CN109709951A - A kind of intelligence storage cart system based on machine learning - Google Patents
A kind of intelligence storage cart system based on machine learning Download PDFInfo
- Publication number
- CN109709951A CN109709951A CN201811404423.4A CN201811404423A CN109709951A CN 109709951 A CN109709951 A CN 109709951A CN 201811404423 A CN201811404423 A CN 201811404423A CN 109709951 A CN109709951 A CN 109709951A
- Authority
- CN
- China
- Prior art keywords
- trolley body
- main control
- module
- article
- control 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.)
- Pending
Links
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The present invention relates to a kind of, and the intelligence based on machine learning stores cart system.Intelligence storage cart system of the present invention based on machine learning includes: including trolley body and the main control module, voice interaction module, image collecting device, electronic compass, moving device and the clamp device that are equipped in the trolley body；The main control module receives the voice command of the voice interaction module, determine that user specifies the classification and positioning of article according to the azimuth information that described image acquisition device acquired image information and the electronic compass are recorded, and it controls the motor drive module and the trolley body is driven to be moved to article position, control the clamp device clamping article and article is transported to initial position.Intelligence storage cart system of the present invention based on machine learning has and can identify to article, sorts out the advantages of placing.
Description
Technical field
The present invention relates to smart home equipment technical fields, small more particularly to a kind of intelligence storage based on machine learning
Vehicle system.
Background technique
Sweeping robot also known as automatic sweeping machine, intelligent dust suction, robot dust suction etc., are one kind of controlling intelligent household appliances,
Certain artificial intelligence can be relied on, complete floor-cleaning work in the room automatically, such as the hair in room, melon seeds can be cleaned
The ground such as shell, ash sundries, and receive into the rubbish storage box of itself, to complete the function of land clearing.With life
Horizontal raising, the diversification of family life articles, the ground environment of Modern Family become complicated, the object being scattered on the ground
Or rubbish is also diversified, such as doll all kinds of, not of uniform size, Che Mo and pop can, brings to parents' cleaning
Certain puzzlement and trouble.Sweeping robot on the market is typically only capable to carry out indiscriminate cleaning to various articles at present, and
All kinds of articles can not be distinguished and carry out identification classification.Obviously, traditional sweeping robot is unable to satisfy Modern Family ground
The cleaning and arrangement demand of complex environment.
Summary of the invention
Based on this, the object of the present invention is to provide a kind of, and the intelligence based on machine learning stores cart system, has
Article can be identified, sort out the advantages of placing.
A kind of intelligence storage cart system based on machine learning, including trolley body and be equipped in the trolley body
Main control module and be electrically connected with the main control module voice interaction module, image collecting device, electronic compass, traveling
Device and clamp device；The voice interaction module receives the voice command of user, carries out identification to voice command and will identify
As a result it is sent to the main control module；Described image acquisition device is used to acquire the image information of article, and by figure collected
As information is sent to the main control module；The electronic compass is used to record the azimuth information of the trolley body rotation, and
Azimuth information is sent to the main control module；The moving device includes the motor for driving the trolley body mobile
Drive module；The main control module receives the speech recognition result of the voice interaction module transmission, is acquired according to described image
The azimuth information that device acquired image information and the electronic compass are recorded determine user specify article classification and
Positioning, and control the motor drive module and the trolley body is driven to be moved to article position, control the clamping dress
It sets clamping article and article is transported to initial position.
Intelligence storage cart system of the present invention based on machine learning, receives user's by voice interaction module
Instruction receives image captured by image collecting device by main control module and identifies to object, and combines electronic compass
Azimuth information positions article, then controls motor drive module driving trolley and march at required object, control clamping
Device clamps required article, and article is transported to initial position, to realize the identification of article and sort out placement.
Further, (SuSE) Linux OS is used inside the main control module, and is built in the (SuSE) Linux OS
Vertical svm classifier model；The main control module carries out HOG feature extraction to described image acquisition device acquired image, and will
The HOG feature inputs the svm classifier model and carries out Classification and Identification, and the classification to determine object and object are in the picture
Geometric position, to identify required article and be positioned to the article.
Further, the svm classifier model foundation image recognition tag library, and identification frame is set, described image is acquired
Device acquired image is scanned, and judges that the characteristic information for the article that the identification frame extracts and described image identify
Whether the label information in tag library is consistent.When the characteristic information and described image of the article that identification frame extracts identify tag library
In image tag it is consistent when, judge the article for the article of required searching.
Further, the moving device further includes the multiple traveling motors for being installed on the trolley body two sides；It is multiple
The traveling motor is connected to the control signal output of the motor drive module；The main control module passes through motor driven mould
Block obtains the turnning circle of the traveling motor, calculates the straight line scalar distance of the trolley body movement, and by described
Straight line scalar distance and azimuth information position the trolley body.
Further, the main control module is recorded the trolley body and searches the path of object and deposited with stacked data
Storage, and drive the trolley body to return according to the inverse process progress path for searching object by executing its stacked data stored
It traces back, object is transported to initial position.
Further, the voice interaction module includes the voice recognition chip being connected with microphone and is connected to described
Controller between voice recognition chip and the main control module；The voice recognition chip knows the phonetic order of user
And to the controller speech recognition result is not sent；The controller receives institute's speech recognition result, and will by serial ports
Institute's speech recognition result is sent to the main control module, and controls the voice recognition chip and issue reply voice.
Further, the clamp device includes being respectively arranged in two steering engines of the trolley body front end and being set to institute
State two mechanical arms on steering engine；Two mechanical arm extends and gathers to form folder object space to the trolley body direction of advance,
To be clamped to article.
It further, further include the infrared sensor being set between two steering engine；What the infrared sensor issued
Whether infrared light is directed toward the folder object space, consolidate to detect mechanical arm described in trolley transport process to the clamping of article.
Further, the X-direction of the electronic compass is parallel with the middle line of the trolley body, Y-direction and the trolley
The median perpendicular of ontology.The design is so that the rotation direction of trolley body and the rotation direction of electronic compass are consistent, convenient for small
The rotation direction of vehicle ontology carries out real-time monitoring.
It further, further include the PWM output module being connected with the main control module；The motor drive module includes
Multiple full bridge driving circuits；The full bridge driving circuit includes the driving chip connecting with the signal output end of PWM output module
With the metal-oxide-semiconductor for being connected to the driving chip signal output end；The output end of the metal-oxide-semiconductor is connected to the traveling motor；Institute
It states driving chip to receive pwm pulse signal and control the on and off of the metal-oxide-semiconductor, to control the rotation of the traveling motor
Speed, to control the travel speed of trolley body.
In order to better understand and implement, the invention will now be described in detail with reference to the accompanying drawings.
Detailed description of the invention
Fig. 1 is total functional block diagram of intelligence storage cart system of the present invention；
Fig. 2 is the circuit diagram of voice interaction module of the present invention；
Fig. 3 is the voice control flow chart of voice interaction module of the present invention；
Fig. 4 is the circuit diagram of electronic compass of the present invention；
Fig. 5 is the partial circuit diagram of motor drive module of the present invention；
Fig. 6 is the booster circuit figure of motor drive module of the present invention；
Fig. 7 is the LM2940 voltage regulator circuit figure of power module of the present invention；
Fig. 8 is the XL4016 voltage regulator circuit figure of power module of the present invention；
Fig. 9 is the software design principle flow chart of intelligence storage cart system of the present invention.
Specific embodiment
In the description of the present invention, it is to be understood that, term " center ", " longitudinal direction ", " transverse direction ", "upper", "lower",
The orientation or positional relationship of the instructions such as "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outside" is
It is based on the orientation or positional relationship shown in the drawings, is merely for convenience of description of the present invention and simplification of the description, rather than instruction or dark
Show that signified device or element must have a particular orientation, be constructed and operated in a specific orientation, therefore should not be understood as pair
Limitation of the invention.In the description of the present invention, unless otherwise indicated, the meaning of " plurality " is two or more.
Referring to Fig. 1, it is total functional block diagram of intelligence storage cart system of the present invention.
The intelligence storage cart system based on machine learning includes that trolley body (is schemed with the trolley body is equipped on
Do not show) on main control module 10 and be electrically connected with the main control module 10 voice interaction module 20, image collecting device
30, electronic compass 40, moving device 50, clamp device 60 and power module 70；The voice interaction module 20 receives user's
Voice command identify to voice command and recognition result is sent to the main control module 10；Described image acquisition device
30 for acquiring the image information of article, and acquired image information is sent to the main control module 10；Electronics sieve
Disk 40 is used to record the azimuth information of the trolley body rotation, and azimuth information is sent to the main control module 10；
The moving device 60 includes the motor drive module for driving the trolley body mobile；The clamp device 70 is for pressing from both sides
Take article；The power module 80 provides power supply to modules；The main control module 10 receives the voice interaction module 20
The speech recognition result of transmission is remembered according to 30 acquired image information of described image acquisition device and the electronic compass 40
The azimuth information of record determines that user specifies the classification and positioning of article, and controls the motor drive module and drive the trolley
Ontology is moved to article position, controls the clamp device 70 and clamps article and article is transported to initial position.
In the present embodiment, intelligence stores the instruction that trolley receives user by voice interaction module, is connect by main control module
It receives image captured by image collecting device to identify object, and the azimuth information of electronic compass is combined to carry out article
Positioning, then control motor drive module driving motor and march at required object, control clamp device presss from both sides required article
It takes, and article is transported to initial position, to realize the identification of article and sort out placement.
The chassis of the trolley body is liftoff about 2 centimetres, center of gravity is lower, with good stability, adopts when turning to
The mode retreated with unilateral advance, opposite side, so that the steering of trolley body is very flexible.
The main control module 10 is raspberry pie, is the micro electric of multicore ARM Cortex A53 based on 64bit a kind of
Brain mainboard, built-in system are based on (SuSE) Linux OS.Using SD or Micro SD card as memory hard disk, there are 4 around mainboard
USB2.0 interface and HDMI HD video output interface support the video format of PAL and NTSC system formula.
Fig. 2 and Fig. 3 are please referred to, wherein Fig. 2 is the circuit diagram of voice interaction module 20 of the present invention；Fig. 3 is this
Invent the voice control flow chart of the voice interaction module.
The voice interaction module 20 stores for intelligence and carries out interactive voice between trolley and user, realizes voice control
Function.The voice interaction module 20 includes the voice recognition chip being connected with microphone and is connected to the speech recognition core
Controller between piece and the main control module；The voice recognition chip identifies the phonetic order of user and to described
Controller sends speech recognition result；The controller receives institute's speech recognition result, and is known the voice by serial ports
Other result is sent to the main control module, and controls the voice recognition chip and issue reply voice.
Using the LD3320 of IC Route company, controller uses STM32 for voice recognition chip in the present embodiment,
It is communicated between controller and voice recognition chip by SPI mode.Voice recognition chip LD3320 uses ASR technology, mentions
GUI operations mode and the voice-based user interface VUI (Voice such as a kind of disengaging key, keyboard, mouse, touch screen are supplied
User Interface) so that user is simpler to the operation of the system, quick and natural.In addition, the voice recognition chip is also
It can be using other voice recognition chips.Controller STM32 is by configuring the register of voice recognition chip LD3320, in advance
By the key words of required identification, such as: position orange square, vat orange square, positioning all objects, reduction all objects,
And other key words, and inputted in voice recognition chip LD3320 in the form of character string.When work, microphone is received
Analog voice signal, speech recognition are inputted by MICP the and MICN pin of voice recognition chip LD3320 after external sound information
Chip carries out frequency-domain transform to the analog voice signal that is inputted, obtains corresponding pinyin character, and with scheduled key words
Identification string matched, if successful match, speech recognition result is returned into controller STM32, controller STM32
It controls the voice recognition chip LD3320 and issues corresponding reply sound, and corresponding behaviour is sent to main control module 10 by serial ports
Make number data；After main control module 10 receives Action number data, control modules execute corresponding task.
In the present embodiment, the voice control flow example of man-machine communication is given, as shown in Figure 3.When work, intelligence is received
Trolley of receiving executes following steps: S1, system boot are initialized, and after initializing successfully, issue " intelligence storage trolley initialization
The voice prompting of success "；S2, instruction is waited；S3, when user inputs phonetic order " starting input service ", system just issues " intelligence
Trolley can be stored wholeheartedly be you service, instruction please be input " voice prompting, and wait user grab object instruct；S4, work as user
When inputting phonetic order " positioning orange square ", intelligence stores trolley and rotates car body searching object, and issues and " position orange side
The reply voice of block "；S5, when user input phonetic order " vat orange square " when, intelligence store trolley object is transported to
Initial position；When S6, user input phonetic order " closing voice control ", system issues the reply language of " exiting service "
Sound, and wait user's input assignment instructions next time.In order to reduce probability of the trolley by false triggering, system has used circulation to wait
Mode and instruction triggers mode, i.e., when only user inputs enabled instruction to trolley, trolley could start what identification needed to grab
Object instruction, it ensure that the safety of the stability of intelligence storage trolley and user.
Described image acquisition device 30 is used to acquire object information in environment, captured by digital image information be
Rgb format, 120*160 pixel, image collecting device 30 carry out the shooting of different angle to object and obtain in different location
Multiple images arrived, and image transmitting to the main control module 10 captured by it is stored.Specifically, in the present embodiment,
The image collecting device 30 is a colour imagery shot, is installed at the one third of the trolley body middle line by upright bar,
In order to provide the broader visual field and prediction, camera is highly set as 30CM from the ground.In addition, trolley moving process in order to prevent
In camera can be made to shake, also support auxiliary rod is increased in camera upright bar two sides, to form stable triangular structure；Meanwhile
Also it is fixed in the junction of each structure using hot melt adhesive.
The electronic compass 40 is used to record azimuth information when trolley body rotation, and azimuth information is passed
It send to the main control module 10.When installation, the X-direction of the electronic compass 40 is parallel with the middle line of the trolley body, Y-direction
With the median perpendicular of the trolley body.The design is so that the rotation direction of trolley body and the rotation direction one of electronic compass
It causes, carries out real-time monitoring convenient for the rotation direction to trolley body.
Referring to Fig. 4, it is the circuit diagram of electronic compass of the present invention.In the present embodiment, electronic compass is adopted
Concrete model is HMC5883L, is connect between main control module 10 by SCL and SDA, is carried out using I2C communication mode
Communication.In addition, electronic compass can also not influence the determination of trolley rotation direction using other models.
The main control module 10 receives 30 acquired image information of described image acquisition device and the electronic compass 40
The azimuth information recorded, to determine that user specifies the classification and positioning of article.The main control module 10 is operated based on Linux
System, and svm classifier model is established in the (SuSE) Linux OS；The main control module 10 is to described image acquisition device
30 acquired images carry out HOG feature extraction, and the HOG feature is inputted the svm classifier model and carries out Classification and Identification,
To determine the geometric position of the classification and object of object in the picture.The svm classifier model foundation image recognition tag library,
And identification frame is set, described image acquisition device acquired image is scanned, and judge the object that the identification frame extracts
Whether the characteristic information of product and the image tag in described image identification tag library are consistent.
HOG (Histogram of Oriented Gradient) feature, is referred to by gradient and edge direction density
The characteristic information of description.Histograms of oriented gradients (Histogram of Oriented Gradient, HOG) is characterized in that one kind exists
It is used to carry out the Feature Descriptor of object detection in computer vision and image procossing, it is by calculating and statistical picture partial zones
The gradient orientation histogram in domain carrys out constitutive characteristic.
SVM (Support Vector Machine) refers to support vector machines, and in machine learning field, being one has prison
The learning model superintended and directed, for carrying out pattern-recognition, classification and regression analysis.Machine learning function in the present embodiment is to utilize
The subject image that camera collects trains SVM model, the subject image training set include camera in different angle, no
Multiple images obtained from object are shot with position.After the completion of the training of SVM model algorithm, system then can be to extraneous environmental information
Carry out analysis identification.
The system of the main control module 10 uses svm classifier model and carries out to the picture that image collecting device 30 acquires back
Classification and Identification.It is scanned in the image that camera acquisition is returned with the identification frame of size 100*100 pixel, when identification frame mentions
When the characteristic information for the article got is consistent with the label information in described image identification tag library, target object will accurately be known
It not and captures, and returns to the label information and the location information of identified object in the picture of identified object.In combination with electricity
Sub- compass 40 determines azimuth that current trolley mentions the earth, thus determines the specific location of identified object.
The moving device 50 includes motor drive module for driving the trolley body to advance and is installed on described
Multiple traveling motors of trolley body two sides.The main control module 10 controls the traveling of motor by motor drive module and turns
To scanning for and transport to control the trolley body to object.
For the movement speed for preferably controlling the trolley body, it is additionally provided with the PWM being connected with the main control module
Output module.Since the PWM output port quantity that main control chip has is limited, only with the pwm signal output port on 16 tunnels, therefore
It has used PCA9685 module as PWM signal generator, has been communicated between main control chip using I2C, i.e. main control chip
Two GPIO be separately connected the SCL and SDA of PCA9685, pass through this two data lines and configure the register of PCA9685 and realize
The pwm pulse signal of PWM output module exports.PWM refers generally to pulse width modulation, and pulse width modulation is a kind of simulation control
Mode, according to the variation of respective loads come the biasing of modulation transistor base stage or metal-oxide-semiconductor grid, Lai Shixian transistor or MOS
The change of pipe turn-on time.
Please refer to Fig. 5 and Fig. 6, wherein Fig. 5 is the partial circuit diagram of motor drive module；Fig. 6 is motor driven mould
The booster circuit figure of block.
The full bridge driving circuit includes the driving chip connecting with the signal output end of PWM output module and is connected to institute
State the metal-oxide-semiconductor of driving chip signal output end；The output end of the metal-oxide-semiconductor is connected to the traveling motor；The driving chip
The on and off that pwm pulse signal controls the metal-oxide-semiconductor is received, to control the velocity of rotation of the traveling motor.Specifically,
The turn-on time of metal-oxide-semiconductor is adjusted by adjusting the duty ratio of pwm pulse signal, to adjust the velocity of rotation of traveling motor.
In the present embodiment, the quantity of traveling motor is four, and the traveling motor is all connected to the control letter of the motor drive module
Number output end, since the trolley has the function of positive and negative rotation, so needing to build two full bridge driving circuits.Specifically, this reality
The model IR2104 of the driving chip in example in motor drive module is applied, the signal input part IN of the driving chip is connected to
PWM signal generator receives pwm pulse signal, and output end HO and LO are connected to the grid of two NMOS tube IR7843, pass through output
Low and high level controls being turned on and off for NMOS tube IR7843, to control the revolving speed of the traveling motor to control trolley
Movement speed.
Since the NMOS tube IR7843 conducting in full bridge driving circuit needs the grid voltage of 10V or more, but battery is defeated
Enter to be up to 8.2V, so needing to boost using voltage of the MC34063A to input, and using boostrap circuit that grid is electric
Pressure is raised, to reach the voltage conditions of NMOS tube IR7843 conducting.
Specifically, the main control module 10 obtains the turnning circle of the traveling motor by motor drive module, calculates
The straight line scalar distance of the trolley body movement out, and by the straight line scalar distance and azimuth information to the trolley
Ontology is positioned.The main control module 10 is recorded the trolley body and searches the path of object and deposited with stacked data
Storage, and drive the trolley body to return according to the inverse process progress path for searching object by executing its stacked data stored
It traces back.
The clamp device 60 includes being respectively arranged in two steering engines of trolley body front end two sides and being set to described
Two mechanical arms on steering engine；Two mechanical arm extends and gathers to form folder object space to the trolley body direction of advance.By
It needs to manipulate mechanical arm in steering engine object is clamped and discharged, therefore steering engine need to be mounted on the most front side of dolly chassis.It is main
The open and-shut mode that module 10 controls steering engine by pwm pulse signal is controlled, so that the rocker arm and mechanical arm that control steering engine are to object
It is clamped or is discharged.In the present embodiment, the model of steering engine can be MG90S, can also use the steering engine of other models.
Trolley object in transport process is fallen in order to prevent, is also provided with and the master control mould in trolley body front end
The infrared sensor 80 that block 10 is electrically connected.The infrared light that the infrared sensor 80 issues is directed toward the folder object space, to examine
Survey whether mechanical arm described in trolley transport process consolidates the clamping of article.When infrared sensor 80 detects in folder object space
When having article, high level signal just is sent to main control module 10, conversely, then sending low level signal to main control module 20.
The power module 80 provides power supply to modules.Camera, the rudder of image collecting device in the present embodiment
Machine, infrared sensor, electronic compass and main control chip ARM Cortex A53 operating voltage be 5V.
Referring to Fig. 7, it is the LM2940 voltage regulator circuit figure of power module of the present invention.Since the steering of steering engine can produce
Raw larger current, can impact the module of other on circuit board, therefore a LM2940 voltage regulator circuit is separately provided and carries out surely
Power supply is provided to steering engine after pressure, circuit design is as shown in Figure 6.
Referring to Fig. 8, it is the XL4016 voltage regulator circuit figure of power module of the present invention.Due to main control chip ARM
To the more demanding of the stability of supply voltage and electric current and numerical values recited when Cortex A53 works, at least it is required to provide
The dc power power supply of 5V, 2A, therefore individually designed voltage regulator circuit all the way provides power supply to main control chip ARM Cortex A53.
Specifically, the linear stabilized power supply based on XL4016 voltage stabilizing chip has been used, input voltage range is in 1.25V to 36V, output
For current peak up to 8A, circuit design is as shown in Figure 7.
Referring to Fig. 9, it is the software design principle flow chart of intelligence storage cart system of the present invention.
Algorithm and program in the intelligence storage trolley course of work will be run on linux system, and device power is
After system initialization is completed, program will wait instruction mode automatically into circulation, and user is waited to trigger voice control using password
After mode, trolley by etc. the instruction of the specified article of clamping assigned of user to be received.Trolley receives instructions later, will in the venue
The specified article of automatic cruising search, while the routing information moved every time by stack record trolley.Camera identification captures
To after specified article, automatic programme path is gone to object attachment by trolley, and final mechanical arm of opening clamps specified article and returns
Object is restored in specified position, everything is finished, and waits next instruction.
Intelligence storage trolley of the invention receives the instruction of user by voice interaction module, is received and is schemed by main control module
The image as captured by acquisition device identifies object, and the azimuth information of electronic compass is combined to determine article
Position, then control motor drive module driving motor and march at required object, control clamp device clamps required article,
And article is transported to initial position, to realize the identification of article and sort out placement.Intelligence storage trolley of the invention is to existing
There are cleaning well and arrangement effect for the complicated ground environment of family, mitigates the housework workload of kinsfolk, improve life
Quality.
The embodiments described above only express several embodiments of the present invention, and the description thereof is more specific and detailed, but simultaneously
It cannot therefore be construed as limiting the scope of the patent.It should be pointed out that coming for those of ordinary skill in the art
It says, without departing from the inventive concept of the premise, various modifications and improvements can be made, these belong to protection of the invention
Range.
Claims (10)
1. a kind of intelligence storage cart system based on machine learning, it is characterised in that: including trolley body and be equipped on described
Main control module in trolley body, the voice interaction module being electrically connected with the main control module, image collecting device, electronics sieve
Disk, moving device and clamp device；The voice interaction module receives the voice command of user, is identified simultaneously to voice command
Recognition result is sent to the main control module；Described image acquisition device is used to acquire the image information of article, and will be adopted
The image information of collection is sent to the main control module；The electronic compass is used to record the azimuth letter of the trolley body rotation
Breath, and azimuth information is sent to the main control module；The moving device includes for driving the trolley body mobile
Motor drive module；The main control module receives the speech recognition result of the voice interaction module transmission, according to the figure
As the azimuth information that acquisition device acquired image information and the electronic compass are recorded determines that user specifies article
Classification and positioning, and control the motor drive module and the trolley body is driven to be moved to article position, described in control
Article is simultaneously transported to initial position by clamp device clamping article.
2. the intelligence storage cart system according to claim 1 based on machine learning, it is characterised in that: the master control mould
(SuSE) Linux OS is used inside block, and svm classifier model is established in the (SuSE) Linux OS；The main control module pair
Described image acquisition device acquired image carries out HOG feature extraction, and the HOG feature is inputted the svm classifier mould
Type carries out Classification and Identification, to determine the geometric position of the classification and object of object in the picture.
3. the intelligence storage cart system according to claim 2 based on machine learning, it is characterised in that: the SVM points
Class model establishes image recognition tag library, and identification frame is arranged and is scanned to described image acquisition device acquired image,
And judge the label information in the characteristic information of article and described image identification tag library that the identification frame extracts whether one
It causes.
4. the intelligence storage cart system according to claim 1 based on machine learning, it is characterised in that: the traveling dress
Setting further includes the multiple traveling motors for being installed on the trolley body two sides；Multiple traveling motors are connected to the motor and drive
The control signal output of dynamic model block；The main control module obtains the turning collar of the traveling motor by motor drive module
Number, calculates the straight line scalar distance of the trolley body movement, and passes through the straight line scalar distance and azimuth information pair
The trolley body is positioned.
5. the intelligence storage cart system according to claim 1 based on machine learning, it is characterised in that: the master control mould
Block is recorded the trolley body and searches the path of object and stored with stacked data, and by executing its storehouse stored
Trolley body described in data-driven carries out path backtracking according to the inverse process for searching object.
6. the intelligence storage cart system according to claim 1 based on machine learning, it is characterised in that: the voice is handed over
Mutual module includes the voice recognition chip being connected with microphone and is connected to the voice recognition chip and the main control module
Between controller；The voice recognition chip, which identifies the phonetic order of user and sends voice to the controller, to be known
Other result；The controller receives institute's speech recognition result, and described in by serial ports being sent to institute's speech recognition result
Main control module, and control the voice recognition chip and issue reply voice.
7. the intelligence storage cart system according to claim 1 based on machine learning, it is characterised in that: the clamping dress
Two mechanical arms setting two steering engines including being respectively arranged in the trolley body front end and being set on the steering engine；Two machine
Tool arm extends and gathers to form folder object space to the trolley body direction of advance.
8. the intelligence storage cart system according to claim 7 based on machine learning, it is characterised in that: further include setting
Infrared sensor between two steering engine；The infrared light that the infrared sensor issues is directed toward the folder object space.
9. the intelligence storage cart system according to claim 1 based on machine learning, it is characterised in that: electronics sieve
The X-direction of disk is parallel with the middle line of the trolley body, the median perpendicular of Y-direction and the trolley body.
10. it is according to claim 1 based on machine learning intelligence storage cart system, it is characterised in that: further include with
The PWM output module that the main control module is connected；The motor drive module includes multiple full bridge driving circuits；The full-bridge
Driving circuit includes the driving chip connecting with the signal output end of PWM output module and to be connected to the driving chip signal defeated
The metal-oxide-semiconductor of outlet；The output end of the metal-oxide-semiconductor is connected to the traveling motor；The driving chip receives pwm pulse signal simultaneously
Control the on and off of the metal-oxide-semiconductor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811404423.4A CN109709951A (en) | 2018-11-23 | 2018-11-23 | A kind of intelligence storage cart system based on machine learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811404423.4A CN109709951A (en) | 2018-11-23 | 2018-11-23 | A kind of intelligence storage cart system based on machine learning |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109709951A true CN109709951A (en) | 2019-05-03 |
Family
ID=66254515
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811404423.4A Pending CN109709951A (en) | 2018-11-23 | 2018-11-23 | A kind of intelligence storage cart system based on machine learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109709951A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110586489A (en) * | 2019-08-06 | 2019-12-20 | 广东工业大学 | Intelligent vehicle for classifying articles based on image recognition and control method thereof |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101976075A (en) * | 2010-11-24 | 2011-02-16 | 重庆大学 | Intelligent self-tracking trolley control system based on voice control |
| CN103529740A (en) * | 2013-10-31 | 2014-01-22 | 天津职业技术师范大学 | Intelligent domestic robot |
| CN105773619A (en) * | 2016-04-26 | 2016-07-20 | 北京光年无限科技有限公司 | Electronic control system used for realizing grabbing behavior of humanoid robot and humanoid robot |
| CN105823552A (en) * | 2016-05-06 | 2016-08-03 | 简燕梅 | Intelligent detection robot car for architectural structure vibration |
| CN107253203A (en) * | 2017-06-29 | 2017-10-17 | 上海理工大学 | Multimode Intelligent toy collects robot |
| US20180321672A1 (en) * | 2016-05-09 | 2018-11-08 | StrongForce IoT Portfolio 2016, LLC | Methods and systems for a data marketplace in an industrial internet of things environment |
-
2018
- 2018-11-23 CN CN201811404423.4A patent/CN109709951A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101976075A (en) * | 2010-11-24 | 2011-02-16 | 重庆大学 | Intelligent self-tracking trolley control system based on voice control |
| CN103529740A (en) * | 2013-10-31 | 2014-01-22 | 天津职业技术师范大学 | Intelligent domestic robot |
| CN105773619A (en) * | 2016-04-26 | 2016-07-20 | 北京光年无限科技有限公司 | Electronic control system used for realizing grabbing behavior of humanoid robot and humanoid robot |
| CN105823552A (en) * | 2016-05-06 | 2016-08-03 | 简燕梅 | Intelligent detection robot car for architectural structure vibration |
| US20180321672A1 (en) * | 2016-05-09 | 2018-11-08 | StrongForce IoT Portfolio 2016, LLC | Methods and systems for a data marketplace in an industrial internet of things environment |
| CN107253203A (en) * | 2017-06-29 | 2017-10-17 | 上海理工大学 | Multimode Intelligent toy collects robot |
Non-Patent Citations (2)
| Title |
|---|
| 劳健涛，等: "智能车竞赛中图像畸变还原的实现", 《物理实验》 * |
| 唐小煜，等: "追寻信标的智能车最优导航策略", 《物理实验》 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110586489A (en) * | 2019-08-06 | 2019-12-20 | 广东工业大学 | Intelligent vehicle for classifying articles based on image recognition and control method thereof |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107139179B (en) | Intelligent service robot and working method | |
| CN105137887B (en) | The material objectization programmed method of plate and its application in robot field based on programming | |
| Wu et al. | View-independent recognition of hand postures | |
| Kumar et al. | Human activity recognition using optical flow based feature set | |
| CN104284065A (en) | Industrial intelligent camera based on FPGA | |
| CN102103409A (en) | Man-machine interaction method and device based on motion trail identification | |
| CN106164562A (en) | Promote that location and displacement calculate the bracket of equipment | |
| CN107123423A (en) | Voice pick device and multimedia equipment | |
| CN109543732B (en) | Assembling system and method based on class feature knowledge base | |
| Mahjoub et al. | Human action recognition using RGB data | |
| CN114063609A (en) | Obstacle avoidance control method and control system of intelligent mowing robot | |
| CN105913034B (en) | Vehicle identification method and device and vehicle | |
| CN109709951A (en) | A kind of intelligence storage cart system based on machine learning | |
| Feng et al. | Garbage disposal of complex background based on deep learning with limited hardware resources | |
| CN105082141A (en) | Execution modules, magic cube resolving robot and use method for magic cube resolving robot | |
| Yi et al. | Long-range hand gesture recognition with joint ssd network | |
| CN103473569A (en) | Medical image classification method based on SVM | |
| Bekel et al. | Adaptive computer vision: Online learning for object recognition | |
| CN109584864A (en) | Image processing apparatus and method | |
| JP2008257353A (en) | Learning system for learning visual representation of object, and computer program | |
| WO2022027865A1 (en) | Double-vision feedback motion control system based on unmanned walking vehicle-washing device | |
| CN204131605U (en) | A kind of industrial intelligent camera based on FPGA | |
| CN107784306A (en) | A kind of automatic shopping car based on OpenCV | |
| CN116098536B (en) | Robot control method and device | |
| CN213621558U (en) | Intelligent garbage classification device |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190503 |