CN109272108A - Control method for movement, system and computer equipment based on neural network algorithm - Google Patents
Control method for movement, system and computer equipment based on neural network algorithm Download PDFInfo
- Publication number
- CN109272108A CN109272108A CN201810961584.7A CN201810961584A CN109272108A CN 109272108 A CN109272108 A CN 109272108A CN 201810961584 A CN201810961584 A CN 201810961584A CN 109272108 A CN109272108 A CN 109272108A
- Authority
- CN
- China
- Prior art keywords
- neural network
- module
- multiple ambient
- network model
- database
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course or altitude of land, water, air, or space vehicles, e.g. automatic pilot
- G05D1/02—Control of position or course in two dimensions
- G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
- G05D1/0212—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory
- G05D1/0221—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory involving a learning process
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course or altitude of land, water, air, or space vehicles, e.g. automatic pilot
- G05D1/02—Control of position or course in two dimensions
- G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
- G05D1/0231—Control of position or course in two dimensions specially adapted to land vehicles using optical position detecting means
- G05D1/0246—Control of position or course in two dimensions specially adapted to land vehicles using optical position detecting means using a video camera in combination with image processing means
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
- G10L13/00—Speech synthesis; Text to speech systems
Abstract
This application involves a kind of control method for movement, system, computer equipment and storage mediums neural network based, wherein multiple ambient images that this method passes through acquisition current environment, multiple ambient images are pre-processed to obtain treated multiple ambient images, by treated, multiple ambient images are input in trained neural network model, neural network model predicts motion track, motion track information is generated according to the prediction probability of neural network output, mobile control is carried out according to motion track information.This programme is by using neural network algorithm, so that movable termination has the ability of self-teaching, and can realize the operation of automation track in complex environment, improves the stability and reliability of movable termination.
Description
Technical field
The present invention relates to artificial intelligence field, more particularly to a kind of control method for movement based on neural network algorithm,
System, computer equipment and storage medium.
Background technique
Currently, the technology with artificial intelligence develops, increasingly carry out the daily life that more artificial intelligence equipment enters user
In work, intelligent carriage be exactly it is one such, intelligent carriage can realize automatic Pilot according to the environment that is presently in, have very
Strong practicability and market prospects.
In the conventional technology, intelligent carriage can be roughly divided into two classes according to the type of sensor, and one is be based on photoelectricity
Class sensor, another kind is based on non-smooth electrical sensor.Automatic Pilot intelligent carriage based on light electrical sensor is big
Part uses infrared sensor, such sensor is analyzed and determined using the track that the principle of reflection of light travels intelligent carriage
Plan, another part use imaging sensor, and such as linear CCD, face battle array CMOS camera, laser etc., such sensor utilizes image
The method of processing carries out analysis and decision to the track of intelligent carriage automatic Pilot, compare and the former, the complexity of latter system
Degree is higher, but the integrated level of system and performance are more superior.The input of unmanned trolley based on non-smooth electrical sensor is believed
It number is no longer just optical information, such as electromagnetic signal, GPS signal etc..Using the intelligent carriage of electromagnetic signal, need in traveling
Electromagnetic information is added on track, intelligent car systems carry out analysis and decision using electromagnetic information of the electromagnetic sensor to acquisition.
Unmanned intelligent carriage based on GPS realizes unmanned, intelligence between two o'clock by inputting different GPS informations
Trolley control system must use under the scene for having GPS signal.Therefore, traditional unmanned intelligent carriage is learned without self
The ability of habit is merely able to run under simple and specific scene, and environmental suitability is poor.
Summary of the invention
Based on this, it is necessary to which in view of the above technical problems, automatic Pilot can be realized in complex scene by providing one kind
Control method for movement, system, computer equipment and storage medium.
A kind of control method for movement neural network based, which comprises
Obtain multiple ambient images of current environment；
The multiple ambient image is pre-processed to obtain treated multiple ambient images；
By treated, multiple ambient images are input in trained neural network model, the neural network model pair
Motion track is predicted；
Motion track information is generated according to the prediction probability of neural network output；
Mobile control is carried out according to the motion track information.
In one of the embodiments, will treated that multiple ambient images are input to trained neural network model
In, before the step of neural network model predicts motion track further include:
Establish neural network model；
Establish track database and neural network test database；
The neural network model is trained to obtain according to the track database and neural network test database
Trained neural network model.
It is described in one of the embodiments, that the multiple ambient image is pre-processed to obtain treated multiple rings
Border image includes:
The multiple ambient image is filtered smoothly；
The multiple ambient image is subjected to color conversion and space multiple ambient images that are converted to that treated.
The neural network model includes: input layer, hidden layer and output layer in one of the embodiments,；
The quantity of input layer is set according to the size of input environment image；
According to the quantity of current application scenarios setting hidden layer；
According to the quantity of the type set output layer of track.
A kind of mobile control system neural network based, the system comprises:
Obtain module, the multiple ambient images for obtaining module and being used to obtain current environment；
Preprocessing module, the preprocessing module are used to be pre-processed to obtain by the multiple ambient image that treated
Multiple ambient images；
Prediction module, the prediction module is for by treated, multiple ambient images to be input to trained neural network
In model, the neural network model predicts motion track；
Track generation module, the prediction probability that the track generation module is used to be exported according to neural network generate moving rail
Mark information；
Mobile control module, the mobile control module are used to carry out mobile control according to the motion track information.
In one of the embodiments, the system also includes:
Model building module, the model building module is for establishing neural network model；
Database module, the Database module is for establishing track database and neural network test data
Library；
Training module, the training module are used for according to the track database and neural network volumes database to described
Neural network model is trained to obtain trained neural network model.
In one of the embodiments, the system also includes:
Voice synthetic module, the voice synthetic module are used to obtain to be generated according to the prediction probability of neural network output and move
Dynamic trace information, and the motion track information is subjected to speech synthesis by online speech synthesis cloud platform and obtains corresponding language
Sound file.
In one of the embodiments, the system also includes:
Voice playing module, the voice playing module, will be described corresponding for obtaining the corresponding voice document
Voice document is played out by audio transcoder, power amplification circuit and voice playing device.
A kind of computer equipment can be run on a memory and on a processor including memory, processor and storage
The step of computer program, the processor realizes above-mentioned any one method when executing the computer program.
A kind of computer readable storage medium, is stored thereon with computer program, and the computer program is held by processor
The step of above-mentioned any one method is realized when row.
Above-mentioned control method for movement neural network based, system, computer equipment and storage medium, it is current by obtaining
Multiple ambient images are pre-processed to obtain treated multiple ambient images, after processing by multiple ambient images of environment
Multiple ambient images be input in trained neural network model, neural network model predicts motion track, root
Motion track information is generated according to the prediction probability of neural network output, mobile control is carried out according to motion track information.This programme
By using neural network algorithm, so that movable termination has the ability of self-teaching, and can be in complex environment
It realizes the operation of automation track, improves the stability and reliability of movable termination.
Detailed description of the invention
Fig. 1 is the application scenario diagram of control method for movement neural network based in one embodiment；
Fig. 2 is the flow diagram of control method for movement neural network based in one embodiment；
Fig. 3 is the flow diagram of the step of neural network model training in one embodiment；
Fig. 4 is the process that multiple ambient images are pre-processed to the step of obtaining ambient image information in one embodiment
Schematic diagram；
Fig. 5 is flow diagram the step of establishing neural network model in one embodiment；
Fig. 6 is the structural block diagram of mobile control system neural network based in one embodiment；
Fig. 7 is the structural block diagram of mobile control system neural network based in another embodiment；
Fig. 8 is the internal structure chart of computer equipment in one embodiment.
Specific embodiment
It is with reference to the accompanying drawings and embodiments, right in order to which the objects, technical solutions and advantages of the application are more clearly understood
The application is further elaborated.It should be appreciated that specific embodiment described herein is only used to explain the application, and
It is not used in restriction the application.
Control method for movement neural network based provided by the embodiment of the present invention may be used on application as shown in Figure 1
In environment.Movable termination 110 is equipped with image capture device 120, which includes: moveable trolley, can
Any of them such as mobile sweeper, the image capture device 120 include: camera, camera, video camera and capable of taking pictures
Any of them such as mobile device.The image capture device 120 can acquire in environment locating for movable termination 110
Multiple ambient images, it is to be appreciated that image capture device 120 can acquire more ambient images so as to obtain more comprehensively
Environmental information.Collected multiple ambient images are sent to movable termination 110 by image capture device 120, are moved eventually
End 110 obtains multiple ambient images, and multiple ambient images are pre-processed to obtain treated multiple rings by movable termination 110
Border image, by treated, multiple ambient images are input in trained neural network model movable termination 110, nerve net
Network model predicts the motion track of movable termination 110, is generated according to the prediction probability of neural network output removable
The motion track information of terminal 110 carries out mobile control accordingly to movable termination 110 according to motion track information.
In one embodiment, as shown in Fig. 2, a kind of control method for movement neural network based is provided, with the party
It is illustrated in the mobile control system neural network based that method is applied in Fig. 1, this method comprises:
Step 202, multiple ambient images of current environment are obtained.
Movable termination uses scene by using environment information acquisition sensor, such as camera, to movable termination
Under running rail carry out Image Acquisition.Such as: the size for acquiring image is 64*48, i.e. the length of image is 64, and width is
48, the format of image is gray level image.Movable termination gets multiple ambient images that environment information acquisition sensing acquisition arrives.
It is understood that carrying out Image Acquisition, the sample of acquisition to running rail to movable termination using sensors such as cameras
This number is the more the better, therefore, can filter out the preferred of arrival standard with multi collect ambient image, and according to preset standard
Ambient image.
Step 204, multiple ambient images are pre-processed to obtain treated multiple ambient images.
Movable termination carries out image preprocessing, including filtering, color space conversion etc. to the gray level image of acquisition.
Specifically, Image Smoothing Skill is details, mutation, edge and the noise in compacting, reduction or elimination image, is exactly image smoothing
Change.Image smoothing is to make low-pass filtering to image, can be realized in spatial domain or frequency domain.
Step 206, by treated, multiple ambient images are input in trained neural network model, neural network mould
Type predicts motion track.
Step 208, motion track information is generated according to the prediction probability of neural network output.
Pretreated image is input in neural network model by movable termination, and neural network is to current driving track
Type is predicted that the prediction probability exported by neural network generates current motion track information.It is wrapped in motion track information
A variety of running rail types are included, such as: straight way, left bend, right bend, zebra stripes etc., it is to be appreciated that the running rail class
Type is not limited to these types.
Step 210, mobile control is carried out according to motion track information.
The motion track information that movable termination is exported according to neural network realizes control to the movement of movable termination.
Specifically, movable termination can be carried out accordingly by motion-control module according to the motion track information that neural network exports
Motion state control.
In the present embodiment, by obtaining multiple ambient images of current environment, multiple ambient images are pre-processed
Multiple ambient images that obtain that treated, will treated that multiple ambient images are input in trained neural network model,
Neural network model predicts motion track, generates motion track information, root according to the prediction probability of neural network output
Mobile control is carried out according to motion track information.This programme is by using neural network algorithm, so that movable termination has certainly
The ability that I learns, and the operation of automation track can be realized in complex environment, improve the stability of movable termination
And reliability.
In one embodiment, a kind of control method for movement neural network based is provided, as shown in figure 3, this method
Will treated that multiple ambient images are input in trained neural network model, neural network model to motion track into
Include the steps that neural network model training before the step of row prediction:
Step 302, neural network model is established；
Step 304, track database and neural network test database are established；
Step 306, the neural network model is trained according to track database and neural network test database
Obtain trained neural network model.
Specifically, movable termination establishes neural network model, including input layer, hidden layer and output layer, learning rate.It is defeated
The quantity for entering layer is determined that the quantity of hidden layer is determined by actual experiment scene, the quantity of output layer by the size of input picture
It is determined by the type of running rail.The database and neural network test database of track are established in the foundation of tranining database.Benefit
The training of neural network is carried out with database.It is understood that can use neural network test database, evaluate trained
Neural network model indicates the neural network model trained completion when the result of evaluation reaches preset standard.
In the present embodiment, by according to track database and neural network test database to the neural network model
It is trained to obtain trained neural network model, realizes the training to neural network, so that trained neural network
The reliability of the application environment being more suitable in the present invention, the trace information finally obtained is higher.
In one embodiment, a kind of control method for movement neural network based is provided, as shown in figure 4, this method
Middle the step of being pre-processed to obtain treated multiple ambient images for multiple ambient images includes:
Step 402, multiple ambient images are filtered smoothly；
Step 404, multiple ambient images are subjected to color conversion and space multiple environment maps that are converted to that treated
Picture.
Specifically, multiple ambient images are filtered smoothly by movable termination first, and then movable termination will be multiple
Ambient image carries out color conversion and space multiple ambient images that are converted to that treated.It is understood that the image
Pretreated process can realize by running corresponding image preprocessing program in the processor of movable termination, can also be with
Cloud Server is sent to by movable termination to realize the pretreatment of image, after the processing for then receiving Cloud Server return again
Multiple ambient images.
In the present embodiment, panorama sketch is obtained by carrying out projection splicing to multiple original images in the same coordinate system
Picture, the calibration effect of calibrating parameters is judged by judging the registration of the panoramic picture, and then has reached raising calibration result
Reliability.
In one embodiment, a kind of control method for movement neural network based, the nerve net in this method are provided
Network model includes: input layer, hidden layer and output layer, as shown in figure 5, the step of establishing neural network model includes:
Step 502, the quantity of input layer is set according to the size of input environment image；
Step 504, according to the quantity of current application scenarios setting hidden layer；
Step 506, according to the quantity of the type set output layer of track.
In the present embodiment, it by setting the quantity of input layer according to the size of input environment image, is answered according to current
With the quantity of scene settings hidden layer, according to the quantity of the type set output layer of track, so that the neural network model established
It is more in line with this programme design scheme, the reliability of the trace information finally obtained is higher.
It should be understood that although each step in the flow chart of Fig. 2-5 is successively shown according to the instruction of arrow,
These steps are not that the inevitable sequence according to arrow instruction successively executes.Unless expressly stating otherwise herein, these steps
Execution there is no stringent sequences to limit, these steps can execute in other order.Moreover, at least one in Fig. 2-5
Part steps may include that perhaps these sub-steps of multiple stages or stage are not necessarily in synchronization to multiple sub-steps
Completion is executed, but can be executed at different times, the execution sequence in these sub-steps or stage is also not necessarily successively
It carries out, but can be at least part of the sub-step or stage of other steps or other steps in turn or alternately
It executes.
In one embodiment, as shown in fig. 6, providing a kind of mobile control system neural network based, the system
Include:
Module 602 is obtained, multiple ambient images that module is used to obtain current environment are obtained；
Preprocessing module 604, preprocessing module are used to be pre-processed to obtain by multiple ambient images that treated to be multiple
Ambient image；
Prediction module 606, prediction module is for by treated, multiple ambient images to be input to trained neural network
In model, neural network model predicts motion track；
Track generation module 608, the prediction probability that track generation module is used to be exported according to neural network generate moving rail
Mark information；
Mobile control module 610, mobile control module are used to carry out mobile control according to motion track information.
In one embodiment, a kind of mobile control system neural network based, the system are provided further include:
Model building module, model building module is for establishing neural network model；
Database module, Database module is for establishing track database and neural network test database；
Training module, training module are used for according to track database and neural network volumes database to neural network model
It is trained to obtain trained neural network model.
In one embodiment, as shown in fig. 7, providing a kind of mobile control system neural network based, the system
Further include:
Voice synthetic module 612, voice synthetic module are used to obtain to be generated according to the prediction probability of neural network output and move
Dynamic trace information, and motion track information is subjected to speech synthesis by online speech synthesis cloud platform and obtains corresponding voice text
Part.
Voice playing module 614, voice playing module lead to corresponding voice document for obtaining corresponding voice document
Audio transcoder, power amplification circuit and voice playing device is crossed to play out.
Specifically, the content of neural network prediction output is input to speech synthesis unit.It is flat using online speech synthesis cloud
Platform completes the speech synthesis of current input content, generates mp3 file, and the program needs to connect internet.Using existing in the market
Chipspeech complete the speech synthesis of current input content, generate mp3 file.Such as on raspberry pie platform, using
Line speech synthesis cloud platform (Baidu's cloud, Iflytek etc.) realizes speech synthesis., the audio file after speech synthesis is mp3 text
Part connects loudspeaker with power amplification circuit using audio transcoder or speaker realizes voice broadcast.Such as on raspberry pie platform,
Voice broadcast is realized using mplayer audio plug and power amplifier driving loudspeaker.
It should be understood that the mobile control system neural network based further include: power management module, including lithium battery
And electric power management circuit, manage the electric energy supply of entire intelligent carriage.Motion-control module: including signal isolation circuit, motor
Driving circuit exports the movement of pulse control motor.Above-mentioned module is all the supplemental functionality of the system.
Specific restriction about mobile control system neural network based may refer to above for based on nerve net
The restriction of the control method for movement of network, details are not described herein.
In one embodiment, a kind of computer equipment is provided, internal structure chart can be as shown in Figure 8.The calculating
Machine equipment includes processor, memory and the network interface connected by system bus.Wherein, the processor of the computer equipment
For providing calculating and control ability.The memory of the computer equipment includes non-volatile memory medium, built-in storage.This is non-
Volatile storage medium is stored with operating system, computer program and database.The built-in storage is non-volatile memory medium
In operating system and computer program operation provide environment.The network interface of the computer equipment is used for and external terminal
It is communicated by network connection.To realize a kind of mobile controlling party neural network based when the computer program is executed by processor
Method.
It will be understood by those skilled in the art that structure shown in Fig. 8, only part relevant to application scheme is tied
The block diagram of structure does not constitute the restriction for the computer equipment being applied thereon to application scheme, specific computer equipment
It may include perhaps combining certain components or with different component layouts than more or fewer components as shown in the figure.
In one embodiment, a kind of computer equipment is provided, including memory, processor and storage are on a memory
And the computer program that can be run on a processor, processor are realized when executing computer program in above each embodiment of the method
The step of.
In one embodiment, a kind of computer readable storage medium is provided, computer program is stored thereon with, is calculated
The step in above each embodiment of the method is realized when machine program is executed by processor.
Those of ordinary skill in the art will appreciate that realizing all or part of the process in above-described embodiment method, being can be with
Relevant hardware is instructed to complete by computer program, the computer program can be stored in a non-volatile computer
In read/write memory medium, the computer program is when being executed, it may include such as the process of the embodiment of above-mentioned each method.Wherein,
To any reference of memory, storage, database or other media used in each embodiment provided herein,
Including non-volatile and/or volatile memory.Nonvolatile memory may include read-only memory (ROM), programming ROM
(PROM), electrically programmable ROM (EPROM), electrically erasable ROM (EEPROM) or flash memory.Volatile memory may include
Random access memory (RAM) or external cache.By way of illustration and not limitation, RAM is available in many forms,
Such as static state RAM (SRAM), dynamic ram (DRAM), synchronous dram (SDRAM), double data rate sdram (DDRSDRAM), enhancing
Type SDRAM (ESDRAM), synchronization link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM
(RDRAM), direct memory bus dynamic ram (DRDRAM) and memory bus dynamic ram (RDRAM) etc..
Each technical characteristic of above embodiments can be combined arbitrarily, for simplicity of description, not to above-described embodiment
In each technical characteristic it is all possible combination be all described, as long as however, the combination of these technical characteristics be not present lance
Shield all should be considered as described in this specification.
The several embodiments of the application above described embodiment only expresses, 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 concept of this application, various modifications and improvements can be made, these belong to the protection of the application
Range.Therefore, the scope of protection shall be subject to the appended claims for the application patent.
Claims (10)
1. a kind of control method for movement neural network based, which comprises
Obtain multiple ambient images of current environment；
The multiple ambient image is pre-processed to obtain treated multiple ambient images；
By treated, multiple ambient images are input in trained neural network model, and the neural network model is to movement
It is predicted track；
Motion track information is generated according to the prediction probability of neural network output；
Mobile control is carried out according to the motion track information.
2. control method for movement neural network based according to claim 1, which is characterized in that will that treated is more
A ambient image is input in trained neural network model, the step that the neural network model predicts motion track
Before rapid further include:
Establish neural network model；
Establish track database and neural network test database；
The neural network model is trained according to the track database and neural network test database and is trained
Good neural network model.
3. control method for movement neural network based according to claim 1, which is characterized in that it is described will be the multiple
Ambient image is pre-processed to obtain, and treated that multiple ambient images include:
The multiple ambient image is filtered smoothly；
The multiple ambient image is subjected to color conversion and space multiple ambient images that are converted to that treated.
4. control method for movement neural network based according to claim 2, which is characterized in that the neural network mould
Type includes: input layer, hidden layer and output layer；
The quantity of input layer is set according to the size of input environment image；
According to the quantity of current application scenarios setting hidden layer；
According to the quantity of the type set output layer of track.
5. a kind of mobile control system neural network based, which is characterized in that the system comprises:
Obtain module, the multiple ambient images for obtaining module and being used to obtain current environment；
Preprocessing module, the preprocessing module are used to be pre-processed to obtain by the multiple ambient image that treated to be multiple
Ambient image；
Prediction module, the prediction module is for by treated, multiple ambient images to be input to trained neural network model
In, the neural network model predicts motion track；
Track generation module, the prediction probability that the track generation module is used to be exported according to neural network generate motion track letter
Breath；
Mobile control module, the mobile control module are used to carry out mobile control according to the motion track information.
6. mobile control system neural network based according to claim 5, which is characterized in that the system also includes:
Model building module, the model building module is for establishing neural network model；
Database module, the Database module is for establishing track database and neural network test database；
Training module, the training module are used for according to the track database and neural network volumes database to the nerve
Network model is trained to obtain trained neural network model.
7. mobile control system neural network based according to claim 5, which is characterized in that the system also includes:
Voice synthetic module, the voice synthetic module, which is used to obtain, generates moving rail according to the prediction probability of neural network output
Mark information, and the motion track information is subjected to speech synthesis by online speech synthesis cloud platform and obtains corresponding voice text
Part.
8. mobile control system neural network based according to claim 7, which is characterized in that the system also includes:
Voice playing module, the voice playing module is for obtaining the corresponding voice document, by the corresponding voice
File is played out by audio transcoder, power amplification circuit and voice playing device.
9. a kind of computer equipment including memory, processor and stores the meter that can be run on a memory and on a processor
Calculation machine program, which is characterized in that the processor realizes any one of claims 1 to 4 institute when executing the computer program
The step of stating method.
10. a kind of computer readable storage medium, is stored thereon with computer program, which is characterized in that the computer program
The step of method described in any one of Claims 1-4 is realized when being executed by processor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810961584.7A CN109272108A (en) | 2018-08-22 | 2018-08-22 | Control method for movement, system and computer equipment based on neural network algorithm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810961584.7A CN109272108A (en) | 2018-08-22 | 2018-08-22 | Control method for movement, system and computer equipment based on neural network algorithm |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109272108A true CN109272108A (en) | 2019-01-25 |
Family
ID=65154256
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810961584.7A Pending CN109272108A (en) | 2018-08-22 | 2018-08-22 | Control method for movement, system and computer equipment based on neural network algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109272108A (en) |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109857118A (en) * | 2019-03-12 | 2019-06-07 | 百度在线网络技术（北京）有限公司 | For planning the method, apparatus, equipment and storage medium of driving strategy |
| CN110275531A (en) * | 2019-06-21 | 2019-09-24 | 北京三快在线科技有限公司 | The trajectory predictions method, apparatus and unmanned equipment of barrier |
| CN110456644A (en) * | 2019-08-13 | 2019-11-15 | 北京地平线机器人技术研发有限公司 | Determine the method, apparatus and electronic equipment of the execution action message of automation equipment |
| CN111523643A (en) * | 2020-04-10 | 2020-08-11 | 商汤集团有限公司 | Trajectory prediction method, apparatus, device and storage medium |
| WO2020164089A1 (en) * | 2019-02-15 | 2020-08-20 | Bayerische Motoren Werke Aktiengesellschaft | Trajectory prediction using deep learning multiple predictor fusion and bayesian optimization |
| CN111562977A (en) * | 2019-02-14 | 2020-08-21 | 上海寒武纪信息科技有限公司 | Neural network model splitting method, device, storage medium and computer system |
| WO2020177417A1 (en) * | 2019-03-01 | 2020-09-10 | 北京三快在线科技有限公司 | Unmanned device control and model training |
| CN112241951A (en) * | 2020-10-19 | 2021-01-19 | 深圳大学 | Agricultural monitoring method, system and computer equipment based on raspberry pi and LORA |
| CN112783151A (en) * | 2019-11-07 | 2021-05-11 | 动态Ad有限责任公司 | Trajectory prediction from a pre-computed or dynamically generated trajectory library |
| CN113031600A (en) * | 2021-03-02 | 2021-06-25 | 京东鲲鹏(江苏)科技有限公司 | Track generation method and device, storage medium and electronic equipment |
| CN113052321A (en) * | 2019-12-26 | 2021-06-29 | 伟摩有限责任公司 | Generating trajectory markers from short-term intent and long-term results |
| CN114047764A (en) * | 2021-11-16 | 2022-02-15 | 北京百度网讯科技有限公司 | Training method of path planning model and path planning method and device |
| CN114080634A (en) * | 2019-07-03 | 2022-02-22 | 伟摩有限责任公司 | Proxy trajectory prediction using anchor trajectories |
| CN114080634B (en) * | 2019-07-03 | 2024-04-26 | 伟摩有限责任公司 | Proxy trajectory prediction using anchor trajectories |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101409072A (en) * | 2007-10-10 | 2009-04-15 | 松下电器产业株式会社 | Embedded equipment, bimodule voice synthesis system and method |
| US20170326726A1 (en) * | 2014-10-02 | 2017-11-16 | Brain Corporation | Apparatus and methods for training path navigation by robots |
| CN108431549A (en) * | 2016-01-05 | 2018-08-21 | 御眼视觉技术有限公司 | The housebroken system of constraint with application |
-
2018
- 2018-08-22 CN CN201810961584.7A patent/CN109272108A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101409072A (en) * | 2007-10-10 | 2009-04-15 | 松下电器产业株式会社 | Embedded equipment, bimodule voice synthesis system and method |
| US20170326726A1 (en) * | 2014-10-02 | 2017-11-16 | Brain Corporation | Apparatus and methods for training path navigation by robots |
| CN108431549A (en) * | 2016-01-05 | 2018-08-21 | 御眼视觉技术有限公司 | The housebroken system of constraint with application |
Cited By (24)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111562977B (en) * | 2019-02-14 | 2022-12-09 | 上海寒武纪信息科技有限公司 | Neural network model splitting method, device, storage medium and computer system |
| CN111562977A (en) * | 2019-02-14 | 2020-08-21 | 上海寒武纪信息科技有限公司 | Neural network model splitting method, device, storage medium and computer system |
| CN113424209B (en) * | 2019-02-15 | 2023-12-22 | 宝马股份公司 | Trajectory prediction using deep learning multi-predictor fusion and Bayesian optimization |
| WO2020164089A1 (en) * | 2019-02-15 | 2020-08-20 | Bayerische Motoren Werke Aktiengesellschaft | Trajectory prediction using deep learning multiple predictor fusion and bayesian optimization |
| CN113424209A (en) * | 2019-02-15 | 2021-09-21 | 宝马股份公司 | Trajectory prediction using deep learning multi-predictor fusion and bayesian optimization |
| WO2020177417A1 (en) * | 2019-03-01 | 2020-09-10 | 北京三快在线科技有限公司 | Unmanned device control and model training |
| CN109857118A (en) * | 2019-03-12 | 2019-06-07 | 百度在线网络技术（北京）有限公司 | For planning the method, apparatus, equipment and storage medium of driving strategy |
| CN109857118B (en) * | 2019-03-12 | 2022-08-16 | 百度在线网络技术（北京）有限公司 | Method, device, equipment and storage medium for planning driving strategy of unmanned vehicle |
| CN110275531A (en) * | 2019-06-21 | 2019-09-24 | 北京三快在线科技有限公司 | The trajectory predictions method, apparatus and unmanned equipment of barrier |
| CN114080634B (en) * | 2019-07-03 | 2024-04-26 | 伟摩有限责任公司 | Proxy trajectory prediction using anchor trajectories |
| CN114080634A (en) * | 2019-07-03 | 2022-02-22 | 伟摩有限责任公司 | Proxy trajectory prediction using anchor trajectories |
| CN110456644B (en) * | 2019-08-13 | 2022-12-06 | 北京地平线机器人技术研发有限公司 | Method and device for determining execution action information of automation equipment and electronic equipment |
| CN110456644A (en) * | 2019-08-13 | 2019-11-15 | 北京地平线机器人技术研发有限公司 | Determine the method, apparatus and electronic equipment of the execution action message of automation equipment |
| CN112783151A (en) * | 2019-11-07 | 2021-05-11 | 动态Ad有限责任公司 | Trajectory prediction from a pre-computed or dynamically generated trajectory library |
| US11912271B2 (en) | 2019-11-07 | 2024-02-27 | Motional Ad Llc | Trajectory prediction from precomputed or dynamically generated bank of trajectories |
| US11900224B2 (en) | 2019-12-26 | 2024-02-13 | Waymo Llc | Generating trajectory labels from short-term intention and long-term result |
| CN113052321A (en) * | 2019-12-26 | 2021-06-29 | 伟摩有限责任公司 | Generating trajectory markers from short-term intent and long-term results |
| CN111523643A (en) * | 2020-04-10 | 2020-08-11 | 商汤集团有限公司 | Trajectory prediction method, apparatus, device and storage medium |
| CN111523643B (en) * | 2020-04-10 | 2024-01-05 | 商汤集团有限公司 | Track prediction method, device, equipment and storage medium |
| CN112241951A (en) * | 2020-10-19 | 2021-01-19 | 深圳大学 | Agricultural monitoring method, system and computer equipment based on raspberry pi and LORA |
| CN113031600B (en) * | 2021-03-02 | 2023-09-05 | 京东鲲鹏(江苏)科技有限公司 | Track generation method and device, storage medium and electronic equipment |
| CN113031600A (en) * | 2021-03-02 | 2021-06-25 | 京东鲲鹏(江苏)科技有限公司 | Track generation method and device, storage medium and electronic equipment |
| CN114047764A (en) * | 2021-11-16 | 2022-02-15 | 北京百度网讯科技有限公司 | Training method of path planning model and path planning method and device |
| CN114047764B (en) * | 2021-11-16 | 2023-11-07 | 北京百度网讯科技有限公司 | Training method of path planning model, path planning method and device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109272108A (en) | Control method for movement, system and computer equipment based on neural network algorithm | |
| CN111914692B (en) | Method and device for acquiring damage assessment image of vehicle | |
| CN111160172B (en) | Parking space detection method, device, computer equipment and storage medium | |
| CN106204435A (en) | Image processing method and device | |
| CN106446782A (en) | Image identification method and device | |
| CN112532897A (en) | Video clipping method, device, equipment and computer readable storage medium | |
| CN110465947A (en) | Multi-modal fusion man-machine interaction method, device, storage medium, terminal and system | |
| US10970909B2 (en) | Method and apparatus for eye movement synthesis | |
| CN111079507B (en) | Behavior recognition method and device, computer device and readable storage medium | |
| US11367196B2 (en) | Image processing method, apparatus, and storage medium | |
| CN115699082A (en) | Defect detection method and device, storage medium and electronic equipment | |
| CN110348418A (en) | Method for tracking target and device, Intelligent mobile equipment and storage medium | |
| CN107392189A (en) | For the method and apparatus for the driving behavior for determining unmanned vehicle | |
| CN111695421A (en) | Image recognition method and device and electronic equipment | |
| CN114641800A (en) | Method and system for forecasting crowd dynamics | |
| CN112000226A (en) | Human eye sight estimation method, device and sight estimation system | |
| CN112115894A (en) | Training method and device for hand key point detection model and electronic equipment | |
| CN109697389B (en) | Identity recognition method and device | |
| CN112906609A (en) | Video important area prediction method and device based on two-way cross attention network | |
| CN114782596A (en) | Voice-driven human face animation generation method, device, equipment and storage medium | |
| WO2023217138A1 (en) | Parameter configuration method and apparatus, device, storage medium and product | |
| CN112306601A (en) | Application interaction method and device, electronic equipment and storage medium | |
| CN110321829A (en) | A kind of face identification method and device, electronic equipment and storage medium | |
| CN110490950B (en) | Image sample generation method and device, computer equipment and storage medium | |
| CN111563465A (en) | Animal behavioristics automatic analysis system |
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 | ||
| CB02 | Change of applicant information |
Address after: 518000 East, 2nd floor, building B, minlida industrial building, zone 4, honghualing Industrial Zone, 1189 Liuxian Avenue, Pingshan community, Taoyuan Street, Nanshan District, Shenzhen City, Guangdong Province Applicant after: SHENZHEN YAHBOOM TECHNOLOGY Co.,Ltd. Address before: 518000 room 306, building C, Guanlong complex building, Guanlong Road, Xili street, Nanshan District, Shenzhen City, Guangdong Province Applicant before: SHENZHEN YAHBOOM TECHNOLOGY Co.,Ltd. |
|
| CB02 | Change of applicant information | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190125 |
|
| RJ01 | Rejection of invention patent application after publication |