CN105956504B - The RFID network planing method of simulated curling motion - Google Patents
The RFID network planing method of simulated curling motion Download PDFInfo
- Publication number
- CN105956504B CN105956504B CN201610247877.XA CN201610247877A CN105956504B CN 105956504 B CN105956504 B CN 105956504B CN 201610247877 A CN201610247877 A CN 201610247877A CN 105956504 B CN105956504 B CN 105956504B
- Authority
- CN
- China
- Prior art keywords
- reader
- working area
- collision
- moving
- rfid network
- 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.)
- Active
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06K—GRAPHICAL DATA READING; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
- G06K7/00—Methods or arrangements for sensing record carriers, e.g. for reading patterns
- G06K7/10—Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation
- G06K7/10009—Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation sensing by radiation using wavelengths larger than 0.1 mm, e.g. radio-waves or microwaves
- G06K7/10019—Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation sensing by radiation using wavelengths larger than 0.1 mm, e.g. radio-waves or microwaves resolving collision on the communication channels between simultaneously or concurrently interrogated record carriers.
- G06K7/10079—Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation sensing by radiation using wavelengths larger than 0.1 mm, e.g. radio-waves or microwaves resolving collision on the communication channels between simultaneously or concurrently interrogated record carriers. the collision being resolved in the spatial domain, e.g. temporary shields for blindfolding the interrogator in specific directions
- G06K7/10089—Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation sensing by radiation using wavelengths larger than 0.1 mm, e.g. radio-waves or microwaves resolving collision on the communication channels between simultaneously or concurrently interrogated record carriers. the collision being resolved in the spatial domain, e.g. temporary shields for blindfolding the interrogator in specific directions the interrogation device using at least one directional antenna or directional interrogation field to resolve the collision
- G06K7/10099—Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation sensing by radiation using wavelengths larger than 0.1 mm, e.g. radio-waves or microwaves resolving collision on the communication channels between simultaneously or concurrently interrogated record carriers. the collision being resolved in the spatial domain, e.g. temporary shields for blindfolding the interrogator in specific directions the interrogation device using at least one directional antenna or directional interrogation field to resolve the collision the directional field being used for pinpointing the location of the record carrier, e.g. for finding or locating an RFID tag amongst a plurality of RFID tags, each RFID tag being associated with an object, e.g. for physically locating the RFID tagged object in a warehouse
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Toxicology (AREA)
- Physics & Mathematics (AREA)
- Computer Networks & Wireless Communication (AREA)
- Electromagnetism (AREA)
- General Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a kind of RFID network planing method of simulated curling motion, simulated curling motion first builds reader model, designs movement and the impact operations of reader；Build a reader；Define moving operation and impact operations of the reader in working region；Reader moves the motion search with collision always in working region, until reader movement number is more than the motor performance that default maximum mobile frequency threshold value completes the reader；New reader is built, reader number adds 1, calculates all readers in now working region and, to the coverage rate COV of label, minimum vertex-covering rate threshold value is preset up to coverage rate is more than；RFID network optimization is completed, elimination operation is carried out to reader, and export RFID network prioritization scheme.The present invention can fast and effectively obtain optimum results, and combine reader and eliminate operation, the quantity of determination reader that can be adaptive, can apply to the RFID network optimization of engineering.
Description
Technical Field
The invention belongs to the technical field of engineering optimization, particularly relates to RFID network optimization, and particularly relates to a quick and effective RFID network planning method for simulating curling motion, which is used for optimizing an RFID network.
Background
Radio Frequency Identification (RFID) technology is an important automatic identification technology. There are many applications of RFID technology, such as in supply chain problems, where RFID tags are attached to each component before production, and when a product is out of order, the source of the problem component can be quickly and accurately tracked and information about the problem component can be obtained by reading the data in the RFID tag. In inventory management, RFID tags are added to each article in a warehouse, a reader in the warehouse reads data in the RFID tags and transmits the data to a central computer, and the central computer integrates information sent by the reader, so that the inventory quantity can be rapidly and accurately counted and information of the articles in the inventory can be obtained. Due to the limited detection range of readers and the large number of tags, rfid systems typically include multiple readers. How to set the number of readers and arrange the positions of the readers is a very important technical problem. If the number of readers is too small, the coverage rate of the tags in the network is too low, and if the number of readers is too large, the resources are wasted. When different readers detect one tag at the same time, interference is caused between the readers, so that data transmission is unreliable. More and more scholars have devised various methods to optimize the RFID network, determine the number of readers required and lay out the locations of the readers. Therefore, designing a reasonable and effective RFID network planning method is still an important direction for research, so that the cost can be saved, and the stability of the RFID network can be enhanced.
In the first scheme, botero et al, "RFID network timing design based on genetic algorithms" ("IEEE International Conference on RFID-technologies and Applications, article number: 300-305 (2011)), the core of the article is to search a solution space by using genetic algorithm codes, specifically, the method is to encode the position and power of a reader by using an individual string in a population, take indexes such as a tag coverage rate, an interference rate and the like as fitness functions, each search is called a generation, and individuals with the maximum fitness functions in the population are reserved in each generation until a certain termination condition is met.
Scheme two, the core of the article is to search a solution space by using a particle swarm algorithm, specifically, the position and the power of a reader of a problem are coded by using particles in a population, the label coverage rate, the interference rate, the load balance and the economic efficiency are used as functions for evaluating the advantages and disadvantages of the particles, the population is optimized by using the optimal solution of the particles and the global optimal solution of the current population of the particles, and the number of the readers in the particles is changed by combining a redundant reader elimination strategy, the information of the particles in the population is continuously updated, and the particles with large function values are reserved until a certain termination condition is met.
In the process of implementing the invention, the inventor finds that the prior art has at least the following problems:
the searching space of the first scheme is large, the specific number of the readers needs to be defined before searching, whether the number of the readers is reasonable or not determines the result of the RFID network planning to a great extent, and in practical engineering application, the estimation of the reasonable number of the readers is difficult; in the second scheme, although the number of readers can be determined in a self-adaptive manner in the searching process, the calculation amount is large, and the method is difficult to apply to a large-scale problem. With the continuous increase of data processing amount in engineering, the problem of mass data processing becomes more and more important, and even if the technical scheme can realize excellent effect, the mass data processing can cause the realization of practical engineering.
Disclosure of Invention
In order to solve the technical problems that the genetic algorithm and the particle swarm algorithm in the prior art are large in calculation amount and cannot determine the number of readers in a self-adaptive mode, the invention provides the RFID network planning method for simulating curling movement, which can determine the number of readers in a self-adaptive mode quickly, effectively and in a self-adaptive mode and is reasonable.
The invention relates to an RFID network planning method for simulating curling motion, which is characterized by comprising the following steps of:
simulating curling motion to construct a reader model, and simulating curling motion to design the moving operation and the collision operation of the reader;
1.1, constructing a reader model, defining an RFID network working area to be optimized as a curling playground, wherein the working area is smooth and flat and has a boundary, labels to be covered in engineering are scattered in the working area, and the reader searches in the working area through movement;
1.2 designing the mobile operation of the reader, wherein the mobile operation of the reader has the following characteristics that the surface friction coefficient of an area with a label is larger, the surface friction coefficient of an area without the label is zero, the reader is under the action of friction force when moving in a working area and meets the label, and the readers have no action with each other;
1.3 designing collision operation of the reader, wherein the collision operation of the reader has the following characteristics that the reader can rebound when meeting the boundary of a working area, the readers have no collision effect, the readers and a label have no collision effect, the collision time is ignored, and the collision times are ignored;
step two, constructing a reader according to the reader model;
step three, moving the reader in the working area, wherein the current position of the reader in the working area is (x) 0 ,y 0 ) Calculating the position (x) of the moving end point of the reader in a given moving time delta t 1 ,y 1 ) If (x) 1 ,y 1 ) Within the working area, the end point position (x) is moved 1 ,y 1 ) Updating the reader's current location (x) 0 ,y 0 ) Adding 1 to the moving times of the reader, and continuously searching in the working area by the reader; if the position of the moving end point of the reader is not in the working area within the given one-time moving time delta t, the reader is defined not to do moving operation, and is not used (x) 1 ,y 1 ) Modifying the current location (x) of the reader 0 ,y 0 )；
Step four, performing collision operation on the reader in the working area, if the movement end point of the reader is not in the working area within the given one-time movement time delta t, determining that the reader performs the collision operation, wherein the reader collides at a certain position at the edge of the working area, the collision position is (x ', y'), the speed of the reader is unchanged after the collision, the direction is changed, the speed v 'of the reader which collides and the speed direction theta' of the reader are calculated and obtained according to elastic collision, the one-time movement time delta t of the next movement operation is corrected, and the reader continues to search in the working area;
step five, repeatedly executing the step three to the step four, wherein the reader carries out motion search in the working area until the movement times of the reader are larger than a preset maximum movement time threshold value, the operation on the reader is completed, and at the moment, the tags in the power radius range of the position of the reader are defined to be covered by the reader;
step six, repeating the step two to the step five, constructing a new reader, adding 1 to the number of the readers, and taking the ratio of the covered labels as a coverage rate COV until the coverage rate is greater than a preset minimum coverage rate threshold value;
and step seven, completing RFID network planning, performing reader elimination operation on all readers in the working area, completing the operation on all readers in the working area, and outputting an RFID network planning scheme.
Compared with the prior art, the invention has the technical advantages that:
(1) The RFID network problem is optimized from a new angle by simulating the sliding process of the curling in the curling movement, the reader is regarded as the curling sliding in the working area for searching and optimizing, the tag in the working area provides resistance to prevent the reader from sliding so that the reader stops and covers the tag, and the kinematics method is applied.
(2) By applying the physical law of the hindering effect of friction force on moving objects in kinematics, the reader is considered to slide in a working area, each label provides a part of friction force, and finally the reader can be effectively stopped in an area with a large number of labels. For the tags which are already covered by the reader, the friction provided by the tags is reduced, so that the blocking effect of the covered tags on other readers is reduced, and the reader is prevented from repeatedly covering the same tags. The characteristics enable the method to effectively and simply optimize the position of the reader.
(3) The invention increases the number of readers in turn at the initial stage until the coverage rate of the RFID network meets the engineering requirement, and combines certain redundant reader elimination operation, so that the invention can effectively and adaptively distribute the number of readers, and meanwhile, because a single reader is put into a second reader for searching after the searching is finished, rather than initializing a large number of readers for searching, the invention has the characteristic that the calculated amount is relatively small, and the layout of the readers can be quickly realized.
Drawings
FIG. 1 is a schematic block diagram of the RFID network planning method of the present invention;
FIG. 2 is a schematic flow diagram of the present invention;
FIG. 3 is a simplified model schematic of a reader provided by the present invention;
FIG. 4 is a schematic view of the mobile operation of the reader of the present invention;
FIG. 5 is a schematic view of a reader collision operation of the present invention;
FIG. 6 is a schematic representation of the work area of the present invention;
FIG. 7 (a) is a diagram illustrating the result of simulation of the data set C _30 according to the present invention;
FIG. 7 (b) is a diagram illustrating the results of a simulation performed on the data set C _50 according to the present invention;
FIG. 7 (C) is a diagram illustrating the results of a simulation performed on the data set C _100 according to the present invention;
FIG. 7 (d) is a diagram illustrating the result of the simulation of the data set R _30 according to the present invention;
FIG. 7 (e) is a diagram illustrating the results of a simulation performed on the data set R _50 according to the present invention;
FIG. 7 (f) is a diagram showing the simulation result of the data set R _100 according to the present invention;
FIG. 7 (g) is a diagram illustrating the results of a simulation performed on the data set C _500 according to the present invention;
FIG. 7 (h) is a diagram illustrating the result of simulation performed on the data set R _500 according to the present invention;
FIG. 8 is a graph comparing the coverage of the data set C _100 simulated by the present invention with the genetic algorithm and the particle swarm algorithm over time;
FIG. 9 is a graph comparing the simulation results of the present invention with the time variation of the interference rate of the data set C _100, which is simulated by the genetic algorithm and the particle swarm algorithm.
Detailed Description
The present invention will be described in detail below with reference to the accompanying drawings.
Example 1
The invention relates to a rapid and effective RFID network planning method for simulating curling motion, which comprises the following steps of referring to figure 1,
the RFID network planning method for simulating curling motion comprises the following steps:
step one, simulating the movement of the curling to construct a reader model, and simulating the movement of the curling to design the movement operation and the collision operation of the reader. The curling can slide on the ice surface, the curling can be decelerated under the action of friction force in the sliding process, if a boundary is arranged around the curling playground, the curling can rebound to the boundary when sliding to the boundary, a certain initial speed is given to enable the curling to slide in the curling playground, the curling can stop in the curling playground after a period of time, according to the movement phenomenon, a reader is regarded as the curling, an RFID network working area is regarded as the curling playground, and tags to be covered in the project are scattered in the working area.
The tag can change the friction coefficient of the ice surface to hinder the reader from sliding, and the reader can rebound when sliding to the boundary of a working area, so that the moving operation and the collision operation of the reader are designed, and the reader can simulate the movement process of the curling to perform movement search in the working area.
1.1, constructing a reader model, defining an RFID network working area to be optimized as a curling playground, wherein the working area is smooth and flat and has a boundary, labels to be covered in engineering are scattered in the working area, the reader searches in the working area through movement, and the reader simulates the characteristics of curling to move in the working area through the model. The mass of the reader has a certain mass, but the mass of the reader is reduced when the acceleration is calculated, and the mass of the reader has no influence on the movement of the reader, so that the mass of the reader does not need to be specially defined.
1.2 design the removal operation of reading the ware, the removal of simulation curling removes, and the ware removal operation has following characteristic, and the regional surface friction coefficient that has the label is great, and the regional surface friction coefficient that does not have the label is zero, and the ware removes and can receive the effect of frictional force when meetting the label in the work area, does not have the effect each other between the ware, and removal operation makes the ware can carry out the mobile search in the work area.
1.3 design collision operation of reading the ware, the collision on simulation curling and boundary collides, curling can be rebounded when sliding to the stadium boundary, read ware collision operation and have following characteristic, read the ware and can be rebounded when running into the work area boundary, read the ware and have no collision effect each other, read also have no collision effect between ware and the label, collision time is ignored, the number of collisions is ignored, the speed direction that the ware was read in the collision operation change, the boundary makes and reads the ware and can search in fixed work area.
And step two, constructing a reader according to the reader model, and initializing parameters of the reader in a given parameter threshold range.
Step three, moving the reader in the working area, wherein the current position of the reader in the working area is (x) 0 ,y 0 ) Calculating the position (x) of the moving end point of the reader in a given moving time delta t 1 ,y 1 ) If (x) 1 ,y 1 ) Within the working area, the end point position (x) is moved 1 ,y 1 ) Updating the reader current location (x) 0 ,y 0 ) Adding 1 to the moving times of the reader, and continuously searching in the working area by the reader; if at a given displacementAnd (x) defining that the reader does not do moving operation if the moving end point position of the reader is not in the working area within the moving time delta t 1 ,y 1 ) Modifying the current location (x) of the reader 0 ,y 0 )。
And step four, performing collision operation on the reader in the working area, determining that the reader performs collision operation if the moving end point of the reader is not positioned in the working area within the given one-time moving time delta t, wherein the moving end point of the reader can be understood to be positioned outside the working area, the reader actually collides at a certain position at the edge of the working area, the collision point is (x ', y'), the speed of the reader after collision is unchanged, the direction is changed, the speed v 'of the reader which collides and the speed direction theta' of the reader are calculated and obtained according to elastic collision, the one-time moving time delta t of the next moving operation is corrected, and the reader continues to search in the working area.
And step five, repeatedly executing the step three to the step four, wherein the reader performs motion search in the working area all the time until the movement times of the reader are larger than a preset maximum movement time threshold value, the operation on the reader is completed, at the moment, the tags in the power radius range of the position of the reader are defined to be covered by the reader, the given movement time and the preset maximum movement time threshold value of the reader are set according to the complexity degree of the practical engineering problem, the more complex the problem is, the smaller the given movement time of the reader is, and the larger the preset maximum movement time threshold value is.
And step six, repeating the step two to the step five, constructing a new reader, adding 1 to the number of the readers, and taking the ratio of the covered labels as the coverage rate COV until the coverage rate is greater than a preset minimum coverage rate threshold value, wherein the preset minimum coverage rate threshold value is determined according to the minimum coverage rate requirement of the engineering problem.
And step seven, completing the RFID network planning, performing reader elimination operation on all readers in the working area, completing the operation on all readers in the working area, and outputting an RFID network planning scheme.
According to the invention, by simulating the sliding process of the curling in curling movement, the reader is regarded as that the curling slides in a working area, and the curling moves and collides in the sliding process; the method meets the requirement of minimum coverage rate by increasing the number of the readers in the initial stage, adds 1 to the number of the readers successively in the process, and performs more and reader elimination operations on the readers in the working area when the requirement of the minimum coverage rate is met, thereby effectively reducing the data processing amount and being capable of obtaining reasonable number of the readers in a self-adaptive manner. The invention carries out motion search by simulating the motion of the curling, the tags in the working area provide resistance to prevent the reader from sliding so that the reader stops and covers the tags, and the RFID network is optimized from a new angle by applying a kinematic method.
Example 2
The RFID network planning method for simulating curling motion is the same as that in embodiment 1, wherein the step one of simulating curling motion to construct a reader model specifically comprises:
a reader model R (x, y, v, theta, R) is defined, including location coordinates (x, y, R) of the reader, a velocity magnitude v of the reader, a direction theta of the reader velocity, and a power radius R of the reader.
Wherein, in the second step, a reader is constructed according to the reader model, and the method specifically comprises the following steps:
according to the reader model and using a random algorithm to construct a reader, randomly initializing (x, y) in a working area as a position coordinate of the reader, randomly initializing v and theta in a speed range as an initial speed size and an initial speed direction of the reader, randomly initializing R in a power radius range as a power radius of the reader, and finishing the construction of the reader, wherein the initialized R (x, y, v, theta, R) represents the reader, the power radius range of the parameter reader is determined by the type of the reader, the position coordinate range of the reader is determined according to the area range of engineering problems, the direction theta of the speed of the reader is [ -pi, pi ], and the speed range of the reader is determined by the complexity of the engineering problems.
Example 3
The RFID network planning method for simulating curling motion is the same as the embodiment 1-2, wherein the moving operation of the reader in the working area in the third step comprises the following steps:
3.1 scanning the tags in the power radius range of the current reader, respectively counting the number of times the tags are covered, wherein the tags respectively provide a part of friction force for the reader, the friction force provided by each tag is related to the number of times the tag is covered, and the friction force received by the reader is the sum f of the friction forces provided by all the tags in the power radius;
3.2 calculating the acceleration a of the reader by the friction force received by the reader, and under the action of the acceleration, defining that the friction force received by the reader is not changed within the time delta t of one movement, and according to a kinematic formula, calculating the current position (x) of the reader by the current position 0 ,y 0 ) Calculating the position (x) of the moving end point of the reader according to the current speed v, the speed direction theta of the reader, the acceleration a and delta t of the reader 1 ,y 1 ) If the moving end point is in the working area, completing a moving process, adding one to the moving times of the reader by (x) 1 ,y 1 ) Updating reader location (x) 0 ,y 0 ) And finishing one moving operation of the reader.
Example 4
The RFID network planning method for simulating curling movement is the same as that in embodiment 1-3, wherein the friction force f provided by each tag in step 3.1 is related to the number of times C that the tag is covered, and the friction force experienced by the reader is the sum f of the friction forces provided by all tags in the power radius, specifically:
f＝-μ×m×g
wherein f is n Is a friction coefficient control factor, C (i) represents the number of times that the ith tag is covered by the reader in the current working area, N r Representing the number of tags in the r power radius of the current reader, m is the mass of the reader, g is the gravitational acceleration, mu is the friction coefficient suffered by the reader, mu min Is the minimum value of the coefficient of friction, mu max Is the maximum value of the coefficient of friction and d is the dimension.
Example 5
The RFID network planning method for simulating curling movement is the same as that in embodiments 1-4, and the collision operation of the reader in the working area in step four includes the following steps:
4.1 reading the moving operation end point position (x) 1 ,y 1 ) From the current position (x) of the reader 0 ,y 0 ) And moving the end point position (x) 1 ,y 1 ) Determining a collision boundary of the reader with the work area, thereby calculating a reader slave (x) 0 ,y 0 ) Time to collision boundary Δ t ″, by the kinematic formula, of (x) 0 ,y 0 ) And delta t ', v, theta, and a, calculating and acquiring the position (x ', y ') of the collision point and the direction theta ' of the speed v ' of the reader after collision and the speed of the reader after collision.
4.2 correcting the next reader moving operation, subtracting the time from the reader to the collision boundary from the one moving time delta t to obtain the remaining moving time delta t ' after the reader collision, updating the one moving time delta t of the next reader moving operation by the remaining moving time delta t ', and updating the position (x ') of the reader by (x ', y ') 0 ,y 0 ) And updating the speed v of the reader by v ', and updating the direction theta of the speed of the reader by theta', so as to finish one-time collision operation of the reader.
Example 6
The RFID network planning method for simulating curling movement is the same as that in embodiments 1 to 5, and the process of completing the RFID network planning in step seven includes the following steps:
7.1 scanning all readers constructed in the working area, counting the number of the covered tags of each reader, and deleting the readers with the number of the covered tags being 0 in the working area; for the remaining readers in the working area, finding the reader with the least number of covering tags from the working area every time, if the tag coverage rate of the reader is not reduced or is reduced to be smaller than a preset threshold value, setting the threshold value according to specific engineering requirements, executing deletion operation to delete the reader, repeatedly executing reader elimination operation for a certain number of times until the coverage rate is basically kept stable, taking the position coordinates and the power radius of the remaining readers in the working area as an RFID network planning scheme, and calculating the performance index of the RFID network planning scheme.
The invention belongs to the field of engineering optimization. The method comprises the steps of constructing a reader; carrying out reader moving operation and reader collision operation on the reader; and performing redundant reader elimination operation on redundant readers in the working area, taking the positions and power radiuses of the redundant readers in the working area as an RFID network planning scheme, and calculating the performance index of the RFID network planning scheme. According to the RFID network planning method based on kinematics, through simulating the movement of the curling in the curling project, a better optimization result can be quickly and effectively obtained, and the number of the readers can be determined in a self-adaptive manner by combining the elimination operation of the redundant readers.
Example 7
The method for planning the RFID network simulating the curling motion is the same as that in the embodiment 1-6, before processing, parameters and nouns designed by the method need to be defined, and R (x, y, v, theta, R) is set to represent a reader, wherein (x, y) is the position coordinate of the reader, v is the speed of the reader, theta is the speed direction of the reader, namely theta is the included angle between the speed and the positive direction of an x axis, and R is the power radius of the reader; COV is the working areaCoverage of the middle label, MC being the minimum coverage threshold; t represents the movement times of the reader, and M represents the maximum movement time threshold of the reader; 0 0 (x 0 ,y 0 ) Moving start point for moving operation of reader, 0 1 (x 1 ,y 1 ) A moving end point of the moving operation of the reader, and 0 ' (x ', y ') is a collision point of the collision operation of the reader; TS is set of tags, RS is set of readers, N t Is the number of labels, d r,t Is the distance from the reader r to the tag t, and radius is the power radius of the reader r; f is the friction force on the reader, m is the mass of the reader, g is the gravitational acceleration, a is the acceleration of the reader, μ is the coefficient of friction, μ min Is the minimum value of the coefficient of friction, μ max Is the maximum value of the friction coefficient, S is the distance of the reader moving through one reader moving operation, delta t is the total time of one movement of the reader, f n Is the coefficient of friction control factor and d is the dimension.
The invention provides a fast and effective RFID network planning method for simulating curling motion, which specifically performs RFID network optimization on an RFID network test sample data set containing 100 tags, wherein the data set is tag position data generated according to the positions of articles in warehouse management, and the RFID network planning method for simulating curling motion comprises the following steps as shown in figure 1:
step one, simulating a curling to construct a reader model.
Step two, constructing a reader according to the reader model:
as shown in FIG. 3, a reader model is defined and a reader R (x, y, v, θ, R) is built, and a random number R is generated to be E [0,1 ]]Multiplying the width of the working area by the random number r to obtain x 0 Multiplying the length of the working area by the random number r to obtain y 0 Multiplying the random number r by the velocity magnitude range to obtain v 0 Multiplying the random number r by the speed direction range to obtain θ 0 Multiplying the random number R by the power radius to obtain R 0 ，R(x 0 ,y 0 ,v 0 ,θ 0 ,R 0 ) Indicating an initialized reader. Reader is alwaysHas position parameter, speed size parameter and speed direction parameter, and power radius parameter.
Step three, moving the reader in the working area:
as shown in FIG. 4, the current position of the reader in the working area is O 0 (x 0 ,y 0 ) Scanning the tags in the current power radius of the reader and respectively counting the covering times of the tags, and firstly calculating a friction coefficient control factor f of the tags to the reader according to the covering times of each tag by the following formula n From f n Calculating the friction coefficient mu of the tag to the reader by the minimum friction coefficient and the maximum friction coefficient, calculating the acceleration a of the reader by combining the mu with the gravity acceleration g, calculating the moving distance S of the reader in the one-time moving time delta t by the motion equation, and calculating the moving end point O of the reader by the distance S and the speed direction of the reader 1 (x 1 ,y 1 ) Adding 1 to the number of times t of the reader movement, and recording the position O of the start point of the movement 0 (x 0 ,y 0 ) And the position of the moving end point O 1 (x 1 ,y 1 ) Reader slave point O 0 (x 0 ,y 0 ) Move to point O 1 (x 1 ,y 1 )；
f＝-μ×m×g
a＝-μ×m×g/m＝-μ×g
S＝v×Δt+0.5×a×Δt 2
x 1 ＝x 0 +S×cosθ
y 1 ＝y 0 +S×sinθ
Wherein C (i) represents the number of times that the ith label is covered by the reader in the current working area, N r Indicating the number of tags within the power radius for the current reader r.
Step four, collision operation of the reader in the working area:
as shown in fig. 5, the reader is moved from the current position O 0 (x 0 ,y 0 ) In the process of moving, collision operation occurs, as shown in fig. 6, the outside of the working area is divided into eight areas and four corners, and the specific division method is as follows: region I, x _ low<x 1 <, x _ up and y 1 >, y _ up; region II, x 1 ≧ x _ up and y 1 ≧ y _ up; region III, x 1 >, x _ up and y _ low<y 1 &Y _ up; region IV, x 1 ≧ x _ up and y 1 Y _ low is ≦ y _ low; region V, x _ low<x 1 <, x _ up and y 1 &Y _ low; region VI, x 1 X _ low and y ≦ x _ low 1 Y _ low; region VII, x 1 <, x _ low and y _ low<y 1 &Y _ up; region VIII, x 1 X _ low and y ≦ x _ low 1 ≧y_up。
According to the current position O of the reader 0 (x 0 ,y 0 ) And a recording position O 1 (x 1 ,y 1 ) Judging a collision boundary if O 1 (x 1 ,y 1 ) If the collision boundary is located in the area I, the collision boundary is an upper boundary; if O is 1 (x 1 ,y 1 ) If the reader is located in the area II, the collision boundary can be an upper boundary, a right boundary or an angle of the area II, the boundary which is reached in a short time is taken as the collision boundary by calculating the time of the reader reaching the upper boundary and the right boundary, and if the time is the same, the angle is taken as the collision boundary; when O is present 1 (x 1 ,y 1 ) And similarly, judging the collision boundary by the method when the collision boundary is positioned in other areas.
Determining the collision boundary, and then determining the time for reaching the collision boundary, the acceleration a of the reader, the speed v of the reader, the speed direction theta of the reader and the current position O of the reader 0 (x 0 ,y 0 ) Calculating the position O '(x', y ') of the collision point and the speed of the reader at the collision point according to the position O' (x ', y') of the collision pointIn the collision operation diagram of fig. 5, the included angle between the speed before and after collision and the collision boundary is the same, the speed of the reader after collision is not changed, the speed direction θ ' of the reader after collision and the speed v ' of the reader after collision are obtained, and the position O of the reader is updated by O ' (x ', y ') 0 (x 0 ,y 0 )。
Correcting the next reader moving operation, subtracting the time from the reader to the collision boundary from the one moving time delta t to obtain the remaining moving time delta t ' after the reader collision, updating the one moving time delta t of the next reader moving operation by using the remaining moving time delta t ', and updating the position (x ') of the reader by using (x ', y ') 0 ,y 0 ) And updating the speed v of the reader by v ', and updating the direction theta of the speed of the reader by theta', thereby completing one-time collision operation of the reader.
Step five, repeatedly executing the step three to the step four, wherein the reader carries out motion search in the working area until the movement times of the reader are larger than a preset maximum movement time threshold value, the operation on the reader is completed, and at the moment, the tags in the power radius range of the position of the reader are defined to be covered by the reader;
step six, repeating the step two to the step five, constructing a new reader, adding 1 to the number of the readers, and calculating the coverage rate COV of the labels in the working area by taking the ratio of the covered labels as the coverage rate COV through the following formula:
wherein for any one tag t e TS, if there is one reader r e RS, d is caused r , t Ramiusr, then label r is covered. Cv (t) =1 if the tag t is covered, otherwise Cv (t) =0;
and when the COV is larger than the minimum coverage rate threshold MC, stopping building the reader.
Step seven, scanning all readers constructed in a working area, counting the number of the covered tags of each reader, finding the reader with the least number of the covered tags from the working area each time, and if the tag coverage rate of the reader is not reduced or is reduced to be smaller than a preset threshold value when the reader is deleted, executing a deleting operation to delete the reader; and executing reader elimination operation for a certain number of times until the coverage rate is basically kept stable, taking the position coordinates and the power radii of the remaining readers in the working area as an RFID network planning scheme and calculating the performance index of the RFID network planning scheme.
Example 8
The RFID network planning method for simulating curling motion is the same as that in the embodiment 1-7, and the effect of the invention is verified by simulation experiments.
1. Experiment operating environment and condition setting:
environment of experimental run: the processor is Intel Pentium B940@2.0GHz, the memory is 4.00GB, the hard disk is 320GB, the operating system is Microsoft windows 10, and the programming environment is Visual Studio Ultimate 2013.
Setting experimental conditions: 8 RFID test samples were tested in the experiment. Tags in C _30, C _50, and C _100 satisfy a positive distribution. The labels in R _30, R _50, and R _100 are randomly distributed. C _500 and R _500 are generated by linear combination of C _100 and R _100, respectively. C _30 and R _30 contain 30 tags, C _50 and R _50 contain 50 tags, C _100 and R _100 contain 100 tags, and C _500 and R _500 contain 500 tags. The working area of the first 6 test samples is a rectangular area of 50m × 50m, and the power radius of the reader ranges from 8m to 15m. The initial speed v of the reader is 10m/s to 20m/s, and the value range of the angle theta is-pi to pi. For the reader moving operator, the number of searches M is set to 100, and the search time Δ t of each search is set to 0.5s. The value of μmin is 0, the value of μmax is 0.5; the working area of the C _500 and R _500 test samples is a rectangular area of 150m × 150 m.
2. Simulation experiment content and result analysis:
simulation experiment 8 RFID test samples are tested by adopting the RFID network planning method for simulating curling motion and compared with the existing method. Existing methods include genetic algorithms and particle swarm algorithms.
To illustrate the effectiveness of the method of the present invention, 8 RFID test samples were tested, respectively, and the results are shown in FIGS. 7 (a) -7 (h).
Fig. 7 (a) is a simulation result of the data set C _30, fig. 7 (b) is a simulation result of the data set C _50, fig. 7 (C) is a simulation result of the data set C _100, fig. 7 (d) is a simulation result of the data set R _30, fig. 7 (e) is a simulation result of the data set R _50, fig. 7 (f) is a simulation result of the data set R _100, fig. 7 (g) is a simulation result of the data set C _500, and fig. 7 (h) is a simulation result of the data set R _500, and it can be seen that, for test samples of different numbers of tags and test samples of different tag distributions, the requirement of the RFID network planning can be effectively realized by the method (CA-RNP) of the present invention, the position of the reader in the working area can be reasonably allocated, the size of the power radius of the reader and the number of readers required for the adaptive determination are set, and the RFID network planning can be effectively realized.
Example 9
The same RFID network planning method for simulating curling motion as in embodiments 1 to 8, and to illustrate the rapidity of the method of the present invention, the C _100 test sample is tested in the following by combining the present invention and the GA and PSO algorithms of the TRE operator in the prior art, and the change curves of the coverage rate and the interference rate with time are counted. The interference ratio ITF is calculated as follows:
wherein, for the tag t belonging to the tag set TS, in (t) represents the number of times the tag is covered by the reader In the working area, N t The number of the labels in the working area is used as the ITF, and the ITF is the ratio of the sum of the covered times of all the labels in the working area to the total number of the labels.
The results of the comparison are shown in fig. 8 and fig. 9, respectively, where fig. 8 is the simulation result of the coverage rate of the three algorithms with time on the data set C _100, and fig. 9 is the simulation result of the interference rate of the three algorithms with time on the data set C _ 100. It can be seen that the method (CA-RNP) of the present invention can quickly optimize the RFID network, so that the coverage of the tags in the working area is rapidly increased in a short time, and when the coverage is up to the requirement, the interference rate is relatively low.
Example 10
The method for planning the RFID network for simulating curling motion is the same as that in embodiments 1-9, and for easy understanding, a brief processing flow of the present invention is provided below, and fig. 2 is referred to for easy understanding of the solution of the present invention.
1. The process is started.
2. The reader is initialized and placed in the work area.
3. And calculating the coverage rate of the label in the current working area, and determining the relation between the coverage rate and a preset minimum coverage rate threshold value.
4. The value of the number of reader movements t is set to 1.
5. And carrying out reader moving operation on the reader, recording the position of a moving starting point of the reader and the position of a moving ending point of the reader, and adding 1 to the moving times t of the reader.
6. It is determined whether a collision operation is performed.
7. And performing collision operation on the reader, acquiring the position of a collision point and the speed of the reader after collision, and correcting the next movement operation of the reader.
8. And repeating the steps from 5 to 7 until the moving times t of the reader are greater than a preset maximum moving time threshold M of the reader.
9. Repeating the steps 2 to 8 until the value of the coverage COV is greater than the preset minimum coverage threshold MC.
10. Redundant reader eliminating operation is performed on the readers in the working area.
11. The position and power radius of the reader in the operating area are output as a result.
The invention designs a RFID network planning method based on kinematics by simulating the movement of a curling in curling items, which comprises the steps of initializing a reader by a reader 2, calculating the current coverage rate COV of an RFID network in a reader 3, comparing the relation between the coverage rate and a minimum coverage rate threshold MC, stopping if the relation is reached, otherwise, carrying out certain times of moving operation on the reader, adding 1 to t after each time of moving operation is finished, judging whether collision occurs or not in the moving process, if the collision occurs, executing collision operation, stopping the operation on the reader when the moving times of the reader reach a maximum moving time threshold of the reader, repeating the operations in the steps 2 to 8, calculating the coverage rate COV of the current RFID network, if the COV is larger than the minimum coverage rate threshold, carrying out redundant reader elimination operation on all readers in the current network, scanning the number of covered tags of each reader in the current RFID network, deleting the readers with the number of covered tags of 0, and finally outputting the positions and power radiuses of the remaining readers in the RFID network as results. The method can quickly and effectively obtain better optimization results, and can search by continuously adding readers into the working area and performing reader moving operation and reader collision operation to increase the coverage rate of the RFID network, when the coverage rate meets the minimum coverage rate threshold value, the readers in the network are selected, redundant readers in the network are deleted by combining with the redundant reader eliminating operation, and the number of the readers is increased firstly to meet the coverage rate operation and the redundant reader eliminating operation, so that the method can determine the number of the readers in a self-adaptive manner.
In short, the RFID network planning method for simulating curling motion comprises the steps of firstly simulating curling motion to construct a reader model, and designing the movement and collision operation of the reader; constructing a reader; defining the moving operation and the collision operation of the reader in a working area; the reader always carries out movement search of movement and collision in the working area until the movement times of the reader is larger than a preset maximum movement time threshold value to finish one movement operation of the reader; constructing a new reader, adding 1 to the number of readers, and calculating the COV (coverage rate) of all readers on the label in the working area until the coverage rate is greater than a preset minimum coverage rate threshold value; and finishing the RFID network planning, eliminating all readers in the working area, and outputting an RFID network planning scheme. The method can quickly and effectively obtain the optimization result, and can determine the number of the readers in a self-adaptive manner by combining the elimination operation of the readers. The method can be applied to the optimization of the engineered RFID network.
In conclusion, the RFID network is optimized by the method, and the RFID network planning scheme can be quickly and effectively obtained.
It should be noted that: the embodiment of performing the RFID network planning by using the RFID network planning method for simulating curling motion is only used as an explanation of the RFID network planning method in a test example, and the RFID network planning method can also be used in other application scenarios according to actual needs, and the specific implementation process is similar to the embodiment.
The sequence numbers in the above embodiments are merely for description, and do not represent the sequence of the assembly or the use of the components.
The above description is only exemplary of the present invention and should not be taken as limiting the invention, as any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (5)
1. An RFID network planning method for simulating curling motion is characterized by comprising the following steps:
simulating curling motion to construct a reader model, and simulating curling motion to design the moving operation and the collision operation of the reader;
1.1, constructing a reader model, defining an RFID network working area to be optimized as an ice kettle sports field, wherein the working area is smooth and flat and has a boundary, tags to be covered in engineering are scattered in the working area, the reader searches in the working area through movement, and the constructing of the reader model specifically comprises the following steps:
defining a reader model R (x, y, v, theta, R), including position coordinates (x, y, R) of the reader, a speed magnitude v of the reader, a direction theta of the speed of the reader, and a power radius R of the reader;
1.2 designing the mobile operation of the reader, wherein the mobile operation of the reader has the following characteristics that the surface friction coefficient of an area with a label is larger, the surface friction coefficient of an area without the label is zero, the reader is under the action of friction force when moving in a working area and meets the label, and the readers have no action with each other;
1.3 designing collision operation of the reader, wherein the collision operation of the reader has the following characteristics that the reader can rebound when meeting the boundary of a working area, the readers have no collision effect, the readers and a label have no collision effect, the collision time is ignored, and the collision times are ignored;
step two, constructing a reader according to the reader model, and specifically comprising the following steps:
constructing a reader according to the reader model by using a random algorithm, randomly initializing (x, y) in a working area as a position coordinate of the reader, randomly initializing v and theta in a speed range, randomly initializing R in a power radius range as an initial speed size and an initial speed direction of the reader, and finishing the construction of the reader, wherein the initialized R (x, y, v, theta, R) represents the reader;
step three, moving the reader in the working area, wherein the current position of the reader in the working area is (x) 0 ,y 0 ) Calculating the position (x) of the moving end point of the reader in a given moving time delta t 1 ,y 1 ) If (x) 1 ,y 1 ) Is located atInside the working area, the position of the end point of the movement (x) 1 ,y 1 ) Updating the reader current location (x) 0 ,y 0 ) Adding 1 to the moving times of the reader, and continuously searching by the reader in a working area; if the position of the moving end point of the reader is not in the working area within the given one-time moving time delta t, the reader is defined not to do moving operation, and is not used (x) 1 ,y 1 ) Modifying the current location (x) of the reader 0 ,y 0 )；
Step four, performing collision operation on the reader in a working area, determining that the reader performs the collision operation if a movement end point of the reader is not in the working area within a given one-time movement time delta t, wherein the reader collides at a certain position at the edge of the working area, the collision point is (x ', y'), the speed of the reader is unchanged after the collision, the direction is changed, the speed v 'of the reader which collides and the speed direction theta' of the reader are calculated and obtained according to elastic collision, the one-time movement time delta t of the next movement operation is corrected, and the reader continues to search in the working area;
step five, repeatedly executing the step three to the step four, wherein the reader carries out motion search in the working area until the movement times of the reader are larger than a preset maximum movement time threshold value, the operation on the reader is completed, and at the moment, the tags in the power radius range of the position of the reader are defined to be covered by the reader;
step six, repeating the step two to the step five, constructing a new reader, adding 1 to the number of the readers, and taking the ratio of the covered labels as a coverage rate COV until the coverage rate is greater than a preset minimum coverage rate threshold value;
and step seven, completing RFID network planning, performing reader elimination operation on all readers in the working area, completing the operation on all readers in the working area, and outputting an RFID network planning scheme.
2. The RFID network planning method for simulating curling motion according to claim 1, wherein the moving operation of the reader in the working area in the third step comprises the following steps:
3.1 scanning the tags in the power radius range of the current reader, respectively counting the number of times the tags are covered, wherein the tags respectively provide a part of friction force for the reader, the friction force provided by each tag is related to the number of times the tag is covered, and the friction force applied to the reader is the sum f of the friction forces provided by all the tags in the power radius;
3.2 calculating the acceleration a of the reader by the friction force received by the reader, and defining that the friction force received by the reader is not changed in the one-time moving time delta t and the current position (x) of the reader 0 ,y 0 ) Calculating the position (x) of the end point of the movement of the reader according to the current speed v, the speed direction theta, the acceleration a and delta t of the reader 1 ,y 1 ) If the movement end point is in the work area, one movement process is completed.
3. The RFID network planning method for simulating curling movement according to claim 2, wherein the friction provided by each tag in step 3.1 is related to the number of times C that the tag is covered, and the friction experienced by the reader is the sum f of the friction provided by all tags in the power radius, specifically:
f＝-μ×m×g
wherein, f n Is a friction coefficient control factor, C (i) represents the number of times that the ith tag is covered by the reader in the current working area, N r Representing the number of tags in the power radius of the current reader r, m is the mass of the reader, g is the gravity acceleration, mu is the friction coefficient suffered by the reader, and mu is min Is the minimum value of the coefficient of friction, mu max Is a massageThe maximum value of the friction coefficient, d is the dimension.
4. The RFID network planning method for simulating curling motion according to claim 1, wherein the collision operation of the reader in the working area in the fourth step comprises the following steps:
4.1 reading the moving operation end point position (x) 1 ,y 1 ) From the current position (x) of the reader 0 ,y 0 ) And moving the end point position (x) 1 ,y 1 ) Determining a collision boundary of the reader with the work area, thereby calculating a reader slave (x) 0 ,y 0 ) Time to collision boundary Δ t ″, consisting of (x) 0 ,y 0 ) Calculating and acquiring a collision point position (x ', y') and a direction theta 'of a speed v' after the reader collides with the reader;
4.2 correcting the next reader moving operation, subtracting the time from the reader to the collision boundary from the one moving time delta t to obtain the remaining moving time delta t ' after the reader collision, updating the one moving time delta t of the next reader moving operation by the remaining moving time delta t ', and updating the position (x ') of the reader by (x ', y ') 0 ,y 0 ) And updating the speed v of the reader by v ', and updating the direction theta of the speed of the reader by theta', thereby completing one-time collision operation of the reader.
5. The RFID network planning method for simulating curling motion according to claim 1, wherein the process of completing RFID network planning in the seventh step comprises the following steps:
7.1 scanning all readers constructed in a working area, counting the number of covered tags of each reader, finding the reader with the least number of covered tags from the working area each time, and if the tag coverage rate of the reader is not reduced or is reduced to be smaller than a preset threshold value when the reader is deleted, executing deletion operation to delete the reader; and executing reader elimination operation for a certain number of times until the coverage rate is basically kept stable, taking the position coordinates and the power radius of the remaining readers in the working area as an RFID network planning scheme, and calculating the performance index of the RFID network planning scheme.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610247877.XA CN105956504B (en) | 2016-04-20 | 2016-04-20 | The RFID network planing method of simulated curling motion |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610247877.XA CN105956504B (en) | 2016-04-20 | 2016-04-20 | The RFID network planing method of simulated curling motion |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105956504A CN105956504A (en) | 2016-09-21 |
| CN105956504B true CN105956504B (en) | 2018-03-06 |
Family
ID=56917878
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610247877.XA Active CN105956504B (en) | 2016-04-20 | 2016-04-20 | The RFID network planing method of simulated curling motion |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105956504B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110781700B (en) * | 2019-11-11 | 2021-05-14 | 湖南大学 | RFID multi-reader coordination method |
| CN111225367B (en) * | 2020-01-08 | 2022-03-04 | 西安电子科技大学 | RFID network planning method based on hybrid particle swarm optimization |
| CN111208823B (en) * | 2020-02-17 | 2023-04-07 | 哈尔滨工业大学 | Motion control system and method for ice cleaning robot of ice pot |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103679073A (en) * | 2012-08-31 | 2014-03-26 | 手持产品公司 | Method for pairing wireless scanner through RFID (Radio Frequency Identification Device) |
| CN103729680A (en) * | 2013-12-24 | 2014-04-16 | 西安电子科技大学 | RFID network topology method based on multi-agent evolutionary algorithm |
-
2016
- 2016-04-20 CN CN201610247877.XA patent/CN105956504B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103679073A (en) * | 2012-08-31 | 2014-03-26 | 手持产品公司 | Method for pairing wireless scanner through RFID (Radio Frequency Identification Device) |
| CN103729680A (en) * | 2013-12-24 | 2014-04-16 | 西安电子科技大学 | RFID network topology method based on multi-agent evolutionary algorithm |
Non-Patent Citations (4)
| Title |
|---|
| A ＲFID network planning method based on genetic algorithm;YANG Y H etal.;《2009 InternationalConference on Networks Security，Wireless Communications and Trusted Computing，2009》;20091231;第534-537页 * |
| ＲFID networks planning using evolutionary algorithms and swarm intelligence;CHEN H N etal.;《4th International Conference on Wireless Communications，Networking and Mobile Computing，2008》;20081231;第1-4页 * |
| 基于多种群共生进化的RFID网络优化;陈瀚宁 等;《解放军理工大学学报(自然科学版)》;20081031;第9卷(第5期);第413-416页 * |
| 用于ＲFID 网络规划的含Expansion关系的多目标云优化算法;高鹰 等;《广州大学学报( 自然科学版)》;20141031;第13卷(第5期);第67-75页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105956504A (en) | 2016-09-21 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105956504B (en) | The RFID network planing method of simulated curling motion | |
| CN106774312B (en) | Method and device for determining moving track of mobile robot | |
| Shahid et al. | Continuous control actions learning and adaptation for robotic manipulation through reinforcement learning | |
| Godoy et al. | Adaptive learning for multi-agent navigation | |
| Franceschetti et al. | Robotic arm control and task training through deep reinforcement learning | |
| Zhao et al. | A path planning method based on multi-objective cauchy mutation cat swarm optimization algorithm for navigation system of intelligent patrol car | |
| Herman et al. | Inverse reinforcement learning of behavioral models for online-adapting navigation strategies | |
| Curtis et al. | Efficient and effective grasping of novel objects through learning and adapting a knowledge base | |
| Zhang et al. | A systematic solution of human driving behavior modeling and simulation for automated vehicle studies | |
| CN108615070A (en) | A kind of TDOA and AOA hybrid locating methods based on Chaos particle swarm optimization algorithm | |
| Liu et al. | Robust individual and holistic features for crowd scene classification | |
| CN115147790A (en) | Vehicle future trajectory prediction method based on graph neural network | |
| Xiong et al. | Analysis of an efficient rule-based motion planning system for simulating human crowds | |
| CN111225367B (en) | RFID network planning method based on hybrid particle swarm optimization | |
| Doellinger et al. | Environment-aware multi-target tracking of pedestrians | |
| CN113406957B (en) | Mobile robot autonomous navigation method based on immune deep reinforcement learning | |
| Hoshino et al. | Motion Planning of Multiple Mobile Robots Based on Artificial Potential for Human Behavior and Robot Congestion | |
| CN109377510B (en) | Particle tracking method and system on supercomputing cluster | |
| Németh et al. | Hierarchical control design of automated vehicles for multi-vehicle scenarios in roundabouts | |
| Wu | Data mining model based on improved ant colony algorithm | |
| Chen et al. | RFID dense reader network anti-collision PSO model and solving | |
| Acuna et al. | Dynamic potential field generation using movement prediction | |
| Uchibe | Deep inverse reinforcement learning by logistic regression | |
| CN113536597B (en) | Speed-based dynamic crowd simulation method optimized through data driving | |
| Godoy et al. | C-nav: Implicit coordination in crowded multi-agent navigation |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |