CN116460830A - Robot intelligent control system and control method based on artificial intelligence - Google Patents
Robot intelligent control system and control method based on artificial intelligence Download PDFInfo
- Publication number
- CN116460830A CN116460830A CN202310262675.2A CN202310262675A CN116460830A CN 116460830 A CN116460830 A CN 116460830A CN 202310262675 A CN202310262675 A CN 202310262675A CN 116460830 A CN116460830 A CN 116460830A
- Authority
- CN
- China
- Prior art keywords
- robot
- individual
- module
- population
- design module
- 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
- 238000000034 method Methods 0.000 title claims abstract description 31
- 238000013473 artificial intelligence Methods 0.000 title claims abstract description 15
- 238000013461 design Methods 0.000 claims abstract description 71
- 210000000349 chromosome Anatomy 0.000 claims abstract description 49
- 230000002068 genetic effect Effects 0.000 claims abstract description 39
- 238000012549 training Methods 0.000 claims abstract description 7
- 230000000007 visual effect Effects 0.000 claims description 10
- 238000012163 sequencing technique Methods 0.000 claims description 7
- 238000003491 array Methods 0.000 claims description 6
- 230000035772 mutation Effects 0.000 claims description 4
- 238000013459 approach Methods 0.000 claims description 3
- 238000001514 detection method Methods 0.000 claims description 3
- 229910002056 binary alloy Inorganic materials 0.000 claims 1
- 238000004088 simulation Methods 0.000 abstract 1
- 238000012804 iterative process Methods 0.000 description 3
- 230000003044 adaptive effect Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000006698 induction Effects 0.000 description 2
- 230000001960 triggered effect Effects 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 230000003542 behavioural effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 230000008447 perception Effects 0.000 description 1
- 230000002787 reinforcement Effects 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
- B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
- B25J9/00—Programme-controlled manipulators
- B25J9/08—Programme-controlled manipulators characterised by modular constructions
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
- B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
- B25J13/00—Controls for manipulators
- B25J13/08—Controls for manipulators by means of sensing devices, e.g. viewing or touching devices
- B25J13/087—Controls for manipulators by means of sensing devices, e.g. viewing or touching devices for sensing other physical parameters, e.g. electrical or chemical properties
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
- B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
- B25J9/00—Programme-controlled manipulators
- B25J9/16—Programme controls
- B25J9/1602—Programme controls characterised by the control system, structure, architecture
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
- B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
- B25J9/00—Programme-controlled manipulators
- B25J9/16—Programme controls
- B25J9/1656—Programme controls characterised by programming, planning systems for manipulators
- B25J9/1664—Programme controls characterised by programming, planning systems for manipulators characterised by motion, path, trajectory planning
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/02—Total factory control, e.g. smart factories, flexible manufacturing systems [FMS] or integrated manufacturing systems [IMS]
Abstract
The invention discloses an intelligent robot control system and a control method based on artificial intelligence, which relate to the technical field of intelligent robot control and comprise a maze design module, a robot attribute setting module, an individual design module, a population design module and a robot advancing module; the simulation digital maze for robot movement design training is designed in advance by arranging a maze design module; the robot attribute setting module is used for setting a plurality of basic attribute setting individual design modules for the robot in advance, and the individual characteristic coding form of each individual in the genetic algorithm is preset; setting a population design module to design population characteristics of a genetic algorithm according to individual characteristics; setting the robot advancing module to make action decisions for the robot in real time according to the chromosome segment data of the individual; the self-adaptive genetic factors and the immune genetic algorithm improved by utilizing the thought of the immune algorithm are established on the basis of the traditional genetic algorithm, so that the reliability and the effectiveness of a robot control system are ensured.
Description
Technical Field
The invention belongs to the field of robot control, relates to genetic algorithm technology, and in particular relates to an intelligent robot control system and method based on artificial intelligence.
Background
Currently robots have evolved to third generation intelligent robots; the intelligent robot is used as a main development direction. According to the external environment and the completion condition of the current task, various environment information can be comprehensively analyzed, and the most suitable solution is intelligently selected from various solutions, so that the problem is more perfectly processed, and the problem solving efficiency and capacity of the intelligent robot are improved; the control method adopted by the third-generation intelligent robot often uses the control technologies such as machine vision perception, reinforcement learning or genetic algorithm and the like;
in the traditional method for controlling the robot by using the genetic algorithm, the problem of overlarge operation complexity caused by long chromosome length or complex coding form is often caused; thus, there is a need to establish adaptive genetic factors and immune genetic algorithms improved using the ideas of immune algorithms;
therefore, an intelligent robot control system and a control method based on artificial intelligence are provided.
Disclosure of Invention
The present invention aims to solve at least one of the technical problems existing in the prior art. Therefore, the invention provides an intelligent control system and a control method of the robot based on artificial intelligence, which are improved on the basis of the traditional genetic algorithm, and build an adaptive genetic factor and an immune genetic algorithm improved by utilizing the thought of the immune algorithm. The improved robot control system ensures the reliability and the effectiveness of the robot control system.
To achieve the above object, an embodiment according to a first aspect of the present invention proposes an artificial intelligence based robot intelligent control system, including a maze design module, a robot attribute setting module, an individual design module, a population design module, and a robot traveling module; wherein, each module is connected by an electric and/or wireless network mode;
the maze design module is mainly used for designing an analog digital maze for training for the movement of the robot in advance;
the analog-digital labyrinth designed by the labyrinth design module comprises walls and a plurality of passages; the analog-digital labyrinth comprises a planned robot travel route;
in the analog-to-digital labyrinth, a visual sensor of the robot is activated upon detection of wall approach; dividing the environment in the analog-digital labyrinth into a starting point, an end point, walls and a passage; the path is in two types, namely, on a planned route designed in advance, is a road through which the robot is planned to pass, and the other path robot can freely pass, but is not on the planned route, and is a road which the robot needs to avoid passing as far as possible;
digitally expressing the environments in the analog-digital maze so as to facilitate calculation by a program; specifically, 1 represents a wall, 2 represents a starting point, and 3 represents a path on a pre-planned travelling route; 4 represents an end point and 0 represents a path that is not on the pre-planned travel route;
the maze design module sends the designed analog-digital maze to the robot running module;
the robot attribute setting module is mainly used for setting a plurality of basic attributes for the robot in advance;
the robot attribute setting module sets basic attributes for the robot, wherein the basic attributes comprise the position of the robot, the movement mode of the robot, the direction of the robot, the maximum moving step number of the robot, the current step number of the robot and the running route of the robot;
it can be understood that the position of the robot, the current step number of the robot and the travel route of the robot are all changed in real time according to the advancing of the robot;
the robot movement mode comprises horizontal movement, clockwise rotation and anticlockwise rotation; the orientation of the robot changes with the clockwise or counterclockwise rotation of the robot;
storing a travel route of the robot by using an array; the maximum moving step number of the robot and the current step number of the robot are used for preventing the algorithm from running infinitely; when the current step number of the robot is continuously increased along with the movement of the robot until the current step number of the robot is equal to the maximum movement step number of the robot, the algorithm terminates the operation;
the robot attribute setting module sends the basic attribute setting of the robot to the robot advancing module;
the individual design module is mainly used for setting individual characteristics of each individual in the genetic algorithm;
the individual design module adopts a binary expression mode for the robot moving instruction: use 00 for stop, 01 for forward, 10 for clockwise rotation, and 10 for counterclockwise rotation;
considering that the robot in the present embodiment has 6 sensing sensors, each of which is located at a different position and each of which has an on/off state, 2 can be given 6 Seed, namely 64 possible sensor input combinations; therefore, to represent the robot actions corresponding to all the induction sensors under different input combinations, a binary number with a certain length is needed, and the relation between the two is corresponding as an individual chromosome;
for 64 different sensor state combinations, the robot may receive one of them at any time, and each action of the known robot may be encoded with a 2-bit binary number, so that the chromosome of the individual needs to store 128 bits of data;
the individual design module uses a numerical value with the length of 128 bits to store chromosomes; in this case, by using the basic principle of the genetic algorithm, by constantly performing mutation and crossover, it is unnecessary to pay attention to which specific instruction is being modified in the iterative process of the algorithm, and only the chromosome sequence as an individual is operated; after the process is finished, the method is used for controlling the robot by performing inverse coding;
since each individual has 64 different combinations of sensor states, the action code requires 64 in total;
after the sensor bit field is converted into decimal, the desired operation can be placed at the corresponding location on the chromosome.
The individual design module sends the chromosome coding design mode of each individual to the population design module;
the population design module is mainly used for designing population characteristics of a genetic algorithm;
in the process of creating the population, firstly, an individual array with the size of the population is created, then individuals are created for the individual array, the individual chromosomes in the array are encoded in sequence, and the individual chromosomes are added into the individual array. Stopping the creation of individuals and giving the individual arrays to the population when the number of the individuals of the array meets the size of the population; the population comprises an important method for obtaining individuals with specific fitness sequences in the population;
the population design module designs population characteristics of the genetic algorithm in the following way:
firstly, sequencing all individuals in an individual array from high to low according to the size of individual fitness, and returning the individuals sequenced by specific fitness; returning to the individual with the highest fitness in the population when the parameter in the method is 0;
the population design module sends all individual arrays after population sequencing to the robot travelling module;
the robot traveling module is mainly used for making action decisions for the robot in real time;
the mode of the robot moving module for deciding the action is as follows:
after finding the starting point, acquiring the initial abscissa and ordinate positions of the robot; judging the next behavior action of the robot; after judging whether each sensor senses the wall, expressing the sensor by using binary digits; each sensor corresponds to a certain digit of six digits binary numbers; when the sensor senses a wall, the corresponding bit of the binary number is represented by 1, and conversely, by 0; each action is represented by two binary numbers in the individual chromosome to form a 128 array for representing each sensor sensing combined action, and the action corresponding to the sensor combination can be found out from the 128 array of the individual chromosome by using the pointer of the individual chromosome; the numerical value of the pointer represents what action in the array should be corresponding to in the chromosome of the individual, so that the robot can judge which action should be executed according to the signal of the sensor;
the behavior of the robot depends on the sensing combination of the current sensor, and the sensing combination of the sensor is related to the current direction of the robot; different robot orientations, the orientations to be judged by the sensors on the robots are different; and the population design module prepares a judgment azimuth table in advance according to the robot orientation logic.
An embodiment according to a second aspect of the present invention proposes an artificial intelligence based robot intelligent control method comprising the steps of:
step one: the maze design module designs an analog digital maze for training for the movement of the robot in advance;
step two: the robot attribute setting module sets a plurality of basic attributes for the robot in advance;
step three: the individual design module presets the coding form of the individual characteristics of each individual in the genetic algorithm;
step four: the population design module designs population characteristics of a genetic algorithm according to individual characteristics;
step five: and the robot travelling module makes action decisions for the robot in real time according to the chromosome segment data of the individual.
Compared with the prior art, the invention has the beneficial effects that:
the invention combines the on-off states of 6 visual sensors of the robot into 64 state combinations; further, a two-bit binary representation is used for 4 actions of the robot; thus, feature information for each individual is expressed using a 64×2=128 bit binary; and then sequencing all individuals in the individual array from high to low according to the individual fitness, and returning the individuals sequenced by the specific fitness. Returning to the individual with the highest fitness in the population when the parameter in the method is 0; for the analog-digital maze which is created in advance, the chromosome pointer is used for controlling the chromosome of the individual; therefore, the improvement is provided on the basis of the traditional genetic algorithm, and the self-adaptive genetic factors are established and the immune genetic algorithm improved by utilizing the thought of the immune algorithm is established. The improved robot control system ensures the reliability and the effectiveness of the robot control system.
Drawings
FIG. 1 is a schematic diagram of the present invention;
fig. 2 is a flow chart of the present invention.
Detailed Description
The technical solutions of the present invention will be clearly and completely described in connection with the embodiments, and it is obvious that the described embodiments are only some embodiments of the present invention, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
As shown in fig. 1, the robot intelligent control system based on artificial intelligence comprises a maze design module, a robot attribute setting module, an individual design module, a population design module and a robot traveling module; wherein, each module is connected by an electric and/or wireless network mode;
the control of the robot must rely on the collection and analysis of external environmental information, and therefore, the robot is required to be provided with some visual sensors, and the robot in this embodiment is provided with 6 visual sensors: 3 in front, 1 on the left, 1 on the right, 1 on the back;
the maze design module is mainly used for designing an analog digital maze for training for the movement of the robot in advance;
in a preferred embodiment, the analog-to-digital labyrinth of the labyrinth design module design comprises walls and a number of passages; it will be appreciated that the robot does not penetrate the wall, but is free to move in the passageway; further, the analog-digital labyrinth comprises a planned robot travel route;
the invention programs the testing robot by using an improved genetic algorithm, so that the robot can make different actions on different sensors according to preset programming, and finally walk out a route consistent with the planned travelling route as far as possible;
in the analog-to-digital labyrinth, a visual sensor of the robot is activated upon detection of wall approach; further, the environment in the analog-digital labyrinth is divided into a starting point, an end point, a wall and a passage; the path is in two types, namely, on a planned route designed in advance, is a road through which the robot is planned to pass, and the other path robot can freely pass, but is not on the planned route, and is a road which the robot needs to avoid passing as far as possible;
in a preferred embodiment, the environments in the analog-to-digital maze are digitally expressed to facilitate computation by a program; specifically, 1 represents a wall, 2 represents a starting point, and 3 represents a path on a pre-planned travelling route; 4 represents an end point and 0 represents a path that is not on the pre-planned travel route;
the maze design module sends the designed analog-digital maze to the robot running module;
the robot attribute setting module is mainly used for setting a plurality of basic attributes for the robot in advance;
in a preferred embodiment, the basic attributes set by the robot attribute setting module for the robot include the position of the robot, the movement mode of the robot, the direction of the robot, the maximum moving step number of the robot, the current step number of the robot and the travel route of the robot;
it can be understood that the position of the robot, the current step number of the robot and the travel route of the robot are all changed in real time according to the advancing of the robot;
the robot movement mode comprises horizontal movement, clockwise rotation and anticlockwise rotation; the orientation of the robot changes with the clockwise or counterclockwise rotation of the robot;
preferably, the travel route of the robot is stored by using an array; the maximum moving step number of the robot and the current step number of the robot are used for preventing the algorithm from running infinitely; when the current step number of the robot is continuously increased along with the movement of the robot until the current step number of the robot is equal to the maximum movement step number of the robot, the algorithm terminates the operation;
the robot attribute setting module sends the basic attribute setting of the robot to the robot advancing module;
the individual design module is mainly used for setting individual characteristics of each individual in the genetic algorithm;
various characteristic information of an individual is written on the chromosome of the individual, in a basic genetic algorithm, a problem is usually converted into binary numbers or integer forms to be encoded, and then the encoded numbers form a chromosome string; the length and coding form of the chromosome have great influence on the operation efficiency and convergence of the genetic algorithm. Encoding data in a correct manner is often the most troublesome problem in genetic algorithms;
in a preferred embodiment, the individual design module uses binary expression of the robot movement instruction in the following manner: use 00 for stop, 01 for forward, 10 for clockwise rotation, and 10 for counterclockwise rotation;
further, considering that the robot in the present embodiment has 6 sensing sensors, each of which is located at a different position and each of which has two states of on/off, 2 can be given 6 Seed, namely 64 possible sensor input combinations; therefore, to represent the robot actions corresponding to all the induction sensors under different input combinations, a binary number with a certain length is needed, and the relation between the two is corresponding as an individual chromosome;
for 64 different sensor state combinations, the robot may receive one of them at any time, and each action of the known robot may be encoded with a 2-bit binary number, so that the chromosome of the individual needs to store 128 bits of data;
the individual design module uses a numerical value with the length of 128 bits to store chromosomes; in this case, by using the basic principle of the genetic algorithm, by constantly performing mutation and crossover, it is unnecessary to pay attention to which specific instruction is being modified in the iterative process of the algorithm, and only the chromosome sequence as an individual is operated; after the process is finished, the method is used for controlling the robot by performing inverse coding;
as one example:
for the following visual sensor states:
visual sensor 1 (front): opening;
vision sensor 2 (left front): closing;
vision sensor 3 (front right): opening;
vision sensor 4 (left): closing;
visual sensor 5 (right): closing;
visual sensor 6 (rear): closing;
then randomly assigning an instruction action, such as left turn, to the user, namely using the '10' in the binary expression to control; the sensor number is then discarded and is directly represented by the front-to-back position: 000101; correspondence 10 indicates a left turn motion. Finally, converting the binary string of the sensor into decimal, and obtaining decimal expression 5; further, since there are 64 different combinations of sensor states per individual, a total of 64 such action codes are required;
to represent it entirely in one chromosome, decimal numbers representing the sensor state may be used as locations in the chromosome, such that when the chromosome is established, each different location represents a commanded action, and the structure of the individual chromosome is as follows:
xxxxxxxx xx xx 10 xxxxxx xx xx are (64 groups together)
In the above-described pseudochromosome, the first pair (position 0) represents the action when the sensor input combination is 0: all off; the second pair (position 1) represents the action when the sensor input combination is 1: only the front sensor detects the wall; the third team (position 2) indicates that only the front left sensor is triggered; a fourth pair (position 3) indicating that both the front and left front sensors are activated; and so on until the last pair (position 63) indicating that all sensors are triggered;
after the sensor bit field is converted into decimal, the desired operation can be placed at the corresponding location on the chromosome.
This coding scheme, while inconvenient for people to read and understand, has several convenient features. First, chromosomes can operate as bit arrays rather than a complex tree structure or hash map, which makes interleaving, mutation, and other operations simple and easy; second, each 128-bit value is a valid solution;
the individual design module sends the chromosome coding design mode of each individual to the population design module;
the population design module is mainly used for designing population characteristics of a genetic algorithm;
in the process of creating the population, firstly, an individual array with the size of the population is created, then individuals are created for the individual array, the individual chromosomes in the array are encoded in sequence, and the individual chromosomes are added into the individual array. Stopping the creation of individuals and giving the individual arrays to the population when the number of the individuals of the array meets the size of the population; the population comprises an important method for obtaining individuals with specific fitness sequences in the population;
in a preferred embodiment, the population design module designs the population characteristics of the genetic algorithm in the following manner:
firstly, sequencing all individuals in an individual array from high to low according to the size of individual fitness, and returning the individuals sequenced by specific fitness; returning to the individual with the highest fitness in the population when the parameter in the method is 0;
the population design module sends all individual arrays after population sequencing to the robot travelling module;
the robot traveling module is mainly used for making action decisions for the robot in real time;
in a preferred embodiment, the manner of the robot motion module to motion decision is:
after finding the starting point, acquiring the initial abscissa and ordinate positions of the robot; judging the next behavior action of the robot; after judging whether each sensor senses the wall, expressing the sensor by using binary digits; each sensor corresponds to a certain digit of six digits binary numbers; when the sensor senses a wall, the corresponding bit of the binary number is represented by 1, and conversely, by 0; each action is represented by two binary numbers in the individual chromosome to form a 128 array for representing each sensor sensing combined action, and the action corresponding to the sensor combination can be found out from the 128 array of the individual chromosome by using the pointer of the individual chromosome; the numerical value of the pointer represents what action in the array should be corresponding to in the chromosome of the individual, so that the robot can judge which action should be executed according to the signal of the sensor;
for example, the front sensor corresponds to the last bit of the binary number, the front left sensor corresponds to the penultimate bit of the binary number, etc.; this allows all sensors to have a unique corresponding number that can be used as a pointer to the individual chromosome;
preferably, the behavioral actions of the robot depend on the sensed combination of current sensors, which are related to the current orientation of the robot; different robot orientations, the orientations to be judged by the sensors on the robots are different; and the population design module prepares a judgment azimuth table in advance according to the robot orientation logic.
As shown in fig. 2, the robot intelligent control method based on artificial intelligence comprises the following steps:
step one: the maze design module designs an analog digital maze for training for the movement of the robot in advance; the analog-digital labyrinth comprises a wall, a plurality of passages and a preset travelling road;
step two: the robot attribute setting module sets a plurality of basic attributes for the robot in advance; the method comprises the steps of a robot position, a robot motion mode, a robot orientation, a maximum moving step number of the robot, a current step number of the robot and a robot travelling route;
step three: the individual design module presets the coding form of the individual characteristics of each individual in the genetic algorithm; preferably, each individual feature is stored as a 128-bit binary digit string; by utilizing the basic principle of the genetic algorithm, the variation and the crossover are continuously carried out, and the specific instruction is not required to be changed in the iterative process of the algorithm, but the chromosome sequence serving as an individual is only operated; after the process is completed, performing inverse coding on the process for robot control;
step four: the population design module designs population characteristics of a genetic algorithm according to individual characteristics;
step five: and the robot travelling module makes action decisions for the robot in real time according to the chromosome segment data of the individual.
The above embodiments are only for illustrating the technical method of the present invention and not for limiting the same, and it should be understood by those skilled in the art that the technical method of the present invention may be modified or substituted without departing from the spirit and scope of the technical method of the present invention.
Claims (8)
1. The robot intelligent control system based on artificial intelligence is characterized by comprising a maze design module, a robot attribute setting module, an individual design module, a population design module and a robot traveling module; wherein, each module is connected by an electric and/or wireless network mode;
the maze design module is used for designing an analog digital maze for training for the movement of the robot in advance; the designed analog digital maze is sent to a robot travelling module;
the robot attribute setting module is used for presetting a plurality of basic attributes for the robot; the basic attribute setting of the robot is sent to a robot advancing module;
the individual design module is used for setting individual characteristics of each individual in the genetic algorithm; the chromosome coding design mode of each individual is sent to a population design module;
the population design module designs population characteristics of a genetic algorithm according to individual characteristics; all individual arrays after population sequencing are sent to a robot travelling module;
the robot advancing module is used for making action decisions for the robot according to the chromosome segment data of the individual in real time.
2. The intelligent robot control system according to claim 1, wherein the analog-to-digital labyrinth designed by the labyrinth design module comprises walls and a number of passages; the analog-digital labyrinth comprises a planned robot travel route;
in the analog-to-digital labyrinth, a visual sensor of the robot is activated upon detection of wall approach; dividing the environment in the analog-digital labyrinth into a starting point, an end point, walls and a passage; the path is two types, namely, on a planned route designed in advance, is a road through which the robot is planned to pass, and is the road through which the robot needs to avoid as much as possible.
3. The artificial intelligence based robot intelligence control system of claim 1, wherein each environment in the analog-to-digital labyrinth uses a digital representation; 1 represents a wall, 2 represents a starting point, and 3 represents a passage on a pre-planned travelling route; 4 represents the end point and 0 represents the path that is not on the pre-planned travel route.
4. The intelligent robot control system according to claim 1, wherein the basic attributes set by the robot attribute setting module include a position of the robot, a movement mode of the robot, an orientation of the robot, a maximum number of moving steps of the robot, a current number of steps of the robot, and a travel route of the robot; the robot movement mode comprises horizontal movement, clockwise rotation and anticlockwise rotation; the orientation of the robot changes as the robot rotates clockwise or counterclockwise.
5. The intelligent robot control system according to claim 1, wherein the individual design module uses binary expression of the robot movement instruction as follows: use 00 for stop, 01 for forward, 10 for clockwise rotation, and 10 for counterclockwise rotation;
presetting 64 possible sensor input combinations; for each different combination of sensor states, using a two-bit binary number to perform an expression of the action; namely, 128-bit binary system is used for storing the chromosomes of the individuals; operating the chromosome sequence as an individual by constantly performing mutation and crossover by using the basic principle of the genetic algorithm; and after the process is finished, performing inverse coding on the process to realize the control of the robot.
6. The intelligent control system of an artificial intelligence based robot of claim 1, wherein the population design module designs the population characteristics of the genetic algorithm in the following manner:
firstly, sequencing all individuals in an individual array from high to low according to the size of individual fitness, and returning the individuals sequenced by specific fitness; when the parameter in the method is 0, returning to the individual with the highest fitness in the population.
7. The artificial intelligence based robot intelligence control system of claim 1, wherein the manner of the robot travel module to action decision is:
after finding the starting point, acquiring the initial abscissa and ordinate positions of the robot; judging the next behavior action of the robot; after judging whether each sensor senses the wall, expressing the sensor by using binary digits; each sensor corresponds to a certain digit of six digits; when the sensor senses a wall, the corresponding bit of the binary number is represented by 1, and conversely, by 0; each action is represented by two binary numbers in an individual chromosome, a 128 array group representing each sensor sensing combined action is formed, and the action corresponding to the sensor combination is found out from the 128 array groups of the individual chromosome by using the pointer of the individual chromosome; the value of the pointer represents what action in the array should be corresponding to in the individual chromosome, so that the robot can judge which action should be executed according to the signal of the sensor.
8. The artificial intelligence based robot intelligent control method according to any one of claims 1 to 7, comprising the steps of:
step one: the maze design module designs an analog digital maze for training for the movement of the robot in advance;
step two: the robot attribute setting module sets a plurality of basic attributes for the robot in advance;
step three: the individual design module presets the coding form of the individual characteristics of each individual in the genetic algorithm;
step four: the population design module designs population characteristics of a genetic algorithm according to individual characteristics;
step five: and the robot travelling module makes action decisions for the robot in real time according to the chromosome segment data of the individual.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310262675.2A CN116460830B (en) | 2023-03-17 | 2023-03-17 | Robot intelligent control system and control method based on artificial intelligence |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310262675.2A CN116460830B (en) | 2023-03-17 | 2023-03-17 | Robot intelligent control system and control method based on artificial intelligence |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116460830A true CN116460830A (en) | 2023-07-21 |
| CN116460830B CN116460830B (en) | 2023-10-20 |
Family
ID=87177908
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310262675.2A Active CN116460830B (en) | 2023-03-17 | 2023-03-17 | Robot intelligent control system and control method based on artificial intelligence |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116460830B (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1888995A (en) * | 2006-07-10 | 2007-01-03 | 北京工业大学 | Intelligent maze robot |
| KR20120013057A (en) * | 2010-08-04 | 2012-02-14 | 중앙대학교 산학협력단 | Method and apparatus for path planning of autonomous mobile robot |
| CN105005823A (en) * | 2015-06-26 | 2015-10-28 | 天津大学 | Ship branch pipeline path planning method based on genetic algorithm |
| CN109176503A (en) * | 2018-07-27 | 2019-01-11 | 东南大学 | A kind of intelligent path detection robot and path detection method based on bionical antenna |
| CN110865642A (en) * | 2019-11-06 | 2020-03-06 | 天津大学 | Path planning method based on mobile robot |
| CN114879494A (en) * | 2022-04-25 | 2022-08-09 | 复旦大学 | Robot self-adaptive design method based on evolution and learning |
| CN115576333A (en) * | 2022-12-08 | 2023-01-06 | 青岛科技大学 | Optimal direction obstacle avoidance strategy |
-
2023
- 2023-03-17 CN CN202310262675.2A patent/CN116460830B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1888995A (en) * | 2006-07-10 | 2007-01-03 | 北京工业大学 | Intelligent maze robot |
| KR20120013057A (en) * | 2010-08-04 | 2012-02-14 | 중앙대학교 산학협력단 | Method and apparatus for path planning of autonomous mobile robot |
| CN105005823A (en) * | 2015-06-26 | 2015-10-28 | 天津大学 | Ship branch pipeline path planning method based on genetic algorithm |
| CN109176503A (en) * | 2018-07-27 | 2019-01-11 | 东南大学 | A kind of intelligent path detection robot and path detection method based on bionical antenna |
| CN110865642A (en) * | 2019-11-06 | 2020-03-06 | 天津大学 | Path planning method based on mobile robot |
| CN114879494A (en) * | 2022-04-25 | 2022-08-09 | 复旦大学 | Robot self-adaptive design method based on evolution and learning |
| CN115576333A (en) * | 2022-12-08 | 2023-01-06 | 青岛科技大学 | Optimal direction obstacle avoidance strategy |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116460830B (en) | 2023-10-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107186713B (en) | Multi-axis motion planning optimization method for mechanical arm passing through path point | |
| CN110275526A (en) | A kind of method for planning path for mobile robot based on improved adaptive GA-IAGA | |
| CN105069220B (en) | Microbial fermentation optimization method based on BP neural network immune genetic algorithm | |
| CN113435025B (en) | Robot high-performance track automatic generation method combined with multistage optimization model | |
| CN110889552B (en) | Apple automatic boxing path optimization method based on optimal parameter genetic algorithm | |
| CN116460830B (en) | Robot intelligent control system and control method based on artificial intelligence | |
| CN103279796A (en) | Method for optimizing genetic algorithm evolution quality | |
| CN110516807B (en) | Semiconductor product yield extreme value calculating method and extreme value calculating system thereof | |
| CN115577647B (en) | Power grid fault type identification method and intelligent agent construction method | |
| CN111489787A (en) | Method for predicting efficiency of targeted knockout of fixed-point DNA (deoxyribonucleic acid) by CRISPR/Cas9 | |
| CN114879494A (en) | Robot self-adaptive design method based on evolution and learning | |
| Chow et al. | On-line fuzzy identification using genetic algorithms | |
| CN107390690A (en) | Merge the robot path planning method of chaos and elite backward learning | |
| CN115439581A (en) | Body animation binding acceleration method based on deep learning | |
| CN103974388B (en) | A kind of wireless sensor network data fusion method and device | |
| Ueyama et al. | Knowledge acquisition and distributed decision making | |
| Hao et al. | Masked imitation learning: Discovering environment-invariant modalities in multimodal demonstrations | |
| CN116127584B (en) | Dynamic aisle arrangement method | |
| Ueyama et al. | Cooperative search using genetic algorithm based on local information-Path planning for structure configuration of cellular robot | |
| Wei | Study on genetic algorithm and evolutionary programming | |
| Yalcin | Evolving aggregation behavior for robot swarms: A cost analysis for distinct fitness functions | |
| CN114326749B (en) | Deep Q-Learning-based cluster area coverage method | |
| CN117556903A (en) | Boolean network reasoning method and system based on MGA-BNI algorithm | |
| CN106874499A (en) | A kind of optimum synergistic team construction method and system based on set covering theory | |
| Ding et al. | A new DNA-based evolutionary algorithm with application to the design of fuzzy controllers |
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 |