CN101685309A - Method for controlling multi-robot coordinated formation - Google Patents

Abstract

The invention provides a method for controlling multi-robot coordinated formation, which is formation control realized by coordinating an accompanying robot with a pilot robot through recurrence predication. The accompanying robot acquires distances and observed azimuths d <k-s>, phi <k-s> (s=0,..., Np max relatively to the pilot robot at current sampling time k and previous Np max times of sampling time by combination with vision and coded disc information; reoccurring to acquire the position L<K>(xL<k>, yL<k>) (see in formula 2) of the pilot robot under a local coordinate system (see formula1) of the accompanying robot at the sampling time k and k-Np max and the position Of<k> (xof<k>, yof<k>) of a photocenter Copt (see formula 3) of a forward camera of the accompanying robot at the sampling time k through conversion of the local coordinate system by means of coded disc information; calculating an angle psiij<k> between the predicted moving direction of the pilot robot and the direction from the pilot robot to the Copt so as to combine an expected angle psiijd, distance Lij<k> (namely d<k>) between the accompanying robot and the pilot robot and the expected distance Lijd, thus combining vision information, and carrying out motion decision to acquire speed vf and a turn angle thetaf. The method is suitable for conditions of communication failures, and provides technical support for application of multi-robot in military and other fields.

Description

Method for controlling multi-robot coordinated formation

Technical field

The present invention relates to the Robotics field, particularly a kind of method for controlling multi-robot coordinated formation is used for multi-robot system.

Background technology

The control of multirobot formation, the team that refers to a plurality of robots composition under the condition of constraint that conforms, keeps predetermined geometric relationship each other in the process of executing the task.The multirobot formation has broad application prospects at aspects such as military affairs, Aero-Space, and its realization helps shortening the task executions time, improves the overall performance of system.Formation control method commonly used has pilotage people's follower method, based on the method and the virtual architecture method of behavior, a lot of research work are devoted to these methods are studied and improved to improve system's operation stability and environmental suitability.

For a system that forms by N robot, a kind of thinking is that the formation control problem that formation control problem is decomposed into two robots of N-1 group is realized, one in two robots of every group are pilot robot for following robot, one, and they are devoted to keep predefined geometric configuration.By the formation control in twos of robot in the system, present the shape of coordination on the whole.In order to realize formation control, follow the relevant information that robot need obtain pilot robot, the means of the information of obtaining commonly used are communications.Consider the problems such as error accumulation of communication network delay, data-bag lost and even communication failures and topworks, design a kind ofly do not have communication, the method by the perceptual inference of its pilot robot being realized coordinate formation control is significant.

Summary of the invention

The purpose of this invention is to provide a kind of method for controlling multi-robot coordinated formation, be based on the method for perceptual inference, make that following robot relies on perceptual inference to realize and the formation control of pilot robot, reaches gratifying effect.

For achieving the above object, technical solution of the present invention is:

A kind of method for controlling multi-robot coordinated formation comprises step:

A) follow robot according to the colour code tube on the pilot robot,, obtain current sampling instant k and N before in conjunction with vision and code-disc information _PmaxInferior sampling instant is with respect to the distance and the observed azimuth d of pilot robot ^K-s,

B) follow robot by code-disc information, by the local coordinate system conversion, recursion obtains sampling instant k and k-N _{P max}The time pilot robot following the robot local coordinate system

Under position L ^k(x _L ^k, y _L ^k), And the photocentre C that follows robot forward direction video camera during current sampling instant k _Opt

Position O under the coordinate system _f ^k(x _Of ^k, y _Of ^k);

C) position-based L ^k,

And O _f ^k, follow robot and calculate the pilot robot direction of motion of prediction and pilot robot to C _OptAngle ψ between the direction _Ij ^k

D) in conjunction with angle ψ _Ij ^kAnd expected angle ψ _Ijd, follow the distance L between robot and the pilot robot _Ij ^k(be d ^k) and desired distance L _Ijd, comprehensive visual information, the decision-making of moving obtains speed v _fAnd rotational angle theta _f, realize following the control of robot speed's size and Orientation.

Described method for controlling multi-robot coordinated formation, the colour code tube on its pilot robot is the cylindrical tube of hollow, is combined up and down by at least two kinds of colors, is put in the robot, the center of its center and robot is consistent.

Described method for controlling multi-robot coordinated formation, the control of its formation are to realize there is not communication between the robot on the basis of robot to the pilot robot perceptual inference following.

Described method for controlling multi-robot coordinated formation, its described B) follows the robot local coordinate system in the step

Be and sampling instant k-N _PmaxCorresponding, its initial point

Get forward direction video camera photocentre C _OptThe position at place,

Direction and sampling instant k-N _PmaxThe time robot motion's direction be consistent.

Described method for controlling multi-robot coordinated formation, its described C) following the prediction of robot to pilot robot direction of motion in the step, is according to current sampling instant k and this moment N before _PmaxThe pilot robot positional information of inferior sampling instant draws, and the pilot robot positional information of these two sampling instants all is based upon the N that follows before the sampling instant k of robot _PmaxUnder the pairing local coordinate system of inferior sampling instant.

Described method for controlling multi-robot coordinated formation, its described D) follows distance L between robot and the pilot robot in the step _Ij ^kBe d ^k

Described method for controlling multi-robot coordinated formation, its described N _Pmax〉=1.

Method for controlling multi-robot coordinated formation of the present invention, the situation of suitable communication failures is for the application of multirobot at aspects such as military affairs provides technical support.

Description of drawings

Fig. 1 is the control block diagram of a kind of method for controlling multi-robot coordinated formation of the present invention;

Fig. 2 is two robot formation control charts;

Fig. 3 is the recursion prognostic chart of a kind of method for controlling multi-robot coordinated formation of the present invention;

Fig. 4 be a kind of method for controlling multi-robot coordinated formation of the present invention follow robot motion's decision diagram;

Fig. 5 is with pilot robot of the inventive method and the movement locus of following robot.

Embodiment

The invention provides a kind of method for controlling multi-robot coordinated formation.Follow robot in conjunction with vision and self code-disc information, obtain current sampling instant k and N before _PmaxInferior sampling instant is with respect to the distance and the observed bearing angle information of pilot robot.By code-disc information, by the local coordinate system conversion, recursion obtains the N before sampling instant k and this moment _PmaxPilot robot during inferior sampling instant is being followed the robot local coordinate system (with sampling instant k-N _PmaxCorresponding) under the position, and the position of photocentre under this local coordinate system of following robot forward direction video camera (its optical axis direction is consistent with robot motion's direction) during current sampling instant k.Based on above-mentioned position, follow robot and calculate the pilot robot direction of motion of prediction and pilot robot to the angle between the photocentre direction of following robot forward direction video camera, and then in conjunction with pilot robot direction of motion and pilot robot to the expected angle between the photocentre direction of following robot forward direction video camera and follow robot and pilot robot between distance and desired distance, comprehensive visual information, the decision-making of moving, obtain following the speed and the corner of robot, realize following the control of robot speed's size and Orientation.The control block diagram of method for controlling multi-robot coordinated formation as shown in Figure 1, wherein, d ^K-s,

Be current sampling instant k and N before _PmaxInferior sampling instant is with respect to the distance and the observed azimuth of pilot robot, L ^k(x _L ^k, y _L ^k) and

Be respectively sampling instant k and k-N _PmaxThe time pilot robot exist

Coordinate system (is followed robot and sampling instant k-N _PmaxCorresponding local coordinate system) position under, O _f ^k(x _Of ^k, y _Of ^k) the photocentre of following robot forward direction video camera during for current sampling instant k exists

Position under the coordinate system, ψ _Ij ^kAnd ψ _IjdBe respectively the pilot robot direction of motion of prediction and pilot robot to angle and expected angle between the photocentre direction of following robot forward direction video camera, L _Ij ^k(be d ^k) and L _IjdBe respectively the distance and the desired distance of following between robot and the pilot robot, v _fAnd θ _fBe respectively the speed and the corner of following robot.

Fig. 2 is two robot formation control chart, wherein R _iAnd R _jRobot and pilot robot are followed in expression respectively, and l is the distance between the robot two driving wheel centers, and XOY is a world coordinate system, X _fO _fY _fBe selected as R _iLocal coordinate system, O _fGet forward direction video camera C _fPhotocentre C _OptThe position at place, d _cBe O _fAnd the distance between the robot center, Y _fDirection and robot motion's direction be consistent.Work as R _iRecursion dopes pilot robot R _jMotion the time, based on L _Ijdψ _Ijd, obtain expecting formation point P _Fd, along with R _iMarch on towards P _Fd, L _Ij→ L _Ijd, ψ _Ij→ ψ _Ijd, finally set up a kind of coordination formation relation, wherein L _Ij, L _IjdBe respectively R _iWith R _jBetween distance and desired distance, ψ _Ij(ψ _Ij∈ [0,2 π)), ψ _IjdBe respectively R _jDirection of motion and R _jTo R _iAngle and expected angle between the photocentre direction of forward direction video camera.

1, pilot robot location estimation:

Pilot robot is equipped with a colour code tube, and it is the cylindrical tube of a hollow, is combined up and down by at least two kinds of colors, and the colour code tube is put on the pilot robot, and its center and robot center are consistent.Follow robot by the visual identity of colour code tube being finished identification,, and then obtain the relative positioning information of estimation in conjunction with vision calibration to pilot robot.Make (d ^k,

) be distance and the observed azimuth of sampling instant k with respect to pilot robot.

The definition (u v) is that image coordinate is fastened a bit, (x, y, z) be its former be the coordinate system w of initial point to the video camera photocentre _CameraOn coordinate, have:

$\mathrm{<figure-callout\; id="1"\; label="z\; u\; v"\; filenames="A2008102227740016H1.png"\; state="\{\{state\}\}">z</figure-callout>}\left[\begin{array}{c}u\\ v\end{array}\right]\left[\begin{array}{c}\\ 1\end{array}\right]=M\left[\begin{array}{c}x\\ y\\ z\end{array}\right]=\left[\begin{array}{ccc}{\mathrm{\&alpha;}}_x& 0& u_0\\ 0& {\mathrm{\&alpha;}}_y& v_0\\ 0& 0& 1\end{array}\right]\left[\begin{array}{c}x\\ y\\ z\end{array}\right]$

Wherein M is the intrinsic parameter matrix of video camera, is obtained by camera calibration.

Note (u ₁, v ₁) and (u ₂, v ₂) be respectively the point of colour code tube top and bottom, (x ₁, y ₁, z ₁) and (x ₂, y ₂, z ₂) be that they are at w _CameraOn coordinate, u ₁=u ₂, z ₁=z ₂, y ₂-y ₁Be the true altitude of colour code tube, then d ^kCan estimate as follows:

$d^k=z_1=z_2=\frac{{\mathrm{\&alpha;}}_y(y_2-y_1)}{v_2-{\mathrm{<figure-callout\; id="1"\; label="v"\; filenames="A2008102227740016H1.png"\; state="\{\{state\}\}">v</figure-callout>}}_1}$

α in the formula _yObtain v by M ₂-v ₁Be that the pixels tall of colour code tube in image is poor.

Note (u _T, v _T) be the center of colour code tube, u _d, θ _vBe respectively the width of image and the width in the visual field, then

Can estimate as follows:

Influenced by the external environment condition or the visual field etc., follow robot and utilize forward direction video camera C _fMay cannot see pilot robot, need estimate the pilot robot position according to code-disc and former visual information this moment.If follow robot through continuous N _CsInferior sampling does not still observe pilot robot, abandons estimating; See pilot robot in case follow robot by other video camera, its overriding concern be to make the forward direction video camera find pilot robot as early as possible by rotation, at this moment, it abandons estimating.

Make (d ^K-1,

) expression previous sampling instant pilot robot positional information, D _l, D _rExpression left and right wheels of the last one-period distance of travelling respectively then has:

${\mathrm{\&alpha;}}_{k-1}=\frac{D_r-D_l}{l}$

$D_{k-1}=\frac{D_l+D_r}{2}$

Wherein, α _K-1The robot corner of representing last one-period, D _K-1Be the robot center distance of travelling of last one-period.

When following the robot craspedodrome, (d ^k,

) estimation as follows:

When following the robot circular motion, have:

2, prediction pilot robot direction of motion:

In order to predict the direction of motion of pilot robot, follow robot and need obtain current sampling instant k and N before _Pmax(N _Pmax〉=1) inferior sampling instant is with respect to the distance and the observed azimuth d of pilot robot ^K-s,

(s=0 ..., N _Pmax) and relevant code-disc information.

Be defined as and follow robot and sampling instant k-N _PmaxCorresponding local coordinate system, note L ^K-s(x _L ^K-s, y _L ^K-s) (s=0 ..., N _Pmax) be sampling instant k ..., k-N _PmaxThe time pilot robot exist

Position under the coordinate system, O _f ^K-s(x _Of ^K-s, y _Of ^K-s) (s=0 ..., N _Pmax) be sampling instant k ..., k-N _PmaxThe time the photocentre C that follows robot forward direction video camera _Opt

Position under the coordinate system, as shown in Figure 3.

Follow robot at sampling instant k-N _PmaxPose

Can be expressed as:

$T_{O_f}^{k-N_{p\max }}=\left[\begin{array}{ccc}{T}_{11}^{k-N_{p\max }}& T_{12}^{k-N_{p\max }}& T_{13}^{k-N_{p\max }}\\ T_{21}^{k-N_{p\max }}& T_{22}^{k-N_{p\max }}& T_{23}^{k-N_{p\max }}\\ T_{31}^{k-N_{p\max }}& T_{32}^{k-N_{p\max }}& T_{33}^{k-N_{p\max }}\end{array}\right]=\left[\begin{array}{ccc}1& 0& 0\\ 0& 1& 0\\ 0& 0& 1\end{array}\right]$

According to code-disc information, draw follow robot sampling instant k-s (s=0 ..., N _Pmax-1) pose

$T_{O_f}^{k-s}=\left[\begin{array}{ccc}{T}_{11}^{k-s}& T_{12}^{k-s}& T_{13}^{k-s}\\ T_{21}^{k-s}& T_{22}^{k-s}& T_{23}^{k-s}\\ T_{31}^{k-s}& T_{32}^{k-s}& T_{33}^{k-s}\end{array}\right]=\left\{\begin{array}{cc}{T}_{O_f}^{k-s-1}{T}_{\mathrm{line}}^{(k-s-1)(k-s)}& \mathrm{whentype}=1\\ T_{O_f}^{k-s-1}{T}_{\mathrm{arc}}^{(k-s-1)(k-s)}& \mathrm{whentype}=0\end{array}\right.$

Wherein,

$T_{\mathrm{line}}^{(k-s-1)(k-s)}=\left[\begin{array}{ccc}1& 0& 0\\ 0& 1& S^{(k-s-1)(k-s)}\\ 0& 0& 1\end{array}\right],$

$T_{\mathrm{arc}}^{(k-s-1)(k-s)}=\left[\begin{array}{cc}\cos \left({\mathrm{\&theta;}}^{(k-s-1)(k-s)}\right)& -\sin \left({\mathrm{\&theta;}}^{(k-s-1)(k-s)}\right)\\ \sin \left({\mathrm{\&theta;}}^{(k-s-1)(k-s)}\right)& \cos \left({\mathrm{\&theta;}}^{(k-s-1)(k-s)}\right)\\ 0& 0\end{array}\right.$

${\left.\begin{array}{c}(\cos \left({\mathrm{\&theta;}}^{(k-s-1)(k-s)}\right)-1)(r^{(k-s-1)(k-s)}+l/2)-d_c\sin \left({\mathrm{\&theta;}}^{(k-s-1)(k-s)}\right)\\ \sin \left({\mathrm{\&theta;}}^{(k-s-1)(k-s)}\right)(r^{(k-s-1)(k-s)}+l/2)+d_c(\cos \left({\mathrm{\&theta;}}^{(k-s-1)(k-s)}\right)-1)\\ 1\end{array}\right]}_{3\&times;3}$

S ^{(k-s-1) (k-s)}The distance of between sampling instant k-s-1 and k-s, travelling when robot rectilinear motion (type=1) is followed in expression, θ ^{(k-s-1) (k-s)}And r ^{(k-s-1) (k-s)}Central angle and the radius of representing the circular arc of correspondence when following robot does circular motion (type=0) from sampling instant k-s-1 to k-s respectively.

In conjunction with distance and observed azimuth, follow robot and estimate that pilot robot exists with respect to pilot robot

Position L under the coordinate system ^K-s(x _L ^K-s, y _L ^K-s) (s=0 ..., N _Pmax) as follows:

Based on

With L ^k, follow robot and obtain

Coordinate system is the pilot robot direction of motion of prediction down, its direction angle alpha ^K+1For:

$\tan {\mathrm{\&alpha;}}^{k+1}=\frac{y_L^k-y_L^{k-N_{p\max }}}{x_L^k-x_L^{k-N_{p\max }}}(x_L^k\&NotEqual;x_L^{k-N_{p\max }})\left|\right|{\mathrm{\&alpha;}}^{k+1}=\frac{\mathrm{\&pi;}}{2}(x_L^k=x_L^{k-N_{p\max }})$

3, calculate pilot robot direction of motion and pilot robot to the photocentre C that follows robot forward direction video camera _OptAngle ψ between the direction _Ij ^k:

Follow the photocentre C of robot forward direction video camera _Opt

Position O under the coordinate system _f ^K-s(x _Of ^K-s, y _Of ^K-s) (s=0 ..., N _Pmax) be calculated as follows:

$\left[\begin{array}{c}{x}_{\mathrm{of}}^{k-s}\\ y_{\mathrm{of}}^{k-\mathrm{<figure-callout\; id="1"\; label="s"\; filenames="A2008102227740016H1.png"\; state="\{\{state\}\}">s</figure-callout>}}\\ 1\end{array}\right]=T_{O_f}^{k-s}\left[\begin{array}{c}0\\ 0\\ 1\end{array}\right]$

Based on O _f ^kAnd L ^k, obtain β ^k(see figure 3):

$\tan {\mathrm{\&beta;}}^k=\frac{y_L^k-y_{\mathrm{of}}^k}{x_L^k-x_{\mathrm{of}}^k}(x_L^k\&NotEqual;x_{\mathrm{of}}^k)\left|\right|{\mathrm{\&beta;}}^k=\frac{\mathrm{\&pi;}}{2}(x_L^k=x_{\mathrm{of}}^k)$

So the pilot robot direction of motion that can predict and pilot robot are to the photocentre C that follows robot forward direction video camera _OptAngle ψ between the direction _Ij ^kAs follows:

${\mathrm{\&psi;}}_{\mathrm{ij}}^k=\mathrm{\&pi;}-{\mathrm{\&alpha;}}^{k+1}+{\mathrm{\&beta;}}^k$

Consider the inexactness of sensing data, ψ _Ij ^kBe limited in one and expected angle ψ _Ij ^dIn the relevant interval, [ψ for example _Ij ^d-ψ _Ij ^d-, ψ _Ij ^d+ ψ _Ij ^D-u].

4, motion decision-making:

By ψ _Ij ^kAnd L _Ij ^k(be d ^k), in conjunction with the ψ of expectation _Ij ^dAnd L _Ij ^d, draw the desirable formation point P that follows robot _Fd ^k, as shown in Figure 4.d _P ^kBe O _f ^kAnd P _Fd ^kBetween distance, γ ₁Be O _f ^kL ^kWith O _f ^kP _Fd ^kBetween angle, then:

$d_P^k=\sqrt{{\left(L_{\mathrm{ij}}^k\right)}^2+{\left(L_{\mathrm{ijd}}\right)}^2-2L_{\mathrm{ij}}^k{L}_{\mathrm{ijd}}\cos ({\mathrm{\&psi;}}_{\mathrm{ijd}}-{\mathrm{\&psi;}}_{\mathrm{ij}}^k)}$

${\mathrm{\&gamma;}}_1=\arccos \frac{{\left(L_{\mathrm{ij}}^k\right)}^2+{\left(d_P^k\right)}^2-{\left(L_{\mathrm{ijd}}\right)}^2}{2L_{\mathrm{ij}}^k{d}_P^k}$

Note

For following the relative distance between robot and the pilot robot and the deviation of desired distance, For pilot robot direction of motion and pilot robot to the angle between the photocentre direction of following robot forward direction video camera and the deviation of expected angle.Work as e _Fld≤ d _FlAnd e _{Fl ψ}≤ ψ _Fl, follow robot and think that formation realizes stop motion; Work as e _Fld≤ d _FlAnd e _{Fl ψ}＞ψ _Fl, follow robot and fall back up to e along its opposite direction to the pilot robot direction _FldSurpass d _FlTill; Work as e _Fld＞d _FlAnd e _{Fl ψ}≤ ψ _Fl, follow robot with the line of pilot robot on move; Work as e _Fld＞d _FlAnd e _{Fl ψ}＞ψ _Fl, its speed v _fAnd rotational angle theta _fFollowing calculating:

${\mathrm{\&theta;}}_f=\left\{\begin{array}{cc}{\mathrm{\&theta;}}_{\max },& {\mathrm{\&theta;}}_f^{\&prime;}\&GreaterEqual;{\mathrm{\&theta;}}_{\max }\\ {\mathrm{\&theta;}}_f^{\&prime;},& -{\mathrm{\&theta;}}_{\max }<{\mathrm{\&theta;}}_f^{\&prime;}<{\mathrm{\&theta;}}_{\max }\\ -{\mathrm{\&theta;}}_{\max },& {\mathrm{\&theta;}}_f^{\&prime;}\&le;-{\mathrm{\&theta;}}_{\max }\end{array}\right.$

In the formula,

Δ T is the sampling period, v _MaxAnd θ _MaxBe respectively maximal rate and corner.

When follow robot unpredictable, when estimating pilot robot, see pilot robot if follow the non-forward direction video camera of robot, it should rotate and make the forward direction video camera find pilot robot as early as possible, if do not find pilot robot, its meeting rotary search, if following robot forward direction video camera sees pilot robot, it can move to pilot robot under the restriction of desired distance.

Embodiment

Method for controlling multi-robot coordinated formation of the present invention is applied in the formation control of two robots, and robot adopts the AIM intelligent robot of Institute of Automation Research of CAS's development.Each robot all is furnished with ccd video camera and code-disc, about 60 ° of single camera field range.Follow robot in the experiment and adopt the forward direction video camera C consistent with direction of motion _fTo pilot robot discern, location estimation; The effect of other video camera is to identify and makes C after the pilot robot as early as possible _fFind pilot robot.v _max＝0.2m/s，θ _max＝20°，N _cs＝5，N _pmax＝5，

l＝0.27m，d _c＝5cm，d _fl＝0.2m，ψ _fl＝20°。Follow the initial distance 1.54m between robot and the pilot robot, 189 ° of initial angles, formation requires that desired distance is 1.2m between two robots, and expected angle is 210 °.Adopt method provided by the present invention, can satisfy the demands, Fig. 5 has provided pilotage people and follower's's (pilot robot and abbreviation of following robot) movement locus, S _lAnd S _fThe starting point of representing pilotage people and follower respectively, G _lAnd G _fThe terminating point of representing pilotage people and follower respectively.

Claims (7)

1. a method for controlling multi-robot coordinated formation is used for multi-robot system; It is characterized in that, comprise step:

A) follow robot according to the colour code tube on the pilot robot,, obtain in conjunction with vision and code-disc information

Current sampling instant k and N before _PmaxInferior sampling instant is with respect to the distance and the observed azimuth d of pilot robot ^K-s, (s=0 ..., N _Pmax);

B) follow robot by code-disc information, by the local coordinate system conversion, recursion obtains sampling instant k and k-N _PmaxThe time pilot robot following the robot local coordinate system

Under position L ^k(x _L ^k, y _L ^k),

And the photocentre C that follows robot forward direction video camera during current sampling instant k _Opt

Position O under the coordinate system _f ^k(x _Of ^k, y _Of ^k);

C) position-based L ^k,

And O _f ^k, follow robot and calculate the pilot robot direction of motion of prediction and pilot robot to C _OptAngle ψ between the direction _Ij ^k

D) in conjunction with angle ψ _Ij ^kAnd expected angle ψ _Ijd, follow the distance L between robot and the pilot robot _Ij ^kAnd desired distance L _Ijd, comprehensive visual information, the decision-making of moving obtains speed v _fAnd rotational angle theta _f, realize following the control of robot speed's size and Orientation.

2. method for controlling multi-robot coordinated formation as claimed in claim 1, it is characterized in that the colour code tube on the pilot robot is the cylindrical tube of hollow, is combined up and down by at least two kinds of colors, be put in the robot, the center of its center and robot is consistent.

3. method for controlling multi-robot coordinated formation as claimed in claim 1 is characterized in that, formation control is to realize there is not communication between the robot on the basis of robot to the pilot robot perceptual inference following.

4. method for controlling multi-robot coordinated formation as claimed in claim 1 is characterized in that, described B) follow the robot local coordinate system in the step Be and sampling instant k-N _PmaxCorresponding, its initial point

Get forward direction video camera photocentre C _OptThe position at place,

Direction and sampling instant k-N _PmaxThe time robot motion's direction be consistent.

5. method for controlling multi-robot coordinated formation as claimed in claim 1 is characterized in that, described C) follow the prediction of robot in the step to pilot robot direction of motion, be according to current sampling instant k and this moment N before _PmaxThe pilot robot positional information of inferior sampling instant draws, and the pilot robot positional information of these two sampling instants all is based upon the N that follows before the sampling instant k of robot _PmaxUnder the pairing local coordinate system of inferior sampling instant.

6. method for controlling multi-robot coordinated formation as claimed in claim 1 is characterized in that, described D) follow distance L between robot and the pilot robot in the step _Ij ^kBe d ^k

7. as claim 1,4 or 5 described method for controlling multi-robot coordinated formation, it is characterized in that described N _Pmax〉=1.

Images