CN107627314A - A kind of pathfinding robot, Pathfinding system and method for searching based on genetic algorithm - Google Patents
A kind of pathfinding robot, Pathfinding system and method for searching based on genetic algorithm Download PDFInfo
- Publication number
- CN107627314A CN107627314A CN201711148700.5A CN201711148700A CN107627314A CN 107627314 A CN107627314 A CN 107627314A CN 201711148700 A CN201711148700 A CN 201711148700A CN 107627314 A CN107627314 A CN 107627314A
- Authority
- CN
- China
- Prior art keywords
- robot
- image
- pathfinding
- signal
- processor
- 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.)
- Pending
Links
Abstract
The invention discloses a kind of pathfinding robot based on genetic algorithm, the robot includes：Machine human body, the machine human body include：Robot head, the machine person and robot chassis；Image collecting device is provided with the robot head；Processor, which is provided with, in the robot body holds area；The processor, which is held, is provided with image processor, master controller and path planning processor in area；Stepper motor, driving wheel, bracing strut and two layers of pillar are installed on the robot chassis, the bracing strut side is provided with the electric machine support for supporting DC brushless motor, two layers of supporting plate are installed, two layers of supporting plate bottom is provided with damping spring on two layers of pillar.Have the advantages that vdiverse in function, intelligence degree is high, efficiency high and accuracy are high.
Description
Technical field
The present invention relates to robotic technology field, more particularly to a kind of pathfinding robot based on genetic algorithm, pathfinding system
System and method for searching.
Background technology
Mobile robot path planning is an important research field of robotics, and artificial intelligence and robotics
One binding site.Whether the mobile robot of which kind of classification, require that (such as track route total length is most according to a certain criterion
It is short, minimum power consumption etc.), walked in working space along the path of one optimal (or suboptimum).
The typical method of path planning has graph search method, Grid Method, Artificial Potential Field Method etc., and these algorithms have certain limitation
Property, locally optimal solution is easily absorbed in, and genetic algorithm has good applicability on solving nonlinear problem, it has also become advise in path
A kind of more method is used in drawing.But the genetic algorithm of standard is easily absorbed in locally optimal solution etc. in itself there is also precocity
Defect, it is impossible to ensure the requirement to computational efficiency on path planning and reliability.
In order to improve the solution quality of path planning and solution efficiency, propose that one kind is based on preselected mechanism niche technique
Improved adaptive GA-IAGA, and be applied to the path planning of mobile robot, the complicated two-dimensional coordinate of use is one-dimensional seat
Target coded system, effectively reduce the search space of genetic algorithm；According to the walking feature of mobile robot, devise adaptive
Answer crossover operator, adaptive mutation rate, insertion operator, deletion operator, disturbing operator and inverse operators.Pass through Computer Simulation
Demonstrate the genetic algorithm after improving and significantly improve search efficiency and convergence rate, and can guarantee that and converge to globally optimal solution,
The shortcomings that overcoming standard genetic algorithm, quickly seek the optimal path that a nothing is touched for robot.
Patent No. CN201710037836.2 patent discloses a kind of navigation method for searching, robot and system, main
Environmental information of the robot in motion process is trained is gathered by collecting device, and according in the training motion process
Environmental information generation training data；Depth nerve net is trained using the method for deeply study using the training data
Network, so that the deep neural network after training is suitable to carry out Analysis of Policy Making according to pathfinding data, and indicate the robot
To navigation target pathfinding is done to move, data that the pathfinding data are gathered in pathfinding application process according to the collecting device and
The navigation target obtains.Although automatic pathfinding can be realized, its intelligence degree is still relatively low, and efficiency is not also high, accurately
Property is relatively low.
The content of the invention
It is an object of the invention to provide a kind of pathfinding robot, Pathfinding system and method for searching based on genetic algorithm,
Have the advantages that vdiverse in function, intelligence degree is high, efficiency high and accuracy are high.
The technical solution adopted by the present invention is as follows：
A kind of pathfinding robot based on genetic algorithm, the robot include：Machine human body, the machine human body include：Machine
The device number of people, the machine person and robot chassis；Image collecting device is provided with the robot head；Set in the robot body
It is equipped with processor and holds area；The processor, which is held, is provided with image processor, master controller and path planning processor in area；
Stepper motor, driving wheel, bracing strut and two layers of pillar are installed, the bracing strut side is provided with use on the robot chassis
In the electric machine support of support DC brushless motor, two layers of supporting plate are installed on two layers of pillar, under two layers of supporting plate
Portion is provided with damping spring.
Further, LMS laser sensors are additionally provided with the chassis.
Further, it is provided with pallet on the robot arm.
Further, user's touch-screen is provided with the robot arm.
A kind of Pathfinding system of the pathfinding robot based on genetic algorithm, the system include：For obtaining surrounding environment
The image collecting device of original image information；Described image harvester signal is connected to for carrying out figure to original image information
As the image processor of processing；Described image processor signal is connected to the master controller of control machine people operation；The master control
Signal is connected to the stepper motor for driving robot motion and the path for planning robot's motion path to device processed respectively
Planning processor.
Further, described image processor includes：For the image sharpening unit being sharpened to image；Described image
Sharpening cell signal is connected to the image segmentation unit for being split to image；Described image cutting unit signal is connected to
For carrying out the binarization unit of binary conversion treatment to image；The binarization unit signal is connected to for carrying out threshold to image
It is worth the threshold skirt detection unit of rim detection；The threshold skirt detection unit signal is connected to master controller.
Further, the path planning processor includes：For establishing initial population and determining the initial of genetic parameter
Change unit, the initialization unit signal are connected to the fitness computing unit for calculating individual adaptation degree；The fitness
Computing unit signal is connected to the genetic manipulation unit for carrying out genetic manipulation；The genetic manipulation cell signal is connected to use
In the comparing unit for comparing substring and father's string size.
Further, the master controller includes：For the power supply powered to whole system；The power supply signal is connected to
The data processing unit of master controller data message is commuted for handling；The data processing unit signal is connected to for passing
The data transmission unit of transmission of data signal and the memory for data storage information.
A kind of method for searching of the Pathfinding system based on genetic algorithm, it is characterised in that the method for searching includes following
Step：
Step 1：System starts, system initialization；
Step 2：Image collecting device starts to gather original picture signal, and picture signal is sent after the processing of master controller
To path planning processor；
Step 3：Path planning processor carries out the action path of signal planning robot according to the image received, will plan
Action path be sent to master controller；
Step 4：Master controller sends control command to stepper motor according to the action path of planning, controls driving stepper motor machine
The motion of device people.
Further, the method for the path planning processor progress path planning comprises the following steps：
Step 1：Initialize population, N number of point chosen along beginning and end line direction is equidistant, on the vertical line of these points with
Machine chooses the ordinate of turning point, and makes these turning points not in barrier；
Step 2：N class will be divided into per generation individual, and the larger individual of some fitness be selected in each class, as a class
Outstanding representative, form a population；Obtaining population scale is：；
Step 3：All individual fitness in population are calculated, its best individual is retained, then using algorithm of tournament selection method,
Father's individual is selected, to perform crossover operation, and whether the offspring individual chromosome length of acquisition is checked more than N, if do not had
Exceed, then retain, otherwise abandon；
Step 4：Row variation, insertion, disturbance, deletion, smooth operation are entered to new caused offspring individual with the probability of setting:；
Meanwhile preselected mechanism is taken, compare the size of substring and father's string fitness, if the fitness of substring is higher than the adaptation of father's string
Degree, just replace father's string；Otherwise maintain father's string constant；
Step 5：Step 3 and step 4 step are repeated until the new individual quantity of acquisition is equal with parent Population；
Step 6：The individual that fitness is worst in new population is replaced with the previous generation optimum individuals of reservation；
Step 7：Check algorithm stop condition.Meet, stop, otherwise jump to:Step 3, algorithm continues.
Beneficial effects of the present invention are as follows：
Using above technical scheme, present invention produces following beneficial effect：
1st, accuracy is high：The image processor of the present invention has carried out improved edge threshold detection, the edge threshold of acquisition to image
It is more accurate to be worth testing result.The path planning processor of the present invention employs Revised genetic algorithum simultaneously, and the algorithm satisfies the need
The planning in footpath is more accurate.
2nd, operational efficiency is high：The present invention genetic algorithm simultaneously taken into account genetic evolution rapidity and colony it is various
Property, the generation of " precocity " phenomenon is restrained effectively, locally optimal solution and globally optimal solution can be searched for well.The algorithm is not
With environment in can converge to optimal solution in less evolutionary generation, the execution speed and success rate of algorithm apparently higher than
The genetic algorithm of standard.In addition, the suitable intersection of different phase selection and mutation probability in evolution have for evolution result
Critical influence.The edge threshold algorithm of the present invention greatly improves the place of system on the premise of accuracy is ensured simultaneously
Efficiency is managed, treatment effeciency is higher.
3rd, cost is low, simple in construction：The robot system of the present invention, the annexation between each processor is simple, image
Identification, master controller and path planning processor can be individually produced and assembled again, greatly reduced complexity, reduced
Cost of manufacture.
Brief description of the drawings
Fig. 1 is the structural representation of the pathfinding robot based on genetic algorithm of the present invention.
Fig. 2 is the structural representation of the Pathfinding system of the pathfinding robot based on genetic algorithm of the present invention.
Fig. 3 is the schematic flow sheet of the pathfinding algorithm of the pathfinding robot based on genetic algorithm of the present invention..
Wherein：1- image collecting devices, 2- robots chassis, 3- processors hold area, 4- machine human bodies, 5- robots
Head.
Specific embodiment
All features disclosed in this specification, or disclosed all methods or during the step of, except mutually exclusive
Feature and/or step beyond, can combine in any way.
This specification（Including any accessory claim, summary）Disclosed in any feature, unless specifically stated otherwise,
Replaced by other equivalent or with similar purpose alternative features.I.e., unless specifically stated otherwise, each feature is a series of
An example in equivalent or similar characteristics.
Embodiment 1：
As shown in figure 1, a kind of pathfinding robot based on genetic algorithm, the robot include：Machine human body, the robot
Body includes：Robot head, the machine person and robot chassis；Image collecting device is provided with the robot head；The machine
Processor, which is provided with, in device human body holds area；The processor, which is held, is provided with image processor, master controller and path in area
Planning processor；Stepper motor, driving wheel, bracing strut and two layers of pillar, the bracing strut are installed on the robot chassis
Side is provided with the electric machine support for supporting DC brushless motor, and two layers of supporting plate are provided with two layers of pillar, described
Two layers of supporting plate bottom are provided with damping spring.
Further, LMS laser sensors are additionally provided with the chassis.
Further, it is provided with pallet on the robot arm.
Further, user's touch-screen is provided with the robot arm.
Embodiment 2：
As shown in Fig. 2 a kind of Pathfinding system of the pathfinding robot based on genetic algorithm, the system include：For obtaining week
The image collecting device of collarette border original image information；Described image harvester signal is connected to for original image information
Carry out the image processor of image procossing；Described image processor signal is connected to the master controller of control machine people operation；Institute
Stating master controller, signal is connected to stepper motor for driving robot motion and for planning robot's motion path respectively
Path planning processor.
Further, described image processor includes：For the image sharpening unit being sharpened to image；Described image
Sharpening cell signal is connected to the image segmentation unit for being split to image；Described image cutting unit signal is connected to
For carrying out the binarization unit of binary conversion treatment to image；The binarization unit signal is connected to for carrying out threshold to image
It is worth the threshold skirt detection unit of rim detection；The threshold skirt detection unit signal is connected to master controller.
Further, the path planning processor includes：For establishing initial population and determining the initial of genetic parameter
Change unit, the initialization unit signal are connected to the fitness computing unit for calculating individual adaptation degree；The fitness
Computing unit signal is connected to the genetic manipulation unit for carrying out genetic manipulation；The genetic manipulation cell signal is connected to use
In the comparing unit for comparing substring and father's string size.
Further, the master controller includes：For the power supply powered to whole system；The power supply signal is connected to
The data processing unit of master controller data message is commuted for handling；The data processing unit signal is connected to for passing
The data transmission unit of transmission of data signal and the memory for data storage information.
Embodiment 3：
A kind of as shown in figure 3, method for searching of the Pathfinding system based on genetic algorithm, it is characterised in that the method for searching bag
Include following steps：
Step 1：System starts, system initialization；
Step 2：Image collecting device starts to gather original picture signal, and picture signal is sent after the processing of master controller
To path planning processor；
Step 3：Path planning processor carries out the action path of signal planning robot according to the image received, will plan
Action path be sent to master controller；
Step 4：Master controller sends control command to stepper motor according to the action path of planning, controls driving stepper motor machine
The motion of device people.
Further, the method for the path planning processor progress path planning comprises the following steps：
Step 1：Initialize population, N number of point chosen along beginning and end line direction is equidistant, on the vertical line of these points with
Machine chooses the ordinate of turning point, and makes these turning points not in barrier；
Step 2：N class will be divided into per generation individual, and the larger individual of some fitness be selected in each class, as a class
Outstanding representative, form a population；Obtaining population scale is：；
Step 3：All individual fitness in population are calculated, its best individual is retained, then using algorithm of tournament selection method,
Father's individual is selected, to perform crossover operation, and whether the offspring individual chromosome length of acquisition is checked more than N, if do not had
Exceed, then retain, otherwise abandon；
Step 4：Row variation, insertion, disturbance, deletion, smooth operation are entered to new caused offspring individual with the probability of setting:；
Meanwhile preselected mechanism is taken, compare the size of substring and father's string fitness, if the fitness of substring is higher than the adaptation of father's string
Degree, just replace father's string；Otherwise maintain father's string constant；
Step 5：Step 3 and step 4 step are repeated until the new individual quantity of acquisition is equal with parent Population；
Step 6：The individual that fitness is worst in new population is replaced with the previous generation optimum individuals of reservation；
Step 7：Check algorithm stop condition.Meet, stop, otherwise jump to:Step 3, algorithm continues.
It is described above, be only presently preferred embodiments of the present invention, any formal limitation not done to the present invention, it is every according to
Any simply modification, the equivalent variations made according to the technical spirit of the present invention to above example, each fall within the protection of the present invention
Within the scope of.
Claims (10)
1. a kind of pathfinding robot based on genetic algorithm, it is characterised in that the robot includes：Machine human body, the machine
Device human body includes：Robot head, the machine person and robot chassis；Image collecting device is provided with the robot head；Institute
State and be provided with processor in robot body and hold area；The processor hold be provided with area image processor, master controller and
Path planning processor；Stepper motor, driving wheel, bracing strut and two layers of pillar, the axle are installed on the robot chassis
Support side is provided with the electric machine support for supporting DC brushless motor, and two layers of supporting plate are provided with two layers of pillar,
Two layers of supporting plate bottom is provided with damping spring.
2. the pathfinding robot based on genetic algorithm as claimed in claim 1, it is characterised in that：Also set up on the chassis
There are LMS laser sensors.
3. the pathfinding robot based on genetic algorithm as claimed in claim 1, it is characterised in that：On the robot arm
It is provided with pallet.
4. the pathfinding robot based on genetic algorithm as claimed in claim 1, it is characterised in that：On the robot arm
It is provided with user's touch-screen.
5. a kind of Pathfinding system of the pathfinding robot based on genetic algorithm based on described in any one of one of Claims 1-4,
Characterized in that, the system includes：For obtaining the image collecting device of surrounding environment original image information；Described image is adopted
Acquisition means signal is connected to the image processor for carrying out image procossing to original image information；Described image processor signal
It is connected to the master controller of control machine people operation；Signal is connected to for driving robot motion's the master controller respectively
Stepper motor and the path planning processor for planning robot's motion path.
6. the pathfinding robot system based on genetic algorithm as claimed in claim 1, it is characterised in that described image processor
Including：For the image sharpening unit being sharpened to image；Described image sharpens cell signal and is connected to for entering to image
The image segmentation unit of row segmentation；Described image cutting unit signal is connected to the two-value for carrying out binary conversion treatment to image
Change unit；The binarization unit signal is connected to the threshold skirt detection unit for carrying out threshold skirt detection to image；
The threshold skirt detection unit signal is connected to master controller.
7. the pathfinding robot system based on genetic algorithm as claimed in claim 2, it is characterised in that at the path planning
Reason device includes：For establishing initial population and determining that the initialization unit of genetic parameter, the initialization unit signal are connected to
For calculating the fitness computing unit of individual adaptation degree；The fitness computing unit signal is connected to for carrying out hereditary behaviour
The genetic manipulation unit of work；The genetic manipulation cell signal is connected to the comparing unit for comparing substring and father's string size.
8. the pathfinding robot system based on genetic algorithm as claimed in claim 3, it is characterised in that the master controller bag
Include：For the power supply powered to whole system；The power supply signal is connected to commutes master controller data message for processing
Data processing unit；The data processing unit signal is connected to the data transmission unit for transmitting data-signal and is used for
The memory of data storage information.
9. a kind of method for searching of the Pathfinding system based on genetic algorithm based on described in one of claim 5 to 8, its feature exist
In the method for searching comprises the following steps：
Step 1：System starts, system initialization；
Step 2：Image collecting device starts to gather original picture signal, and picture signal is sent after the processing of master controller
To path planning processor；
Step 3：Path planning processor carries out the action path of signal planning robot according to the image received, will plan
Action path be sent to master controller；
Step 4：Master controller sends control command to stepper motor according to the action path of planning, controls driving stepper motor machine
The motion of device people.
10. the method for searching based on genetic algorithm as claimed in claim 6, it is characterised in that the path planning processor
The method for carrying out path planning comprises the following steps：
Step 1：Initialize population, N number of point chosen along beginning and end line direction is equidistant, on the vertical line of these points with
Machine chooses the ordinate of turning point, and makes these turning points not in barrier；
Step 2：N class will be divided into per generation individual, and the larger individual of some fitness be selected in each class, as a class
Outstanding representative, form a population；Obtaining population scale is：；
Step 3：All individual fitness in population are calculated, its best individual is retained, then using algorithm of tournament selection method,
Father's individual is selected, to perform crossover operation, and whether the offspring individual chromosome length of acquisition is checked more than N, if do not had
Exceed, then retain, otherwise abandon；
Step 4：Row variation, insertion, disturbance, deletion, smooth operation are entered to new caused offspring individual with the probability of setting:；
Meanwhile preselected mechanism is taken, compare the size of substring and father's string fitness, if the fitness of substring is higher than the adaptation of father's string
Degree, just replace father's string；Otherwise maintain father's string constant；
Step 5：Step 3 and step 4 step are repeated until the new individual quantity of acquisition is equal with parent Population；
Step 6：The individual that fitness is worst in new population is replaced with the previous generation optimum individuals of reservation；
Step 7：Check algorithm stop condition.Meet, stop, otherwise jump to:Step 3, algorithm continues.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711148700.5A CN107627314A (en) | 2017-11-17 | 2017-11-17 | A kind of pathfinding robot, Pathfinding system and method for searching based on genetic algorithm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711148700.5A CN107627314A (en) | 2017-11-17 | 2017-11-17 | A kind of pathfinding robot, Pathfinding system and method for searching based on genetic algorithm |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107627314A true CN107627314A (en) | 2018-01-26 |
Family
ID=61108707
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711148700.5A Pending CN107627314A (en) | 2017-11-17 | 2017-11-17 | A kind of pathfinding robot, Pathfinding system and method for searching based on genetic algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107627314A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113759922A (en) * | 2021-09-14 | 2021-12-07 | 安徽工程大学 | Robot path planning method based on spring algorithm |
-
2017
- 2017-11-17 CN CN201711148700.5A patent/CN107627314A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113759922A (en) * | 2021-09-14 | 2021-12-07 | 安徽工程大学 | Robot path planning method based on spring algorithm |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Orozco-Rosas et al. | Mobile robot path planning using membrane evolutionary artificial potential field | |
| CN107403426B (en) | Target object detection method and device | |
| CN104794214B (en) | A kind of method for designing big data driving cloud robot | |
| CN103699136B (en) | Intelligent home service robot system based on the algorithm that leapfrogs and method of servicing | |
| US20200319609A1 (en) | Action Control Method and Apparatus | |
| CN103412490A (en) | Polyclone artificial immunity network algorithm for multirobot dynamic path planning | |
| CN112327821A (en) | Intelligent cleaning robot path planning method based on deep reinforcement learning | |
| Niroui et al. | Robot exploration in unknown cluttered environments when dealing with uncertainty | |
| US11776409B2 (en) | Methods, internet of things systems and storage mediums for street management in smart cities | |
| CN113532461B (en) | Robot autonomous obstacle avoidance navigation method, equipment and storage medium | |
| CN108871341A (en) | A kind of concurrently positioning of global optimization and build drawing method | |
| KR102425657B1 (en) | Mobile robot capable of evaluating self-training based traversability | |
| Sadhu et al. | A modified imperialist competitive algorithm for multi-robot stick-carrying application | |
| CN110497419A (en) | Building castoff sorting machine people | |
| Agah et al. | Phylogenetic and ontogenetic learning in a colony of interacting robots | |
| Harun et al. | A genetic algorithm based task scheduling system for logistics service robots | |
| Lou et al. | Path planning in an unknown environment based on deep reinforcement learning with prior knowledge | |
| CN107627314A (en) | A kind of pathfinding robot, Pathfinding system and method for searching based on genetic algorithm | |
| Moon et al. | Path planning of cleaning robot with reinforcement learning | |
| CN109976158B (en) | AUV energy optimization path searching method based on distance evolution N-PSO | |
| CN113342029A (en) | Maximum sensor data acquisition path planning method and system based on unmanned aerial vehicle cluster | |
| Racz et al. | Artificial neural network for mobile robot topological localization | |
| Shi et al. | Path planning for mobile robots in complex environments based on improved ant colony algorithm. | |
| CN108227718A (en) | A kind of automatic transporting trolley path planning method adaptively switched | |
| Chidambaran et al. | Multi-criteria evolution of neural network topologies: Balancing experience and performance in autonomous systems |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20180126 |