CN110673826A - Random search method and system for swarm robots in area search - Google Patents
Random search method and system for swarm robots in area search Download PDFInfo
- Publication number
- CN110673826A CN110673826A CN201911023609.XA CN201911023609A CN110673826A CN 110673826 A CN110673826 A CN 110673826A CN 201911023609 A CN201911023609 A CN 201911023609A CN 110673826 A CN110673826 A CN 110673826A
- Authority
- CN
- China
- Prior art keywords
- robot
- random
- search
- step length
- distribution
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F7/00—Methods or arrangements for processing data by operating upon the order or content of the data handled
- G06F7/58—Random or pseudo-random number generators
- G06F7/588—Random number generators, i.e. based on natural stochastic processes
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Computational Mathematics (AREA)
- Mathematical Analysis (AREA)
- Mathematical Optimization (AREA)
- Pure & Applied Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Manipulator (AREA)
Abstract
The present disclosure provides a random search method and system for swarm robots in area search, comprising: establishing a robot random walking model, and controlling the random motion of the robot by using the model; constructing a random number generator to be able to generate step sizes that satisfy a given distribution and ensure that the generated step sizes are within a specified range; when the robot executes the regional search task by using a random walking method, selecting a proper step length according to the characteristics of the search task to realize effective control on the movement distance of the robot; and measuring the searching efficiency of the robot by using the mean square displacement as a performance index.
Description
Technical Field
The disclosure relates to the technical field of intelligent robots, in particular to a random search method and a system for swarm robots in area search.
Background
The swarm robot system is a system established by simulating the behaviors of social insects or other social biological groups, and consists of a plurality of undifferentiated autonomous robots under the completely distributed control. The swarm robot is an emerging discipline, belongs to the category of multi-robot systems, and mainly researches a mechanism of social interaction between robots and between the robots and the surrounding environment, and how to emerge complex swarm behaviors and swarm intelligence in the social interaction process.
Area search is a basic task in robotic research, and the search of the environment by the robot can be used for subsequent tasks of mapping, positioning and the like. The core problem of area search research is how to traverse the entire unknown area quickly and efficiently. When the area of the region is large, a single robot is difficult to effectively search the whole region. Because the swarm robots have better robustness, flexibility and expandability, the swarm robots are widely applied to the area search of unknown environments. The research of using swarm robots to execute the regional search task has important significance in both theoretical research and practical application. In practical application, swarm robot search can be used for tasks such as survivor rescue, mine disaster search and rescue, mine clearance and explosion elimination after disasters such as space exploration, earthquakes and the like. Theoretically, the research on swarm robot area search can not only realize the analysis and prediction of swarm robot search behaviors, but also improve the search efficiency of the robot. Furthermore, understanding of the swarm robot self-organization emergence mechanism can be deepened, an emergence model of the swarm robot self-organization behaviors is established, and key factors influencing the evolution law of the swarm robot self-organization behaviors are researched by the emergence model so as to design and control the swarm robot self-organization behaviors.
In the field of robots, the existing search method mainly depends on a sensor system (such as a odometer, an ultrasonic radar and the like) with higher precision and a complex mapping and search algorithm, but since the swarm robot only has limited individual capacity and cannot complete complex mapping and positioning, a simple random walking method is usually adopted when the swarm robot executes a search task. The random walk theory arises from the random motion of particles suspended in a liquid, i.e., Brownian motion. The randomness in random walking is mainly embodied in two aspects, namely the randomness of the motion direction of random walking and the randomness of the step size of walking along the direction. Random walking can be divided into two categories according to whether there is correlation between the directions of motion: the walking is carried out randomly, the continuous moving directions are independent, and the moving directions are completely random; in relation to random walks, there is a correlation between successive directions of motion, in which case the direction of motion of the robot may be correlated to a previous direction or the direction of motion of the robot points in a given direction. The random walk can be divided into a fixed step length and a randomly changed step length according to the difference of the step length. The random walking method commonly used in swarm robots is Brownian motion, levy flight, etc. Brownian motion and levey flight can produce randomly varying step sizes to control the distance of movement of the robot.
The inventor finds in research that random walking is widely applied to swarm robots at present, but the effectiveness of the random walking method in the task of area search is still to be researched. On the one hand, there is a lack of proper and effective theory for describing random walk models. On the other hand, there is no effective performance index for measuring the search efficiency of random walking in area search. In addition, the existing random walking method (such as Brownian motion and levy flight) can not effectively control the moving distance of the robot due to the generated large and small step length, so that the research on the swarm robot searching efficiency is more difficult.
Disclosure of Invention
The invention aims to provide a random search method of swarm robots in area search, which realizes effective control of the movement distance of the robots and effectively measures the search efficiency.
The embodiment of the specification provides a random search method of swarm robots in area search, which is realized by the following technical scheme:
the method comprises the following steps:
establishing a robot random walking model, and controlling the random motion of the robot by using the model;
constructing a random number generator to be able to generate step sizes that satisfy a given distribution and ensure that the generated step sizes are within a specified range;
when the robot executes the regional search task by using a random walking method, selecting a proper step length according to the characteristics of the search task to realize effective control on the movement distance of the robot;
and measuring the searching efficiency of the robot by using the mean square displacement as a performance index.
In the further technical scheme, random walking under the N-dimensional condition is deduced based on the condition that the robot moves in the one-dimensional lattice, and a probability density function of the position where the robot is located is obtained; for real random walking in a two-dimensional state, the moving direction of the robot is described by using von mises distribution.
According to the further technical scheme, when the mean square displacement of the random walking method using the von mises distribution is calculated, the mean square displacement of the robot is described by utilizing the sine mean value and the cosine mean value of the motion direction generated by the distribution and the probability density function of the position where the robot is located.
Further technical solution, a random number generator is constructed to generate step sizes satisfying a given distribution and ensure that the size of the generated step size is within a specified range:
according to the range of the step length specified by the actual region search task, wherein L is the minimum value of the step length, L is the maximum value of the step length, and the random number generator generates an integer value between L and L as the step length;
supposing that the step length s (L is not more than s and not more than L) of the robot meets the probability distribution P(s), wherein P represents the probability of generating the step length s, the probability P(s) of other integer step lengths is calculated in sequence, s belongs to [ L, L ], the step length with the minimum probability is found through comparison, the number of the step lengths generated by the random number generator is set to be 1, the number of the other step lengths is calculated according to the probability comparison, the number of the integer step lengths in the [ L, L ] range can be calculated finally, the elements form a step length set, and the step length is selected randomly from the set when the robot executes the region search task by using a random walking method.
The embodiment of the specification provides a random search system of swarm robots in area search, which is realized by the following technical scheme:
the method comprises the following steps:
the model building module is used for building a robot random walking model and controlling the random motion of the robot by using the model;
the random number generator constructing module is used for constructing the random number generator to generate the step size meeting the given distribution and ensuring that the generated step size is in a specified range;
the robot motion control module selects a proper step length according to the characteristics of the search task when the robot executes the area search task by using a random walking method, so as to realize effective control on the motion distance of the robot;
and measuring the searching efficiency of the robot by using the mean square displacement as a performance index.
In the further technical scheme, random walking under the N-dimensional condition is deduced based on the condition that the robot moves in the one-dimensional lattice, and a probability density function of the position where the robot is located is obtained; for real random walking in a two-dimensional state, the moving direction of the robot is described by using von mises distribution.
According to the further technical scheme, when the mean square displacement of the random walking method using the von mises distribution is calculated, the mean square displacement of the robot is described by utilizing the sine mean value and the cosine mean value of the motion direction generated by the distribution and the probability density function of the position where the robot is located.
Further technical solution, a random number generator is constructed to generate step sizes satisfying a given distribution and ensure that the size of the generated step size is within a specified range:
according to the range of the step length specified by the actual region search task, wherein L is the minimum value of the step length, L is the maximum value of the step length, and the random number generator generates an integer value between L and L as the step length;
supposing that the step length s (L is not more than s and not more than L) of the robot meets the probability distribution P(s), wherein P represents the probability of generating the step length s, the probability P(s) of other integer step lengths is calculated in sequence, s belongs to [ L, L ], the step length with the minimum probability is found through comparison, the number of the step lengths generated by the random number generator is set to be 1, the number of the other step lengths is calculated according to the probability comparison, the number of the integer step lengths in the [ L, L ] range can be calculated finally, the elements form a step length set, and the step length is selected randomly from the set when the robot executes the region search task by using a random walking method.
The invention also discloses computer equipment which comprises a memory, a processor and a computer program stored on the memory and capable of running on the processor, and is characterized in that the processor realizes the steps of the random search method of the swarm robots in the area search when executing the program.
The invention also discloses a computer readable storage medium, which stores a computer program, characterized in that the program is executed by a processor to realize the steps of the random search method of the swarm robots in the area search.
Compared with the prior art, the beneficial effect of this disclosure is:
the invention firstly provides the mathematical description of the random walk model, and provides a performance index for measuring the search efficiency according to the given random walk mathematical model. In order to obtain the effective random step length, the invention also constructs a random number generator based on the random walking method, and the random number generator can generate the random step length in a certain range based on the given random walking method, thereby realizing the effective control of the movement distance of the robot. In addition, the performance index provided by the invention is combined with the random number generator, so that the search efficiency can be more effectively measured.
Drawings
The accompanying drawings, which are included to provide a further understanding of the disclosure, illustrate embodiments of the disclosure and together with the description serve to explain the disclosure and are not to limit the disclosure.
FIG. 1 is a graph of walking trajectories of 10 robots when the step size of the robots of one or more embodiments of the present disclosure follows a normal distribution;
FIG. 2 illustrates walking trajectories of 10 robots when the robot steps follow a power law distribution according to one or more embodiments of the present disclosure;
FIG. 3 is a probability density function of a von Missels distribution for one or more implementation examples of the disclosure;
FIG. 4 shows a random walking trajectory of 10 robots according to one or more embodiments of the present disclosure, wherein the robot motion directions follow a von Missels distribution θ0Pi/6, k 0, step size obeyedNormal distribution μ ═ 5 and δ ═ 1;
FIG. 5 shows a random walking trajectory of 10 robots according to one or more embodiments of the present disclosure, wherein the robot motion directions follow a von Missels distribution θ0Pi/6, k 1, the step size follows a normal distribution μ 5, δ 1;
FIG. 6 shows a random walking trajectory of 10 robots according to one or more embodiments of the present disclosure, wherein the robot motion directions follow a von Missels distribution θ0Pi/6, k 2, step size follows normal distribution μ 5, δ 1;
FIG. 7 shows a random walking trajectory of 10 robots according to one or more embodiments of the present disclosure, wherein the robot motion directions follow a von Missels distribution θ0Pi/6, k 4, the step size follows a normal distribution μ 5, δ 1.
Detailed Description
It should be noted that the following detailed description is exemplary and is intended to provide further explanation of the disclosure. Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this disclosure belongs.
It is noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of example embodiments according to the present disclosure. As used herein, the singular forms "a", "an" and "the" are intended to include the plural forms as well, and it should be understood that when the terms "comprises" and/or "comprising" are used in this specification, they specify the presence of stated features, steps, operations, devices, components, and/or combinations thereof, unless the context clearly indicates otherwise.
Example of implementation 1
The embodiment discloses a random search method of swarm robots in area search, which specifically comprises the following steps:
firstly, a random number generator is used for generating step length in a given range, then mean square displacement is used for measuring the search efficiency of several random walking methods, and a proper random walking method is selected. Generating a random motion direction complying with von Misses distribution, generating a random step length by using a random number generator, and searching; then generating random motion direction and random step length, and searching.
In the implementation example, firstly, a mathematical description of the random walk model is given, and a performance index, namely, mean square displacement, which can be used for measuring the search efficiency is provided according to the given random walk mathematical model, wherein the value of the mean square displacement is related to the step length generated by the random walk method and the probability corresponding to the step length. In order to obtain effective random step length, the invention also constructs a random number generator based on a random walking method, and the random number generator can generate random step length within a certain range based on a given random walking method, thereby not only obtaining more effective step length, but also realizing effective control of the movement distance of the robot. In addition, the performance index provided by the invention is combined with the random number generator, so that the search efficiency can be more effectively measured.
The performance index is used for measuring and comparing the search efficiency of the robot using various random walking methods, namely comparing the effectiveness of the random walking methods, and then selecting a more effective random walking method according to task requirements.
The invention firstly provides mathematical description of a robot random walking model and deduces a probability density function of the position of the robot.
For easy calculation and derivation, the invention firstly gives the situation that the robot moves in a one-dimensional lattice. Assuming that the robot is located at the origin (x is 0) at the initial time (t is 0), the probabilities of the robot moving straight and moving backward are r and l, respectively, and the movement distance of the robot in a short time τ is δ. If the robot is located at x at time t + τ, there are two possibilities: (1) the robot moves to the right at x- δ and the next time, (2) the robot moves to the left at x + δ and the next time. Can obtain the product
p(x,t+τ)＝p(x-δ,t)r+p(x+δ,t)l
When the distance delta and the time interval tau are small, the formula can be expanded into a Taylor series at (x, t), and therefore, the formula can be obtained
Wherein, epsilon is r-l, O (tau)2),O(δ3) Representing higher order terms.
Let δ, τ → 0, then O (τ)2),O(δ3) Approaching 0, the above formula can be expressed as
Wherein
The formula is a translation-diffusion square scale, whereinIs a translation term, representing the movement of the robot in a desired direction;
a diffuse term represents a random movement of the robot towards the peripheral area. When the equation has an initial solution x-0 and t-0, there is an analytical solution for the equation
It can be seen that the position of the robot at time t follows a normal distribution.
Generally, the farther the distance between the position of the robot and the initial position is, the larger the area searched by the robot is, which indicates that the robot has higher search efficiency. The mean square displacement is the square of the distance between the position of the robot and the initial position at the moment t, so the method measures the search efficiency of the robot by using the mean square displacement as a performance index. The mean square displacement can be expressed as
In addition, the average position of the robot can also be used to analyze the search efficiency of the robot, and is noted as
From the above formula, can be derived
For the related random walk, i.e. the probability of straight walk and backward walk is not equal (r ≠ l), when t is larger
Mean square displacement and t of robot random walking2Is in direct proportion. For irrelevant random walk, i.e. the probability of straight walk and backward walk is equal (r ═ l), the mean square displacement and the mean position of the robot random walk satisfy
The mean square displacement of the robot walking randomly is proportional to t. That is, when the robot performs the area search using the correlation random walk method, the search efficiency and t are2Is in direct proportion; when the robot performs the area search using the unrelated random walk method, the search efficiency is proportional to the time t.
Random walk under the N-dimensional condition can be deduced according to the random walk of the robot in the one-dimensional lattice, and a translation-diffusion equation can be obtained
Where u is the N-dimensional average translational velocity,
is a gradient operator, which is a linear operator,is the laplacian operator. When the equation has an initial solution x-0 and t-0, there is an analytical solution for the equation
From the calculation formula of the average position and the mean square displacement of the robot, E (X) can be obtainedt) Ut. Let Rt＝|xtL, mean square displacement of
Can obtain the product
For related random walk, mean square displacement and t of robot random walk2Is in direct proportion; for unrelated random walks, the mean square displacement of the robot's random walk is proportional to t.
The method firstly deduces and obtains the probability density function of the position where the robot is located under the ideal condition of 1-dimensional lattice and N-dimensional lattice (the robot can only select one dimension each time and can select one direction to move from the front direction and the back direction of the dimension). Whereas the actual motion of the robot is typically a 2-dimensional space, the direction of motion of the robot is completely random, i.e. the robot can move in any direction, subject to a distribution that is a circular distribution, which the present invention uses von mises (von mises) distributions to describe,
wherein I0Bessel (Bessel) function representing the first 0 th order, which is denoted as
θ0∈[-π,π]Is the mean of the generated motion directions, which can be taken as the desired motion direction; kappa can measure the direction of production theta withMean value theta0The larger the κ value is, the more the distribution of θ is concentrated on θ0(ii) a Conversely, the smaller the k value, the more dispersed the distribution of θ; when k is 0, there is no correlation between the generated directions θ, and the robot moves in an unrelated random manner. The probability density function of the von mises distribution can be seen in fig. 3. Von mises distribution through theta0The overall movement direction of the robot is controlled, and the concentration degree of the movement direction of the robot is controlled through kappa.
In order to calculate the mean square displacement of the random walk method using von mises distribution, it is necessary to study the law of the direction of motion θ generated by the distribution. In a circular distribution, the sine mean s and cosine mean c of the direction of motion θ are commonly used indicators, where
Using a sine mean value s and a cosine mean value c of theta and a translation-diffusion equation, after n steps, the mean square displacement of the robot is
Wherein γ ═ ((1-c)2-s2)cos((n+1)θ0)-2s(1-c)sin((n+1)θ0)。
In nature, a living being is usually searched in an environment along a desired direction θ0Left-right symmetrical, i.e. s is 0, in which case the mean square displacement of the above equation can be reduced to
Wherein
When the step length L is constantWhen, E (L)2)＝(E(L))2If b is 0; when the step size L is varied, E (L)2)>(E(L))2Then b is>0; the formula shows that when the robot executes a search task by using a random walking method, the search efficiency of the variable step length is greater than the fixed step length.
When the robot has a desired motion direction and the sine mean s of the motion direction θ is 0, the mean square displacement of the above equation can be further simplified to
When n is large, it is preferable that,
i.e. the search efficiency and t of the robot2It is related.
For unrelated random walks, f (θ) is uniformly distributed and s ═ c ═ 0, where the mean square shift can be simplified to
From the definition of mean square displacement, it can be obtained
From this equation, the search efficiency of the robot is determined by the mean and variance of the step size, and the mean e (l) of the step size has a larger influence on the search efficiency than the variance d (l).
When performing area search, the commonly used random walking methods are Brownian motion and levy flight, i.e. the step length of the robot follows normal distribution and power law distribution respectively. When the robot performs area search by using a Brownian motion method, the normal distribution can generate a plurality of smaller step lengths and a plurality of larger step lengths; when the robot performs a zone search using the levy flight method, the power law distribution will generate some smaller steps at a higher frequency and occasionally some larger steps. When the robot executes the region searching task, the effective motion of the robot cannot be realized by a smaller step length, and the larger step length has no practical significance to a limited region. On the other hand, when the mean square displacement is used to measure the search efficiency of the robot, the mean and variance of the step length of the robot are needed, and the larger step length generated by the random walking method such as normal distribution and power law distribution is not necessarily the step length of the actual walking of the robot, so in order to measure the search efficiency better by the mean square displacement, the step length generated by the conventional random walking method must be properly adjusted. The present invention constructs a random number generator that is capable of generating step sizes that satisfy a given distribution and ensuring that the generated step sizes are within a specified range.
The range of step sizes is specified according to the actual region search task, L is the minimum value of the step sizes, L is the maximum value of the step sizes, and the random number generator generates an integer value between L and L as the step size. Suppose that the robot step length s (L ≦ s ≦ L) satisfies the probability distribution P(s), where P represents the probability of generating step length s. And sequentially calculating the probabilities P(s) of other integer step lengths, wherein s belongs to [ L, L ], comparing and finding the step length with the minimum probability, setting the number of the step lengths generated by the random number generator to be 1, and calculating the number of other step lengths according to the probability comparison. And finally, calculating the number of step sizes of each integer in the range of [ L, L ], forming a set of step sizes by the elements, and randomly selecting the step sizes from the set when the robot executes the region search task by using a random walking method.
In order to more clearly illustrate the way of generating the random step size by the random number generator, taking the levy random number generator as an example, a specific process of the generation way is given:
the probability of step length satisfaction is P(s) -s-μWhere μ is 2, the minimum value L is 1, and the maximum value L is 1000. 1 for P (1), 0.25 for P (2), 1.0 × 10 for P (1000)-6Step length 1000 has the smallest probability, the step length number with the setting value of 1000 is 1, the step length number with the value of 2 is 250000, the step length number with the value of 1 is 1000000, the number of other integer step lengths is calculated in sequence, finally, all elements form a set of step lengths, and the step length is randomly selected from the set when the robot executes the area search task by using a random walking method.
Robot performs area search using random walk methodWhen a task is searched, a proper step length needs to be selected according to the characteristics of the search task, and when the step length of the robot is in normal distribution, more and smaller step lengths can be generated, as shown in fig. 1; when the robot's step size follows a power law distribution, some smaller step sizes may be generated and a few larger step sizes may be generated, as shown in fig. 2. The robot cannot achieve an efficient search using too small and too large step sizes, and therefore it is necessary to use a random number generator to generate step sizes within a certain range and subject to a certain distribution in order to achieve an efficient search of the robot. When the robot executes area search, the search direction needs to be changed after the robot moves for a given step length, and the angle theta obeying von mises distribution is a new movement direction. As shown in fig. 3, the value of the parameter κ in the von mises distribution is adjusted so that the robot can realize independent random walking and also can realize dependent random walking. FIG. 4 is θ0Pi/6, k 0, then 10 robots perform the area search task using a random walk method, and the robot motion appears as completely unrelated random walks. FIG. 5 is θ0Pi/6, k 1, when 10 robots perform the area search task using the random walking method, the robot motion appears as the related random walking. FIG. 6 is θ0Pi/6, k 2 von mises distribution, and at this time, 10 robots perform the area search task by using a random walking method, and robot motions are expressed as related random walks. FIG. 7 is θ0Pi/6, k 4, when 10 robots perform the area search task using the random walking method, the robot motion appears as the related random walking. And as the parameter value k becomes larger, the relevance of the random walking of the robot is enhanced, namely the trend of the robot moving along the expected direction is more and more obvious. When the robot executes a search task in an unknown environment, an irrelevant random walking method is generally used, and the search efficiency of the robot can be measured by mean square displacement. At this time, mean square displacement
From this equation, the search efficiency of the robot is determined by the mean and variance of the step lengthAnd the mean of the step sizes has a greater influence on the search efficiency.
It is to be understood that throughout the description of the present specification, reference to the term "one embodiment", "another embodiment", "other embodiments", or "first through nth embodiments", etc., is intended to mean that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the terms used above do not necessarily refer to the same embodiment or example. Furthermore, the particular features, structures, or materials described may be combined in any suitable manner in any one or more embodiments or examples.
The above description is only a preferred embodiment of the present disclosure and is not intended to limit the present disclosure, and various modifications and changes may be made to the present disclosure by those skilled in the art. Any modification, equivalent replacement, improvement and the like made within the spirit and principle of the present disclosure should be included in the protection scope of the present disclosure.
Claims (10)
1. The random search method of swarm robots in the area search is characterized by comprising the following steps:
establishing a robot random walking model, and controlling the random motion of the robot by using the model;
constructing a random number generator to be able to generate step sizes that satisfy a given distribution and ensure that the generated step sizes are within a specified range;
when the robot executes the regional search task by using a random walking method, selecting a proper step length according to the characteristics of the search task to realize effective control on the movement distance of the robot;
and measuring the searching efficiency of the robot by using the mean square displacement as a performance index.
2. The method for random search of the swarm robots in the regional search according to claim 1, wherein the random walk in the N-dimensional situation is derived based on the movement of the robot in the one-dimensional lattice, and the probability density function of the position where the robot is located is obtained; for real random walking in a two-dimensional state, the moving direction of the robot is described by using von mises distribution.
3. The method for the swarm robot to randomly search for the area according to claim 1, wherein the mean square displacement of the robot is described by means of the mean sine and cosine values of the moving direction generated by the von mises distribution and the probability density function of the robot position when calculating the mean square displacement of the random walking method using the von mises distribution.
4. The method for random search of the swarm robots of claim 1, wherein the random number generator is configured to generate the step size satisfying a given distribution and to ensure the generated step size is within a specified range:
according to the range of the step length specified by the actual region search task, wherein L is the minimum value of the step length, L is the maximum value of the step length, and the random number generator generates an integer value between L and L as the step length;
supposing that the step length s (L is not more than s and not more than L) of the robot meets the probability distribution P(s), wherein P represents the probability of generating the step length s, the probability P(s) of other integer step lengths is calculated in sequence, s belongs to [ L, L ], the step length with the minimum probability is found through comparison, the number of the step lengths generated by the random number generator is set to be 1, the number of the other step lengths is calculated according to the probability comparison, the number of the integer step lengths in the [ L, L ] range can be calculated finally, the elements form a step length set, and the step length is selected randomly from the set when the robot executes the region search task by using a random walking method.
5. The random search system of swarm robot in the area search is characterized by comprising:
the model building module is used for building a robot random walking model and controlling the random motion of the robot by using the model;
the random number generator constructing module is used for constructing the random number generator to generate the step size meeting the given distribution and ensuring that the generated step size is in a specified range;
the robot motion control module selects a proper step length according to the characteristics of the search task when the robot executes the area search task by using a random walking method, so as to realize effective control on the motion distance of the robot;
and measuring the searching efficiency of the robot by using the mean square displacement as a performance index.
6. The system of claim 5, wherein the random walk in N-dimension is derived based on the motion of the robot in one-dimensional lattice, and the Von-Misses distribution is used to describe the motion direction of the robot to describe the real random walk of the robot in two-dimensional state.
7. The system for random search in an area search by a swarm robot as claimed in claim 5, wherein the mean square displacement of the robot is described by a sine mean and a cosine mean of a moving direction generated by the Von Milsses distribution when calculating the mean square displacement of the random walking method using the Von Milsses distribution.
8. The system for random search of the swarm robots of claim 5, wherein the random number generator is configured to generate the step size satisfying a given distribution and to ensure that the generated step size is within a specified range:
according to the range of the step length specified by the actual region search task, wherein L is the minimum value of the step length, L is the maximum value of the step length, and the random number generator generates an integer value between L and L as the step length;
supposing that the step length s (L is not more than s and not more than L) of the robot meets the probability distribution P(s), wherein P represents the probability of generating the step length s, the probability P(s) of other integer step lengths is calculated in sequence, s belongs to [ L, L ], the step length with the minimum probability is found through comparison, the number of the step lengths generated by the random number generator is set to be 1, the number of the other step lengths is calculated according to the probability comparison, the number of the integer step lengths in the [ L, L ] range can be calculated finally, the elements form a step length set, and the step length is selected randomly from the set when the robot executes the region search task by using a random walking method.
9. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor when executing the program implements the steps of the method for random search in area search by swarm robots of any of claims 1 to 4.
10. A computer-readable storage medium having stored thereon a computer program, characterized in that the program, when being executed by a processor, implements the steps of the method for random search in area search by swarm robots as claimed in any one of claims 1 to 4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911023609.XA CN110673826B (en) | 2019-10-25 | 2019-10-25 | Random search method and system for swarm robots in area search |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911023609.XA CN110673826B (en) | 2019-10-25 | 2019-10-25 | Random search method and system for swarm robots in area search |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110673826A true CN110673826A (en) | 2020-01-10 |
| CN110673826B CN110673826B (en) | 2023-02-28 |
Family
ID=69084407
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911023609.XA Active CN110673826B (en) | 2019-10-25 | 2019-10-25 | Random search method and system for swarm robots in area search |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110673826B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2023193424A1 (en) * | 2022-04-07 | 2023-10-12 | 哈尔滨工业大学（深圳） | Global navigation method for mobile robot in man-machine coexistence environment following pedestrian norm |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107066705A (en) * | 2017-03-27 | 2017-08-18 | 东莞理工学院 | The Group Robots searching algorithm and verification method looked for food based on Physarum Polycephalum |
| CN108958028A (en) * | 2018-05-24 | 2018-12-07 | 大连理工大学 | Based on the group robot collaboratively searching method for improving particle swarm algorithm |
| CN109270905A (en) * | 2018-10-22 | 2019-01-25 | 山东大学 | A kind of method that group robot neural network based is looked for food using pheromones communication realization cooperation |
| CN109298725A (en) * | 2018-11-29 | 2019-02-01 | 重庆大学 | A kind of Group Robots distributed multiple target tracking method based on PHD filtering |
| CN109615057A (en) * | 2018-11-29 | 2019-04-12 | 山东大学 | Group robot look for food in the self-organizing method for allocating tasks based on dynamic response threshold value |
-
2019
- 2019-10-25 CN CN201911023609.XA patent/CN110673826B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107066705A (en) * | 2017-03-27 | 2017-08-18 | 东莞理工学院 | The Group Robots searching algorithm and verification method looked for food based on Physarum Polycephalum |
| CN108958028A (en) * | 2018-05-24 | 2018-12-07 | 大连理工大学 | Based on the group robot collaboratively searching method for improving particle swarm algorithm |
| CN109270905A (en) * | 2018-10-22 | 2019-01-25 | 山东大学 | A kind of method that group robot neural network based is looked for food using pheromones communication realization cooperation |
| CN109298725A (en) * | 2018-11-29 | 2019-02-01 | 重庆大学 | A kind of Group Robots distributed multiple target tracking method based on PHD filtering |
| CN109615057A (en) * | 2018-11-29 | 2019-04-12 | 山东大学 | Group robot look for food in the self-organizing method for allocating tasks based on dynamic response threshold value |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2023193424A1 (en) * | 2022-04-07 | 2023-10-12 | 哈尔滨工业大学（深圳） | Global navigation method for mobile robot in man-machine coexistence environment following pedestrian norm |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110673826B (en) | 2023-02-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Zafar et al. | Methodology for path planning and optimization of mobile robots: A review | |
| Mohanty et al. | Cuckoo search algorithm for the mobile robot navigation | |
| Jun et al. | Path planning for unmanned aerial vehicles in uncertain and adversarial environments | |
| Pang et al. | A swarm robotic exploration strategy based on an improved random walk method | |
| Reynolds | Bridging the gulf between correlated random walks and Lévy walks: autocorrelation as a source of Lévy walk movement patterns | |
| Kecskés et al. | PSO and GA optimization methods comparison on simulation model of a real hexapod robot | |
| Gorbenko et al. | Anticipation in Simple Robot Navigation and Learning of Effects of Robot’s Actions and Changes of the Environment | |
| Gorbenko et al. | Computational Experiments for the Problem of Selection of a Minimal Set of Visual Landmarks | |
| Zainal et al. | Glowworm swarm optimization (GSO) algorithm for optimization problems: A state-of-the-art review | |
| Arora et al. | Robotic path planning using genetic algorithm in dynamic environment | |
| Oikawa et al. | Distributed formation control for swarm robots using mobile agents | |
| CN110673826B (en) | Random search method and system for swarm robots in area search | |
| Andreevich et al. | Designing algorithms for service robots on the basis of mivar approach | |
| Gorbenko et al. | Anticipation in simple robot navigation and finding regularities | |
| Volos et al. | Chaotic autonomous mobile robot for military missions | |
| Omar et al. | 3D path planning for unmanned aerial vehicles using visibility line based method | |
| Zhao et al. | Design and Analysis of Multi-robot Grouping Aggregation Algorithm. | |
| Ranó et al. | A biologically inspired controller to solve the coverage problem in robotics | |
| Tuncer et al. | Chromosome coding methods in genetic algorithm for path planning of mobile robots | |
| Biswas et al. | Simultaneous replanning with vectorized particle swarm optimization algorithm | |
| Nazarzehi et al. | A distributed bio-inspired algorithm for search of moving targets in three dimensional spaces | |
| Petrenko et al. | Evaluation of the iterative method of task distribution in a swarm of unmanned aerial vehicles in a clustered field of targets | |
| Pang et al. | Analysis of random walk models in swarm robots for area exploration | |
| Xu et al. | Stochastic adaptive sampling for mobile sensor networks using kernel regression | |
| Chen et al. | Real time path planning via alternating minimisation through image information |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |