Soccer Robotics An Overview Computer Science Essay

Abstract-Soccer robotics is a new fad that is catching up non merely with game partisans, but chiefly with research workers who believe that developing a MiroSoT will take to solutions for bing restrictions and come up a new dimension to the universe of robotics. The functionality is self explanatory where, two squads of automatons play against each other a game of association football.

Soccer automatons come in assorted classs, but we will chiefly concentrate on MiroSoT. The whole thought of association football automatons is to develop an intelligent automaton that can Feel, DECIDE AND ACT depending on a existent clip state of affairs.

A robot uses detectors to feel and actuators to move on the determination made. The determination procedure goes through a instead extended process. These basic behavioural features are combined to strike, support, undertake etc. Each automaton of the squad communicates with the other with the aid of a host computing machine that is connected to the overhead camera on top of the field along with a RF faculty.

Get quality help now
Marrie pro writer
Verified writer

Proficient in: Camera

5 (204)

“ She followed all my directions. It was really easy to contact her and respond very fast as well. ”

+84 relevant experts are online
Hire writer

Soccer robotics is a good platform to research and analyze multi agents systems. In this paper we will concentrate on how computing machine vision and control techniques are used to observe the present scenario and execute consequently.

Keywords: Robots, Intelligent control, Computer vision, PID accountant.


Soccer robotics is played in several classs, but we will concentrate on MiroSoT ( Micro Robot Soccer Tournament ) . There are two sorts of conferences, Small conference which consists of three automatons in each squad and a Middle conference with five automatons in a squad and they are both inclusive of a end keeper [ 1 ] .

Get to Know The Price Estimate For Your Paper
Number of pages
Email Invalid email

By clicking “Check Writers’ Offers”, you agree to our terms of service and privacy policy. We’ll occasionally send you promo and account related email

"You must agree to out terms of services and privacy policy"
Write my paper

You won’t be charged yet!

Each automaton size is defined as a three-dimensional box of parametric quantities 7.5cms x 7.5cms ten 7.5cms [ 1 ] . Hence the field size is relative to the automaton sizes.

Intelligent automatons are those that can work in existent clip environment with lower limit or no human invasion [ 2 ] . Intelligent automatons have the undermentioned characteristic characteristics:

Reactive: The automatons should be able to feel the alterations in existent clip environment and react to them [ 2 ] .

Proactive: The automatons should be in a place to do determinations [ 2 ] .

Communicative: The automatons should be able to pass on with each other and work as a squad [ 2 ] .

These are of import characteristics for the edifice of an intelligent automaton.

Sense is defined as the perceptual experience of the environment by the automatons [ 2 ] . DECIDE is defined when the sensed information is processed and an action to be performed is construed [ 2 ] . ACT is non every bit simply simple. It falls under two classs of control and propulsion where maps that produce bids to the automatons are control and maps that make the hardware tally is actuation [ 2 ] .

The criterions of the game are set by FIRA ( Federation of International Robot soccer Association ) . FIRA is a non net income administration that conducts one-year association football competitions for automatons that are built by representatives of take parting state [ 1 ] . In order to keep a basic criterion and to contradict favourability of winning, FIRA sets some eligibility criteria the squad has to go through in order to vie [ 2 ] . The trials set by FIRA are chiefly to find if a automaton is able to execute basic association football actions. These actions are cardinal for a winning squad. The undermentioned describes the trials.

Kicking the Ball:

This action is absolute BASIC of playing the game. A automaton at the place ( x, y, I? ) is to kick the stationary ball at ( x ‘ , y ‘ ) through the place ( ten ” , y ” ) [ 2 ] . Basically the automaton should be able to get down running from any place towards the stationary ball and kick it in any way. This is necessary for striking a end, taking punishments etc.

Passing the ball to another automaton:

This involves two automatons. The first automaton at the place ( x, y, I? ) kicks a stationary ball at ( x ‘ , y ‘ ) to ( x ” , y ” ) which is to be collected by another automaton get downing from ( ten ” ‘ , y ” ‘ , I? ” ‘ ) [ 2 ] . This is the first benchmark that tests the working of a automaton as a squad.

Striking a traveling ball:

This is a hard status because the ball is invariably traveling and the captured image has to make the vision system faster than the ball traveling which poses a challenge. A automaton at ( x, y, I? ) moves to ( x ‘ , Y ‘ , I? ‘ ) to strike the ball through ( ten ” , y ” ) [ 2 ] . The motion of the automaton is based on the image captured by the overhead camera. The trying rate of the camera has to be at its upper limit as this will assist in make up one’s minding the speed of the ball.

Passing a traveling ball to traveling automaton:

A automaton at ( x, y, I? ) strikes the ball at ( x ‘ , Y ‘ , I? ‘ ) to go through through ( ten ” , y ” ) where it is struck by another automaton gazing from ( ten ” ‘ , y ” ‘ , I? ” ‘ ) to go through through ( ten ” ” , y ” ” ) [ 2 ] .

Trickling through an obstruction:

This is the most hard trial to go through as the automaton demand to acknowledge obstructions and dribble the ball past them. For this trial there is test automaton and a ball. The trial automaton is placed one robot infinite distance from the first obstruction. The 2nd obstruction is in line with the trial automaton and first obstruction. The automaton needs to observe these two obstructions and base on balls through them in a S-shaped way in order to go through the trial [ 2 ] .


As seen in the Fig. 1 there is an overhead camera placed above the field such that it covers the full drama country. Each squad has camera SENSING the motion of each automaton and the ball [ 2 ] . The determination doing faculty based on the information sensed will Decide the undermentioned class of program [ 2 ] . Now the host computing machine, the intelligent faculty will direct bids to the automatons to Act on the strategic program to strike a end [ 2 ] .

One full processing rhythm consists of these three elements ( feeling, determination devising and action ) . The trying rate of the camera captures need to be so high in order non to lose any frame. This besides requires the processing rhythm clip to be least in order to catch every frame. Hence this is what makes association football robotics disputing.


The automatons and the ball have predefined colorss [ 1 ] . This basic thought is used to recognize the entity on the field.

The chief constituent responsible for computing machine vision is an equivalent of the human oculus, overhead camera. The 3D scene is captured into a 2D image which is fed into the system. It goes through following procedure. The camera has Charge Coupled Device ( CCD ) which is an image detector [ 3 ] . CCD converts the image into an electrical parallel signal [ 3 ] . Frame grabber samples these signals utilizing ADC [ 2 ] . These sampled signals have a taking information which in common footings is called a pel. There are assorted color theoretical accounts that are used before change overing them into binary image for image cleavage. This paper will see the YUV. “ They code the information about chrominance in two dimensions ( U and V ) and merely one dimension includes the information about strength ( Y ) ” [ 4 ] .

Fig.1. Soccer robotics apparatus ( MiroSoT )

Object-background separation

“ Thresholding is a method to change over a grey scale image into a binary image so that objects of involvement are separated from the background ” [ 2 ] .

The demand to concentrate on the automaton or the ball is done by polishing the image to merely project the entity in inquiry. In a binary image the objects are considered a binary 1 and background a 0 [ 5 ] . The colored image is converted to its binary image. The object and the background are required to be at high contrast. This is achieved by the layout of the game where the automatons are color coded, whereas the field is green or black [ 1 ] . This is one of the layouts of several bing 1s, but all follow similar construction of changing contrasts.

The threshold choice is more frequently test and mistake, based on the cognition of the size of automatons and the ball.

Position and orientation

It is easier to turn up the place of the object as compared to finding its position. The object pels are highlighted on a binary image. The object place is calculated utilizing the undermentioned expression where N is the figure of rows and m is figure of columns of a grid n x m binary image [ 2 ] . A is the country of the object.

( 1 )

( 2 )

( 3 )

Equation 2, 3 give the Centre of the object [ 2 ] . This is a really of import technique after executing image labelling. These points are converted to existent physical place by utilizing another set of similar equations [ 2 ] . We now know where mark object is located but we do n’t cognize where it is orientated as the system uses an overhead camera.

Each automaton has standard two colored squares on top of it [ 6 ] . One square is used to stand for the coloring material of the squad and the other as personal single designation [ 6 ] . This is represented in Fig 2. These are criterions set by FIRA. Both the squads evidently choose different colorss. These two squares are considered in executing mathematical computation to find the position of the automaton. An axis is drawn between the spots as shown in Fig 2.

Fig.2. Top position of the automaton and orientation

The angle in which it is traveling frontward is I? and the orientation angle is [ 2 ] as shown in Fig.2. They are both deliberate as ( 4 ) , ( 5 ) :

( 4 )

Where and

( 5 )

Connectivity and size filtering

Binary image nowadayss natural informations. The needed information is extracted from this image. This is done by utilizing a construct of labelling and so filtrating it. Pixels that are following to each other are represented as either 4 neighbors or 8 neighbors. Pixels that portion one side with another are 4 neighbors and 1s that portion at least one corner are 8 neighbors [ 6 ] . This is the connectivity of pels. Neighbouring pels are given one label.

As shown in the Fig.3 connected pels are given single individuality. There are certain sorts of algorithms that run in order to observe each pel and group them. The most efficient one is consecutive algorithm [ 2 ] . First the image is scanned. The pel is given a label X [ 4 ] . The image is scanned once more. If the following pel has a neighbour so same label is given [ 4 ] . If it has more than one neighbor with different labels so the pel is given the label of the neighbors above it [ 4 ] . Finally if it has no neighbors it is given a new label, X+1 [ 4 ] . This rhythm is continued till all pels are numbered.

Fig.3. Size filtrating

Fig.4. Size filtrating

In world an image has batch of noise that is contributed by bad lighting or the overhead camera. They appear like merely another labeled object, but merely they are more ragged or conspicuous compared to objects in inquiry. They are besides lot smaller in country as compared to aim objects. Hence filtering is done based on country of the constituent [ 6 ] . Fig.4 shows an illustration of size filtering with the threshold invariable as 3 [ 6 ] . This discards all the pels which are less than or equal to 3 in size, retaining merely the needed image.

Now we have a perfect binary image of entities under survey. The equation 2, 3 are used to happen the image carbon monoxide ordinates. The physical carbon monoxide ordinates are found based on the length and comprehensiveness of the association football field as discussed earlier. At this phase we know the location and position of the automaton.


The following measure after turn uping the automaton is, driving the automaton to that location to execute the needed action, which could be kick, block, defend etc. The function of each automaton is defined based on the country it is present in. The field is defined in parts. Depending on the part the automaton is positioned the function is assigned. The automaton in guardian kingdom acts as a guardian, one near the end station Acts of the Apostless as goalkeeper and so on and so forth. These functions are non fixed as the automatons are invariably traveling. In order to avoid struggles of function it plays, whenever a automaton is unable to execute a given undertaking, the parts are re constructed and functions are assigned based on new design [ 2 ] . The action follows this determination devising.

The most common and efficient control method used is PID control system. This control system is used for guaranting successful completion of association football actions. It is used to rectify mistakes that could take to losing of the game.

PID accountant

The automatons have assigned marks to execute some action. Automatons that work on motor controlled wheels tend to exhibit batch of mistakes which leads to losing the mark [ 7 ] . In order to avoid this, a feedback control system is used so the mistakes can be corrected.

“ PID is an algorithm used in a control feedback cringle to modulate a procedure such as the gesture of a motor. Using PID control will do the automaton design more stable, robust, and has the potency to better response features [ 8 ] ” .

The PID accountant is used at two phases. First one is used for driving the automaton to a mark place. Here the mention input is the mark place. This represents the intelligent control of the system. Second it is used to cipher the left and right wheel speeds of the automaton [ 11 ] .

Fig.5. PID Controller block diagram

P – This is called the relative [ 9 ] . This is the chief constituent in cut downing mistakes and accomplishing the needed end product [ 9 ] . In our instance this is defined as velocity of the automaton.

I – This is the built-in. This represents truth of the system.

D – This is the derivative. It gives the action that needs to be taken based on the rate at which mistakes are measuring [ 10 ] .

Let us see the system end product of Fig.5 as O ( T ) , the mensural mistake as vitamin E ( T ) and mention as Y ( T ) .

The mistake signal is defined as ( 6 ) .

( 6 )

This mistake is multiplied with each of the additions of relative, built-in and derivative which are K, Kand Krespectively. Equation ( 7 ) is derived from the Fig.5, which determines the system input s ( T ) [ 10 ] .

( 7 )

The existent required end product is brought highly close to the mention or set value of the system. For illustration if the automaton is expected to travel at the velocity of 5m/s, but alternatively is traveling at 3m/s [ 7 ] . The mistake is 2m/s [ 7 ] . This mistake is brought really near to zero, doing the automaton move in right way or right velocity.

The end product received is linear and needs to be converted to digital in order to run on the host system. The digital value is obtained by distinguishing the equation 7 with regard to clip.


Fig.6. Plot of amplitude versus clip

The subsiding clip is the clip required for the mistake to make a value near to zero [ 7 ] . In a system it ‘s advantageous to hold every bit low a settling clip as possible, as the longer it takes slower will be the motion of the automaton. FS in Fig.6 is the fixed mention value that is required to accomplish. Any value traveling over that is overshoot. In a practical system there ever remains an mistake set [ 7 ] . It can ne’er be one individual value. As we can see the concluding end product is highly close to the needed value.

This is the basic construct used to travel towards the mark and execute the distinct action. The automaton now is able to observe the ball utilizing the vision system and move towards it and execute the needed action based on PID control.

Further Improvements

The automaton can be made faster and sharper by utilizing more advanced intelligent controlling of the information. The undermentioned is brief debut to how the intelligent control can be upgraded.

Fuzzy logic

Using fuzzed logic in multi agents systems has reaped better consequences. The PID accountant determines the location, orientation and speed of the automaton. This information is fed into a fuzzed accountant, which so based on the set of fuzzy regulations decides what action is to be taken [ 12 ] . As discussed earlier, each automaton in a squad is assigned its function in the game and based on the function appropriate action is taken. Based on the function its playing, fuzzy regulations are applied to make up one’s mind. Rules are framed sing all existent clip environmental parametric quantities [ 12 ] .


In this paper we have discussed the construct of computing machine vision and control technique of a multi agents system. The other intelligent control techniques being tested are systems like Petri cyberspaces, nervous webs, Q larning etc.

Soccer robotics challenges so many facets of robotics. This field does n’t yet boast of the best possible solution. One of the chief grounds is owing to its disablement of non cut downing the processing rhythm velocity.

This multi agent systems thought can be extended in using it to industrial automatons, functioning automatons etc where automatons work in an environment without any human supervising. The automatons can be used to work in mines, eating houses, as nurses, as constructional workers and in several more countries. But this will take to economic ruin with cut down in labor. Whenever this is impending, allow us trust we will be in a place to strike a balance.

Cite this page

Soccer Robotics An Overview Computer Science Essay. (2020, Jun 02). Retrieved from

Soccer Robotics An Overview Computer Science Essay

👋 Hi! I’m your smart assistant Amy!

Don’t know where to start? Type your requirements and I’ll connect you to an academic expert within 3 minutes.

get help with your assignment