c _i2t+c _i1, in like manner, at adjacent set point sequence E _iwith E _i+1between utilize cubic spline interpolation to obtain light-source brightness grade time dependent cubic polynomial function expression E _i(t)=λ _i4t ³+ λ _i3t ²+ λ _i2t+ λ _i1, wherein i=1,2 ..., n, c _i4, c _i3, c _i2, c _i1, λ _i4, λ _i3, λ _i2, λ _i1the constant coefficient of representative polynomial function expression.

(2) two target functions of Optimized model are set,

$\left\{\begin{array}{c}\min \underset{i=1}{\overset{n-1}{\mathrm{\&Sigma;}}}{h}_i\\ {\min }_T{\mathrm{\&Delta;A}}^2={\left(F\right\{E(t+T),D(t+T)\}-F\{E\left(t\right),D\left(t\right)\left\}\right)}^2\end{array}\right.$

H _irepresent the time interval of adjacent set point sequence, Δ A represents the knots modification of image brightness quality index in sampling time T; min _trepresent oeprator of minimizing, E (t+T) represents the light-source brightness value in t+T moment, D (t+T) represents the observed range in t+T moment, and E (t) represents the light-source brightness value of t, and D (t) represents the observed range of t.

(3) constraints of given optimization object function:

$\left\{\begin{array}{c}\max \left(\right|{\stackrel{\&CenterDot;}{\mathrm{\&theta;}}}_j\left(t\right)\left|\right)\&le;\sup \left({\mathrm{Vc}}_j\right)\\ \max \left(\right|{\stackrel{\&CenterDot;\&CenterDot;}{\mathrm{\&theta;}}}_j\left(t\right)\left|\right)\&le;\sup \left({\mathrm{Ac}}_j\right)\\ D=D\left(t\right)\\ 0\&le;E\left(t\right)\&le;E_H\\ A_L<A\left(t\right)<A_H\\ \max \left(\right|\stackrel{\&CenterDot;}{E}\left(t\right)\left|\right)\&le;\mathrm{\&zeta;}\end{array}\right.$

Wherein, sup (Vc _j) and sup (Ac _j) be the kinematic constraint upper limit in a jth joint in joint space, represent speed respectively, the higher limit of acceleration.A _l, A _hrepresent the image brightness quality evaluation index upper lower limit value that artificial subjective judgement image brightness quality is suitable, E _hrepresent the higher limit of light-source brightness grade, ζ is then the higher limit of light-source brightness change of rank amount, represent the speed in a jth joint in t joint space, represent the acceleration in a jth joint in t joint space, A (t) represents t image brightness quality evaluation index, represent t light-source brightness rate of change.

Step 5 is specifically implemented to comprise: the Model for Multi-Objective Optimization in solution procedure 4, obtains robot motion's track of shortest time and the optimum light-source brightness value track of the guarantee image brightness quality corresponding with movement locus.

Step 6 is specifically implemented to comprise: issue regulating and controlling instruction according to the movement locus and optimum light-source brightness value track that obtain shortest time in step 5 to Robot Motion Controller and light source controller, thus ensure the image quality of robot end's camera in robot kinematics.

Above specific embodiments of the invention are described.It is to be appreciated that the present invention is not limited to above-mentioned particular implementation, those skilled in the art can make a variety of changes within the scope of the claims or revise, and this does not affect flesh and blood of the present invention.

Claims (5)

1., based on a light-source brightness control method of tackling dark surrounds in robot kinematics, it is characterized in that, comprise the following steps:

Step 1: set up image brightness quality evaluation index A;

Step 2: carry out calibration experiment to robot vision detection system, for the Mathematical Modeling set up between light-source brightness grade E, image brightness quality A and observed range D three provides experimental data;

Step 3: the experimental data provided according to step 2 is carried out data fitting and set up Mathematical Modeling A=F (E between light-source brightness grade E, image brightness quality A and observed range D, D), wherein, F (E, D) represents the function about E, D;

Step 4: according to obtain in step 3 about the Mathematical Models robot motion track optimizing model between A, E, D;

Step 5: by the Optimized model in solution procedure 4, obtains the track of robot motion's track of shortest time, the optimum light-source brightness value corresponding with robot motion's track of shortest time;

Step 6: according to robot motion's track, the optimum light-source brightness value track corresponding with robot motion's track of shortest time of the shortest time that Optimization Solution in step 5 obtains, regulating and controlling instruction is issued to Robot Motion Controller and light source controller, thus observed range can be solved change the image brought and cross dark or excessively bright problem, ensure the image quality of camera in robot kinematics.

2. the light-source brightness control method based on tackling dark surrounds in robot kinematics according to claim 1, it is characterized in that, described step 1 specifically comprises:

After original image is HSV color space by RGB color space conversion, the average gray value of computed image luminance component V is as image brightness quality evaluation index A, and computing formula is:

$A=\frac{1}{\mathrm{MN}}\underset{u=0}{\overset{M-1}{\mathrm{\&Sigma;}}}\underset{v=0}{\overset{N-1}{\mathrm{\&Sigma;}}}I(u,v)$

Wherein, the size of original image is the distribution of gray value at the plane of delineation that M*N, I (u, v) represent V component, the lateral coordinates under u presentation video plane pixel coordinates system, the along slope coordinate under v presentation video plane coordinate system.

3. the light-source brightness control method based on tackling dark surrounds in robot kinematics according to claim 1, it is characterized in that, described step 2 specifically comprises:

Step 2.1: the first observed range D of the camera distance examined object of stationary machines people vision detection system, then the brightness degree scope of light source is divided into K part, change light-source brightness grade and obtain corresponding original image, calculating the image brightness quality evaluation index A of original image;

Step 2.2: fixed light source brightness degree, then changes observed range D, obtains corresponding original image and calculates the image brightness quality evaluation index A of original image.

4. the light-source brightness control method based on tackling dark surrounds in robot kinematics according to claim 3, it is characterized in that, described step 3 specifically comprises:

Step 3.1: taken the logarithm lnA=lnF (E, D) in A=F (E, D) formula both sides, be designated as:

a＝lnA,e＝lnE,d＝lnD，

Obtain a=f (e, d), wherein, f (e, d) represents the function about e, d;

Step 3.2: by the experimental data obtained in step 2.1, to each variable E, after A takes the logarithm, carries out multinomial linear fit, obtains linear fit result a=f ₁(e)=P ₁e ³+ P ₂e ²+ P ₃e+P ₄; Wherein, f ₁e () represents the result that linear fit obtains, the functional relation expression formula namely between a and e, P ₁, P ₂, P ₃, P ₄polynomial constant coefficient in representative function expression formula;

Step 3.3: by the experimental data obtained in step 2.2, to each variables D, after A takes the logarithm, carries out multinomial linear fit, obtains linear fit result wherein, f ₂d () represents linear fit result, i.e. the functional relation expression formula of a and d, polynomial constant coefficient in representative function relational expression;

Step 3.4: combining step 3.2 and step 3.3 fitting result are by following formula:

The expression of a=f (e, d) can be obtained:

$a=f(e,d)=P_1{e}^3+P_2{e}^2+P_{3}e+p_1^1{d}^2+p_2^{1}d+C$

Wherein, a ₀presentation video brightness quality evaluation index A initial value take the logarithm after value, e ₀represent light-source brightness E initial value take the logarithm after value, d ₀represent observed range D initial value take the logarithm after value, C represents the constant entry value in a=f (e, d);

Step 3.5: a=lnA, e=lnE, d=lnD are substituted in the expression of the described a=f (e, d) obtained in step 3.4 and obtain A=F (E, D) expression formula.

5. the light-source brightness control method based on tackling dark surrounds in robot kinematics according to claim 4, it is characterized in that, described step 4 specifically comprises:

Step 4.1: establish Q ₁, Q ₂..., Q _nbe the point sequence of n point of trajectory planning process in operating space, provide perfect light source brightness degree E corresponding to described n point respectively ₁, E ₂..., E _n, by the point sequence Q in operating space ₁~ Q _nafter being converted to the joint variable in joint of mechanical arm space, at adjacent set point sequence Q _iwith Q _i+1obtain the cubic polynomial function expression s of joint variable t change in time according to cubic spline interpolation between corresponding joint variable _i(t):

s _i(t)＝c _i4t ³+c _i3t ²+c _i2t+c _i1，i＝1,2,……,n-1

Wherein, c _i4, c _i3, c _i2, c _i1the constant coefficient of representative polynomial function expression, t represents through adjacent set point sequence Q _iwith Q _i+1the time interval;

At adjacent set point sequence Q _iwith Q _i+1corresponding E _iwith E _i+1between utilize cubic spline interpolation obtain light-source brightness grade in time t change cubic polynomial function expression E _i(t):

E _i(t)＝λ _i4t ³+λ _i3t ²+λ _i2t+λ _i1，i＝1,2,……,n-1

Wherein, λ _i4, λ _i3, λ _i2, λ _i1the constant coefficient of representative polynomial function expression, t represents through adjacent set point sequence Q _iwith Q _i+1the time interval;

Step 4.2: two target functions setting described Optimized model:

$\left\{\begin{array}{c}\min \underset{i=1}{\overset{n-1}{\mathrm{\&Sigma;}}}{h}_i\\ {\min }_T\mathrm{\&Delta;}{A}^2={\left(F\right\{E(t+T),D(t+T)\}-F\{E\left(t\right),D\left(t\right)\left\}\right)}^2\end{array}\right.$

Wherein, h _irepresent through adjacent set point sequence Q _iwith Q _i+1the time interval, Δ A represents the knots modification of image brightness quality index A in sampling time T; min _trepresent oeprator of minimizing, E (t+T) represents the light-source brightness value in t+T moment, D (t+T) represents the observed range in t+T moment, and E (t) represents the light-source brightness value of t, and D (t) represents the observed range of t;

Step 4.3: given described bound for objective function:

$\left\{\begin{array}{c}\max \left(\right|{\stackrel{.}{\mathrm{\&theta;}}}_j\left(t\right)\left|\right)\&le;\sup \left({\mathrm{Vc}}_j\right)\\ \max \left(\right|{\stackrel{..}{\mathrm{\&theta;}}}_j\left(t\right)\left|\right)\&le;\sup \left({\mathrm{Ac}}_j\right)\\ D=D\left(t\right)\\ 0\&le;E\left(t\right)\&le;E_H\\ A_L<A\left(t\right)<A_H\\ \max \left(\right|\stackrel{.}{E}\left(t\right)\left|\right)\&le;\mathrm{\&zeta;}\end{array}\right.$

Wherein, sup (Vc _j) and sup (Ac _j) be the kinematic constraint upper limit of a jth joint of mechanical arm in joint space, represent the higher limit of speed, the higher limit of acceleration respectively; A _l, A _hhigher limit, the lower limit of presentation video brightness quality evaluation index; E _hrepresent the higher limit of light-source brightness grade; ζ is then the higher limit of light-source brightness change of rank amount; represent the speed of a jth joint of mechanical arm in t joint space, represent the acceleration of a jth joint of mechanical arm in t joint space, A (t) represents t image brightness quality evaluation index, represent t light-source brightness rate of change; J is positive integer.