CN106595633B - Indoor orientation method and device - Google Patents
Indoor orientation method and device Download PDFInfo
- Publication number
- CN106595633B CN106595633B CN201611070529.6A CN201611070529A CN106595633B CN 106595633 B CN106595633 B CN 106595633B CN 201611070529 A CN201611070529 A CN 201611070529A CN 106595633 B CN106595633 B CN 106595633B
- Authority
- CN
- China
- Prior art keywords
- pedestrian
- indoor
- probability
- movement
- default node
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
- G01C21/005—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 with correlation of navigation data from several sources, e.g. map or contour matching
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
- G01C21/20—Instruments for performing navigational calculations
- G01C21/206—Instruments for performing navigational calculations specially adapted for indoor navigation
Abstract
The present invention provides a kind of indoor orientation method and device, belongs to indoor positioning technologies field.Method includes: to predict the location information of pedestrian according to the collected data of multiple sensor;Based on indoor sport model, the indoor sport state of pedestrian is obtained;Based on indoor environment cartographic model, is calibrated according to indoor sport state and the location information of indoor default node, the pedestrian position information obtained to prediction, obtain the final position information of pedestrian.The location information that the present invention passes through prediction pedestrian.Based on indoor sport model, the indoor sport state of pedestrian is obtained.Based on indoor environment cartographic model, is calibrated according to indoor sport state and the location information of indoor default node, the pedestrian position information obtained to prediction, obtain the final position information of pedestrian.Due to not having to installation external equipment, to can also reduce hardware cost consumption, cost is relatively low so that expending when indoor positioning while avoiding design complexities higher system.
Description
Technical field
The present invention relates to indoor positioning technologies fields, more particularly, to a kind of indoor orientation method and device.
Background technique
In current many application fields, obtains people or object location information is all extremely important.With science and technology with
And communication technology etc. is constantly progressive, the location technology based on LBS (Location Based Service, location-based service) is also swift and violent
Development.LBS mainly solves the problems, such as it is that already existing resource is supplied to user, and the surrounding enviroment locating for it provide
Specific service.
In an outdoor environment, it can be obtained by GPS (Global Positioning System, global positioning system)
Location information.As information technology develops rapidly, GPS positioning technology can satisfy the daily demand of general public.Correspondingly,
GPS technology is always the mainstream studied.Since GPS is positioned by satellite, before four satellites position simultaneously
Best positioning accuracy can just be acquired by putting.But indoors in the environment of, since building blocks GPS signal,
Satellite signal strength and quality dramatic decrease.Simultaneously as indoor environment is sufficiently complex, it is accurately fixed not can be carried out indoors
Position.Therefore, indoor positioning technologies difficulty is larger.
In addition, research shows that 90% time is all indoors in people's daily life, that is to say, that people's most of the time is all
Except GPS signal positioning.It can be seen that indoor positioning technologies are also the focused on demand of people.With intelligent communication device
Development, the research of indoor positioning technologies has also been developed.Smart phone and tablet computer used in people's daily life
Equal intelligent mobile terminals, are integrated with MEMS (Micro-Electro-Mechanical System, MEMS), acceleration
The high-precision sensors elements such as meter, gyroscope and barometer.Being widely used for these sensor elements, mentions for indoor positioning technologies
The precision of basis and sensor element has been supplied also to can satisfy the demand of indoor positioning technologies.In addition, assistant positioning system is also same
When obtain extensively and adequately study.For example, the location technology of indoor positioning technologies, view-based access control model based on wireless fingerprint signal
Deng.These modes are either used in combination with sensor, or as independent location technology.Correspondingly, the essence of indoor positioning
Degree also increases accordingly.In recent years, the companies such as Google begin setting up indoor map, gradually cover some metropolitan significant build
Object is built, so that indoor map can also become the supplementary means of positioning.
Since 2000, intelligent terminal gradually popularize on the basis of, LBS be gradually used in Emergency Assistance,
The fields such as disaster prevention, logistics management, equipment inspection and health care.In this context, for positioning from outdoor to interior
Seamless connection demand also be proposed out.
In public safety guarantee, commercialized services and in Warehouse Service field, high-precision indoor positioning technologies have
Very important meaning.For example, in terms of social public security guarantee, accurately indoor positioning technologies can for fireman,
The personnel such as police provide indoor navigation service, and position etc. in prison to convict.In commercialized services, accurately
Indoor positioning technologies can provide positioning service for family, can provide navigation in the indoor environments such as market, museum, cure
It may provide for the nurse positioning service of old man, child and patient in institute, family.It, can be to article in Warehouse Service
It is positioned.Therefore, indoor positioning technologies, which possess, is widely applied scene.
On this basis, the research direction of pedestrian's indoor positioning and airmanship mainly towards it is inexpensive, easily it is portable and
Improve positioning accuracy development.Existing indoor orientation method mainly installs multiple external equipments in advance, such as AP (Acess
Piont, access point), according to the mapping relations of signal strength between mobile terminal and external equipment and distance, pedestrian is carried out
Positioning.
In the implementation of the present invention, the existing technology has at least the following problems for discovery: need to be mounted so that multiple outer
Portion's equipment, equipment cost expend higher.In addition, usually requiring the higher system of design complexities between multiple external equipments to assist
With work.Therefore, the higher cost expended when indoor positioning.
Summary of the invention
The present invention provide a kind of indoor orientation method for overcoming the above problem or at least being partially solved the above problem and
Device.
According to an aspect of the present invention, a kind of indoor orientation method is provided, this method comprises:
According to the collected data of multiple sensor, the location information of pedestrian is predicted;
Based on indoor sport model, the indoor sport state of pedestrian is obtained;
Based on indoor environment cartographic model, according to indoor sport state and the location information of indoor default node, to prediction
Obtained pedestrian position information is calibrated, and the final position information of pedestrian is obtained.
According to another aspect of the present invention, a kind of indoor positioning device is provided, which includes:
Prediction module, for predicting the location information of pedestrian according to the collected data of multiple sensor;
Module is obtained, for being based on indoor sport model, obtains the indoor sport state of pedestrian;
Calibration module, for being based on indoor environment cartographic model, according to indoor sport state and the position of indoor default node
Confidence breath, the pedestrian position information obtained to prediction are calibrated, and the final position information of pedestrian is obtained.
The technical solution that the application proposes has the benefit that
By predicting the location information of pedestrian according to the collected data of multiple sensor.Based on indoor sport model, obtain
Take the indoor sport state of pedestrian.Based on indoor environment cartographic model, according to indoor sport state and the position of indoor default node
Confidence breath, the pedestrian position information obtained to prediction are calibrated, and the final position information of pedestrian is obtained.It is outer due to not having to installation
Portion's equipment, so that hardware cost consumption can be also reduced, so that indoor while avoiding design complexities higher system
Expend that cost is relatively low when positioning.
In addition, having chosen accelerometer, barometer and gyro data when due to carrying out motion feature classification, improving
Accuracy when motion feature is classified, while can be avoided the appearance of long-time accumulated error.Since position fixing process navigates pedestrian
Position speculates that algorithm, indoor pedestrian movement's feature and Hidden Markov Model matching process are combined togather, thus guaranteeing
While located higher accuracy rate, the robustness of indoor positioning can also be promoted.
Detailed description of the invention
Fig. 1 is a kind of flow diagram of indoor orientation method of the embodiment of the present invention;
Fig. 2 is a kind of flow diagram of indoor orientation method of the embodiment of the present invention;
Fig. 3 is a kind of structural schematic diagram of indoor positioning device of the embodiment of the present invention.
Specific embodiment
With reference to the accompanying drawings and examples, specific embodiments of the present invention will be described in further detail.Implement below
Example is not intended to limit the scope of the invention for illustrating the present invention.
Existing indoor orientation method mainly installs multiple external equipments in advance, such as AP (Acess Piont, access
Point), according to the mapping relations of signal strength between mobile terminal and external equipment and distance, to be positioned to pedestrian.Due to
Need to install multiple external equipments, equipment cost expends higher.In addition, usually requiring design complexities between multiple external equipments
Higher system cooperates.Therefore, the higher cost expended when indoor positioning.
For the problems of the prior art, a kind of indoor orientation method is present embodiments provided.This method is applied to movement
Terminal, mobile terminal include but is not limited to mobile phone, tablet computer and smartwatch etc..In addition, since the embodiment of the present invention needs
The collected data of sensor are applied to, so that mountable in mobile terminal have acceleration transducer, gyroscope and barometer etc.,
The present embodiment is also not especially limited this.
Referring to Fig. 1, method flow provided in this embodiment includes: 101, according to the collected data of multiple sensor, in advance
Survey the location information of pedestrian;102, it is based on indoor sport model, obtains the indoor sport state of pedestrian;103, it is based on indoor environment
Cartographic model, according to indoor sport state and the location information of indoor default node, pedestrian position information that prediction is obtained into
Row calibration, obtains the final position information of pedestrian.
Method provided in an embodiment of the present invention, by predicting the position of pedestrian according to the collected data of multiple sensor
Information.Based on indoor sport model, the indoor sport state of pedestrian is obtained.Based on indoor environment cartographic model, according to indoor fortune
The location information of dynamic state and indoor default node, the pedestrian position information obtained to prediction are calibrated, and obtain pedestrian most
Whole location information.Due to not having to installation external equipment, to can also be reduced while avoiding design complexities higher system
Hardware cost consumption, so that expending when indoor positioning, cost is relatively low.
In addition, having chosen accelerometer, barometer and gyro data when due to carrying out motion feature classification, improving
Accuracy when motion feature is classified, while can be avoided the appearance of long-time accumulated error.Since position fixing process navigates pedestrian
Position speculates that algorithm, indoor pedestrian movement's feature and Hidden Markov Model matching process are combined togather, thus guaranteeing
While located higher accuracy rate, the robustness of indoor positioning can also be promoted.
The location information of pedestrian is predicted according to the collected data of multiple sensor as a kind of alternative embodiment, is wrapped
It includes:
Determine pedestrian from the total step number begun to move between stopping movement;
The mobile each step of pedestrian is calculated according to the location information of the step-length of pedestrian, steering angle and pedestrian before mobile
Pedestrian moves the location information after moving a step, until calculation times reach total step number, using final calculation result as the position of pedestrian
Information.
As a kind of alternative embodiment, determine that pedestrian before the total step number begun to move between stopping movement, also wraps
It includes:
According to the acceleration of each sampled point, starts mobile to pedestrian and stop mobile detecting.
Movement is started to pedestrian and is detected, is wrapped according to the acceleration of each sampled point as a kind of alternative embodiment
It includes:
Any sampled point is determined when detecting the acceleration of any sampled point not less than the first preset threshold
Pedestrian starts to move on any sampled point.
Movement is stopped to pedestrian and is detected, is wrapped according to the acceleration of each sampled point as a kind of alternative embodiment
It includes:
For any sampled point, when detecting the acceleration of any sampled point less than the first preset threshold, to from any
Sampling is lighted, and is continuously counted less than the sampled point quantity of the first preset threshold;
When statistical result reaches preset quantity, the last one sampled point when statistical result reaches preset quantity is obtained,
Determine that pedestrian stops movement on the last one sampled point.
As a kind of alternative embodiment, determine pedestrian from the total step number begun to move between stopping movement, comprising:
For pedestrian from beginning to move into any sampled point stopped in mobile this period, when detecting any sampled point
Acceleration be greater than the second preset threshold, and the corresponding acceleration of next sampled point of any sampled point is less than the second preset threshold
When, using next sampled point of a upper sampled point for any sampled point, any sampled point and any sampled point as a paces week
Phase, and the total step number of pedestrian is added one.
As a kind of alternative embodiment, by a upper sampled point for any sampled point, any sampled point and any sampled point
Next sampled point is as a paces period, and before the total step number of pedestrian is added one, further includes:
The acceleration correlation in the period is cut down according to previous step, calculates the second preset threshold.
Row is calculated according to the location information of the step-length of pedestrian, steering angle and pedestrian before mobile as a kind of alternative embodiment
People moves before the location information after moving a step, further includes:
Based on space coordinates, the angular acceleration of current pedestrian in three directions is obtained;
Based on the projection relation between space coordinates and earth axes, according to the angle of current pedestrian in three directions
Acceleration and acceleration calculate the steering angle of pedestrian.
The location information of pedestrian is predicted according to the collected data of multiple sensor as a kind of alternative embodiment, is wrapped
It includes:
According to the air pressure value of pedestrian present position, floor locating for pedestrian is determined.
As a kind of alternative embodiment, it is based on indoor sport model, obtains the indoor sport state of pedestrian, comprising:
According to the acceleration of each sampled point in first time window, the corresponding characteristic value of first time window is calculated;
The motion feature of pedestrian is determined according to the corresponding characteristic value of first time window based on motion state classifier.
As a kind of alternative embodiment, it is based on indoor sport model, obtains the indoor sport state of pedestrian, comprising:
According to the steering angle of the second time window one skilled in the art, the angle of turn of pedestrian is calculated;
According to the angle of turn of pedestrian, the turning feature of pedestrian is determined.
As a kind of alternative embodiment, it is based on indoor environment cartographic model, according to indoor sport state and indoor default section
The location information of point, the pedestrian position information obtained to prediction are calibrated, and the final position information of pedestrian is obtained, comprising:
Based on indoor environment cartographic model, determine that pedestrian is moved to the shifting of adjacent default node in indoor environment cartographic model
Dynamic probability;
The movement probability of each adjacent default node is ranked up, maximum two movements of numerical value in ranking results are chosen
Probability, respectively first movement probability and the second movement probability, first movement probability are greater than the second movement probability;
When the ratio of first movement probability and the second movement probability is greater than third predetermined threshold value, by first movement probability pair
Final position information of the location information for the adjacent default node answered as pedestrian.
As a kind of alternative embodiment, it is based on indoor environment cartographic model, determines pedestrian's with being moved to indoor environment artwork
The movement probability of adjacent default node in type, comprising:
For adjacent default node any in indoor environment cartographic model, believed according to the position of any adjacent default node
Breath calculates the emission probability that pedestrian is moved to any adjacent default node;
According to movement identification probability matrix, determine that the motion state of any adjacent default node shows as indoor sport state
State recognition probability;
The shifting of any adjacent default node is moved to using the product between emission probability and state recognition probability as pedestrian
Dynamic probability.
As a kind of alternative embodiment, the movement probability of each adjacent default node is ranked up, chooses ranking results
Middle maximum two movement probabilities of numerical value, respectively after first movement probability and the second movement probability, further includes:
When the ratio of first movement probability and the second movement probability is not more than third predetermined threshold value, obtained row will be predicted
Final position information of people's location information as pedestrian.
All the above alternatives can form alternative embodiment of the invention using any combination, herein no longer
It repeats one by one.
Based on content provided by above-mentioned Fig. 1 corresponding embodiment, the embodiment of the invention provides a kind of indoor orientation methods.
Referring to fig. 2, method flow provided in this embodiment includes: 201, determines pedestrian from the total step begun to move between stopping movement
Number;202, each step mobile for pedestrian is calculated according to the location information of the step-length of pedestrian, steering angle and pedestrian before mobile
Pedestrian moves the location information after moving a step, until calculation times reach total step number, using final calculation result as the position of pedestrian
Information;203, it is based on indoor sport model, obtains the indoor sport state of pedestrian;204, indoor environment cartographic model, root are based on
It calibrates, obtains according to indoor sport state and the location information of indoor default node, the pedestrian position information obtained to prediction
The final position information of pedestrian.
Wherein, 201, determine pedestrian from beginning to move into the total step number stopped between mobile.
Method provided in this embodiment is mainly first to predict the position letter of pedestrian according to the collected data of multiple sensor
Breath, then the pedestrian position information obtained to prediction are calibrated to realize the process positioned.Wherein, this step 201 is to step 202
Mainly according to the collected data of multiple sensor, the process of pedestrian position information is predicted.
Since when pedestrian starts movement and when stop movement to be unknown indoors, thus executing sheet
Before step 201, mobile and stopping movement can also being started to pedestrian according to the acceleration of each sampled point and detected, this reality
It applies example and this is not especially limited.
About the acceleration according to each sampled point, the mobile mode detected is started to pedestrian, the present embodiment is to this
It is not especially limited, including but not limited to: for any sampled point, when the acceleration for detecting any sampled point is not less than first
When preset threshold, determine that pedestrian starts to move on any sampled point.
Wherein, sampled point it is corresponding be sensor sampling period.First preset threshold can value according to the actual situation, this
Embodiment is not especially limited this.For example, being 1.5m/s with the first preset threshold2For.If the sampling of acceleration transducer
Period is 20ms, then every 20ms be exactly a sampled point.I.e. acceleration transducer acquires acceleration value on each sampled point
When, mobile terminal also judges whether collected acceleration value on each sampled point is greater than or equal to 1.5m/s2.If detecting
The acceleration of one sampled point is greater than 1.5m/s2, it is determined that pedestrian moves since being lighted the sampling.
The present embodiment to the acceleration according to each sampled point, does not stop the mobile mode detected to pedestrian and makees specifically
It limits, including but not limited to: for any sampled point, when detecting the acceleration of any sampled point less than the first preset threshold
When, it lights to from any sampling, is continuously counted less than the sampled point quantity of the first preset threshold;When statistical result reaches pre-
If when quantity, obtaining the last one sampled point when statistical result reaches preset quantity, determining pedestrian in the last one sampled point
Upper stopping movement.Wherein, preset quantity can also be configured according to the actual situation, and the present embodiment is not especially limited this.
For example, being 1.5m/s with the first preset threshold2, for preset quantity is 30.If detecting the 10th sampled point
Acceleration be less than 1.5m/s2, then continue to test the 11st, the 12nd ..., the acceleration of the 40th sampled point.When the 10th
The acceleration of continuous 30 sampled points is both less than 1.5m/s after a sampled point2When, i.e., when the 11st, the 12nd ..., the 40th
The acceleration of sampled point is less than 1.5m/s2When, obtain the last one sampled point when having reached 30 sampled points, i.e., the 40th
Sampled point.Correspondingly, on the 40th sampled point, pedestrian stops movement.
Determine pedestrian start it is mobile and stop it is mobile after, it may be determined that pedestrian stops mobile this period from beginning to move into
The total step number of interior walking.The present embodiment is not made to have to determining pedestrian from the mode for stopping the total step number between mobile is begun to move into
Body limits, including but not limited to: for pedestrian from any sampled point stopped in mobile this period is begun to move into, working as detection
Acceleration to any sampled point is greater than the second preset threshold, and the corresponding acceleration of next sampled point of any sampled point is less than
When the second preset threshold, next sampled point of a upper sampled point for any sampled point, any sampled point and any sampled point is made
For a paces period, and the total step number of pedestrian is added one.
When walking due to people, acceleration of the one leg when carrying away is biggish, and prepares acceleration when landing after carrying away
Degree is lesser, to based on the above principles, can detect in above process to continuous two sampled points.Two are adopted
Sampling point is compared with the second preset threshold, when previous sampled point is greater than the second preset threshold and the latter sampled point less than the
When two preset thresholds, it can be considered that pedestrian has walked a step.It is greater than the sampled point of the second preset threshold accordingly for acceleration, when
It, can be by a upper sampled point to next sampled point when the corresponding acceleration of next sampled point of the sampled point is less than the second preset threshold
Between this period as a paces period.For example, if the acceleration of the 3rd sampled point is greater than the second preset threshold and the
The acceleration of 4 sampled points, then can be by the 2nd sampled point, the 3rd sampled point and the 4th sampled point less than the second preset threshold
As a paces period, and it is considered as pedestrian and has walked a step within this paces period.Correspondingly, total step number can add one.It needs
Illustrate, the initial value of total step number is 0 before counting total step number.
In addition, can also cut down the acceleration correlation in the period according to previous step before determining total step number, it is pre- to calculate second
If threshold value, the present embodiment is not especially limited this.According to the experimental result to pedestrian's dead reckoning, the second preset threshold can lead to
Dynamic threshold equation is crossed to calculate, the dynamic threshold equation (1) is as follows:
Wherein, α and β is the parameter pre-set, and can distinguish value is 0.25 and 0.75, and the present embodiment does not make this to have
Body limits.γ be ambient noise variance, can value be 0.09, the present embodiment is also not especially limited this.ToldIt is cut down for previous step
Second preset threshold in period, A1And A2For acceleration correlation.A1And A2It each can represent previous step and cut down acceleration in the period
Maximum value and minimum value, A1And A2It can also be the mean value or variance of acceleration, the present embodiment is not especially limited this.
Wherein, 202, the position for the mobile each step of pedestrian, before being moved according to the step-length of pedestrian, steering angle and pedestrian
Information calculates pedestrian and moves the location information after moving a step, until calculation times reach total step number, using final calculation result as row
The location information of people.
Before executing this step, can first obtain the step-length of pedestrian, pedestrian it is mobile before location information and pedestrian steering
Angle.Wherein, the step-length of pedestrian estimates that the present embodiment is not especially limited this according to pedestrian's height.Due to the present embodiment
The method of offer is iterative process, i.e., pedestrian's last time it is mobile before the side that can also provide through this embodiment of location information
Method obtains, thus obtain pedestrian it is mobile before location information when, can obtain last time executes method institute provided in this embodiment
Corresponding calculated result.It should be noted that initial position can be according to reality when first time carrying out indoor positioning to pedestrian
Border situation is configured, and the present embodiment is not especially limited this.
In addition, be possible to do when moving due to pedestrian is curvilinear motion, thus in order to more accurately to pedestrian position
It sets and is predicted, the steering angle of pedestrian can also be obtained.The present embodiment does not limit the mode for the steering angle for obtaining pedestrian specifically
It is fixed, including but not limited to: being based on space coordinates, obtain the angular acceleration of current pedestrian in three directions;It is sat based on space
Projection relation between mark system and earth axes, according to the angular acceleration and acceleration of current pedestrian in three directions, meter
Calculate the steering angle of pedestrian.
Wherein, the angular speed that can be divided into according to space coordinates on three directions of XYZ axis.Angular acceleration on three directions
It can be obtained by the gyroscope in mobile terminal, the present embodiment is not especially limited this.In the steering angle for calculating pedestrian
Before, the angular displacement on three directions, the present embodiment can be calculated according to the angular acceleration on three directions by way of integral
This is not especially limited.Integral process can refer to following formula (2):
Wherein,AndAngular acceleration respectively on tri- directions XYZ.tbeginWhen to move the starting to move a step
It carves, tstopTo move the end time to move a step.θx、θyAnd θzAngular displacement respectively on tri- directions XYZ.
It should be noted that if when the angular displacement in a time window on tri- directions XYZ is smaller, such as less than
Angle threshold, then can be considered that pedestrian makees in this time window is linear motion.For example, by taking angle threshold is 15 ° as an example.
When the angular displacement in a time window on tri- directions XYZx、θyAnd θzAt respectively less than 15 °, then it can determine pedestrian at this
What is made in time window is linear motion.
Based on the above principles, pedestrian can calculate the acceleration arithmetic average on three directions during straight line is walked
Value.Correspondingly, the steering of pedestrian can be calculated according to the angular displacement of current pedestrian in three directions and acceleration arithmetic mean of instantaneous value
Angle, the present embodiment are not especially limited this.Above-mentioned calculating process can pass through such as undefined pedestrian's dead reckoning model (3)
It is indicated:
Wherein, OzFor the steering angle of pedestrian.θx、θyAnd θzAngular displacement on respectively three directions.AndPoint
It Wei not acceleration arithmetic mean of instantaneous value on three directions.
After the steering angle of pedestrian is calculated, each step mobile for pedestrian can step-length according to pedestrian, steering angle
And the location information before pedestrian's movement, it calculates pedestrian and moves the location information after moving a step.The calculating process can refer to following formula
(4):
Wherein, lkFor the step-length of pedestrian.Based on earth axes, xk-1And yk-1The position letter before moving a step is being moved for pedestrian
Breath, xkAnd ykThe location information after moving a step is being moved for pedestrian.
Based on the total step number determined in above-mentioned steps 201, it can determine that pedestrian is walking so multistep according to above-mentioned formula (4)
Location information afterwards, that is, the pedestrian position information predicted.
It should be noted that mainly location information of the pedestrian based on earth axes of above process prediction.Due to
When carrying out indoor positioning to pedestrian, it may be desirable to be positioned to the floor of building where it, thus the row that prediction obtains
People's location information can also include the floor where pedestrian.The present embodiment additionally provides a kind of side of determining pedestrian place floor
Method, including but not limited to: according to the air pressure value of pedestrian present position, determining floor locating for pedestrian.
Since air pressure is at being highly inversely proportional in earth atmosphere, i.e., when height increases, air pressure can be reduced,
So as to the air pressure obtained using barometric surveying, height where pedestrian is calculated.According to the layer of building where pedestrian
The high height with where pedestrian, it may be determined that the floor where pedestrian, the present embodiment are not especially limited this.Based on ICAO
(International Civil Aviation Organization, International Civil Aviation Organization) model would know that height is every
Increase about 8.7 meters of atmospheric pressures and reduces 1mbar.According to normal atmospheric pressure in 1993, the height where calculating pedestrian can join
Examine following formula (5):
Wherein, P0That represent is normal atmospheric pressure (1013.25mbar).H is the height where pedestrian, and unit is rice.
Wherein, 203, it is based on indoor sport model, obtains the indoor sport state of pedestrian.
Before executing this step, pedestrian's motion state indoors first can be determined according to the exercise habit of pedestrian indoors,
The present embodiment is not especially limited this.In the present embodiment, the motion state of pedestrian is divided into 7 according to real life scene
It is a, be respectively as follows: walking, sit down, stand, upstairs, downstairs, turning and U-shaped turning.
Wherein, it walks, sit down, standing, being upstairs and downstairs motion feature in pedestrian's moving process.It turns and U-shaped
Turning is turning feature of the pedestrian in moving process.Since pedestrian is in turning, it may be possible to which the small turning of adjustment direction also has
It may be the big turning turned around, thus in order to distinguish both turnings, divided when above-mentioned turning feature divides in order to turn and U
Type turning.When determining turning with U-shaped turning, can according to angle of turn of the pedestrian in time window and preset threshold range it
Between comparison result determine.For example, when angle of turn is at 50 °~135 °, it is believed that be a common turning.Work as turning angle
When degree is greater than 135 °, it is believed that be a u turn.
Based on above content, this step when obtaining the indoor design condition of pedestrian, can obtain respectively pedestrian motion feature and
Turning feature.Not to indoor sport model is based on, the mode for obtaining the motion feature of pedestrian specifically limits the present embodiment, including
But it is not limited to: according to the acceleration of each sampled point in first time window, calculates the corresponding characteristic value of first time window;Base
The motion feature of pedestrian is determined according to the corresponding characteristic value of first time window in motion state classifier.
Wherein, motion state classifier can be instructed by KNN (k-Nearest-Neighbor, K arest neighbors) classifier
Practice, the present embodiment is not especially limited this.The experimenter that specifically can first choose preset quantity practices above-mentioned five kinds of movements spy
Sign walks, sits down, standing, upstairs and downstairs, and obtaining characteristic value of the experimenter when practicing motion feature.Passing through will
A kind of characteristic value of motion feature and each experimenter when practicing the motion feature every time is input in KNN model, to KNN
Model is ceaselessly trained, until frequency of training reaches preset times.Finally, according to the available movement of training result
State classifier and movement identification probability matrix.Wherein, motion state classifier is used to determine pedestrian according to the characteristic value of input
Motion feature.
Movement identification probability matrix is identified pair of every kind of motion feature of pedestrian and is identified as other motion features
Probability matrix.For example, by taking the current motion feature of pedestrian is actually " sitting down " as an example.Based on motion state classifier, root
The motion feature that can recognize that the pedestrian according to the characteristic value of input is " sitting down " really.At this point, recognition result be correctly, this
Correspond to a probability.But originally it is " sitting down ", it is also possible to be identified as " walking ".At this point, recognition result be it is wrong, this
Also a probability is corresponded to.Based on above-mentioned theory, by ceaselessly training to KNN model, motion state classifier is being obtained
At the same time it can also obtain movement identification probability matrix.
Correspondingly, after obtaining motion state classifier, pedestrian need to only be obtained in the corresponding characteristic value of first time window,
It can be based on motion state classifier, determine the motion feature of pedestrian.Wherein, characteristic value can be according to each in first time window
The acceleration of sampled point is calculated, and the present embodiment is not especially limited this.Characteristic value may include in first time window
On three directions on the mean value Yu variance, air pressure value of acceleration and three directions acceleration amplitude etc., the present embodiment pair
This is not especially limited.The length of first time window can take 2s~3s, and the present embodiment is not especially limited this.It considers
The continuity that pedestrian acts when walking, first time window may be arranged as 50% overlapping covering, and the present embodiment is to this
It is not especially limited.For example, first time window can be 0s~2s with value, 1s~3s, 2s~4s ....
Motion feature in addition to determining pedestrian, this can determine the turning feature of pedestrian.The present embodiment is not to determining pedestrian
The mode of turning feature specifically limit, including but not limited to: according to the steering angle of the second time window one skilled in the art, calculating row
The angle of turn of people;According to the angle of turn of pedestrian, the turning feature of pedestrian is determined.
Wherein, the length of the second time window can also take 2s~3s, and the present embodiment is not especially limited this.In addition,
First time window and the length of the second time window may be the same or different, and the present embodiment does not also limit this specifically
It is fixed.Specifically when calculating the angle of turn of pedestrian, the first steering angle of the second time window initial time can be first calculated, then calculate
Second steering angle of the second time window end time.Using the difference between the first steering angle and the second steering angle as pedestrian's
Angle of turn.Wherein, the process for calculating steering angle can refer to the process of above-mentioned formula (3), and details are not described herein again.
For example, being 2s with the second time window length, for the second time window is 1s~3s.In the turning for calculating pedestrian
When angle, the steering angle of pedestrian when can first calculate 1s, then when calculating 3s pedestrian steering angle, thus by the difference of two steering angles
Angle of turn as pedestrian.
By the above process, the motion feature and turning feature of pedestrian may finally be obtained.Wherein, motion feature is row
Walk, sit down, standing, upstairs and downstairs one of, turning feature be turning and it is U-shaped turning one of.It is movement character combined
And turning feature, the indoor sport state of pedestrian can be obtained.
Wherein, 204, it is based on indoor environment cartographic model, is believed according to the position of indoor sport state and indoor default node
Breath, the pedestrian position information obtained to prediction are calibrated, and the final position information of pedestrian is obtained.
Due to above-mentioned steps 201 predicted into step 202 pedestrian position information may there is a certain error,
To this step can based in indoor environment cartographic model preset node location information, in step 201 to step 202
Predict that obtained pedestrian position information is calibrated.The present embodiment is not to indoor environment cartographic model is based on, according to indoor sport
The location information of state and indoor default node, the pedestrian position information obtained to prediction are calibrated, and the final of pedestrian is obtained
The mode of location information specifically limits, including but not limited to: being based on indoor environment cartographic model, determines that pedestrian is moved to interior
The movement probability of adjacent default node in environmental map model;The movement probability of each adjacent default node is ranked up, is selected
Take maximum two movement probabilities of numerical value, respectively first movement probability and the second movement probability, first movement in ranking results
Probability is greater than the second movement probability;It, will when the ratio of first movement probability and the second movement probability is greater than third predetermined threshold value
Final position information of the location information of the corresponding adjacent default node of first movement probability as pedestrian.
Wherein, indoor environment cartographic model is to be obtained by real life scenario building.Specifically, stair in a solitary building
The position of the nodes such as corner point and passageway corner point is usually can be predetermined.Node coordinate, each according to each node
The motion state of direction and distance, pedestrian on each node between node and its adjacent node that can spatially go directly, can
Establish indoor environment cartographic model.
It is being based on indoor environment cartographic model, is determining that pedestrian is moved to adjacent default node in indoor environment cartographic model
Location information before movement probability, before can first obtaining pedestrian's movement.By above-mentioned steps 201 to the content in step 204 it is found that
Since method provided in this embodiment is iterative calculation, i.e., continues deduction pedestrian according to last calculated result and move next time
Location information after dynamic, so that this step when obtaining the location information before pedestrian's movement, can obtain the last time by this implementation
The pedestrian position information that the method that example provides is calculated, the present embodiment are not especially limited this.
After the location information before determining that pedestrian is mobile, according to the position for each presetting node in indoor environment cartographic model
Information can determine that pedestrian can directly reach which of indoor environment cartographic model from the position before movement and preset node.
These default nodes are adjacent default node corresponding to position before pedestrian is mobile.Based on above content, for any phase
The default node of neighbour, the present embodiment are not moved to the movement probability of adjacent default node in indoor environment cartographic model to determining pedestrian
Mode specifically limit, including but not limited to: for adjacent default node any in indoor environment cartographic model, according to any
The location information of adjacent default node calculates the emission probability that pedestrian is moved to any adjacent default node;It is identified according to movement
Probability matrix determines that the motion state of any adjacent default node shows as the state recognition probability of indoor sport state;It will hair
The product penetrated between probability and state recognition probability is moved to the movement probability of any adjacent default node as pedestrian.
Wherein, emission probability is the probability that an observation state is generated by a hidden state, the position before pedestrian is mobile
Information is a hidden state, and the location information of adjacent default node is observation state.When calculating emission probability, can refer to as
Lower formula (6):
In above-mentioned formula (6), P (zt|ri) it is emission probability.ztFor the location information of adjacent default node, riFor pedestrian
Location information before movement, zt-riIndicate Euclidean distance between the two.σ is the standard deviation of pedestrian's moving distance calculated value, this reality
The value for applying σ in example is 0.1.
Based on the related content for moving identification probability matrix in above-mentioned steps 203, for any adjacent default node, true
It, can be according to indoor sport shape when the motion state of the fixed adjacent default node shows as the state recognition probability of indoor sport state
The motion feature of motion feature and the adjacent default node in state searches corresponding probability in movement identification probability matrix.It will
The probability found is as state recognition probability.
In addition, in the movement for being moved to any adjacent default node according to emission probability and state recognition probability calculation pedestrian
When probability, calculating process can refer to following formula (7):
P(zt,mt|ri)=P (zt|ri)P(mt|ri) (7)
Wherein, P (zt|ri) it is emission probability, P (mt|ri) it is state recognition probability, P (zt,mt|ri) it is movement probability.
Due to default node adjacent with the mobile front position of pedestrian in environmental map model indoors might have it is multiple, from
And the movement probability of each adjacent default node can be calculated according to above-mentioned calculating process.After multiple movement probabilities are calculated,
All movement probabilities can be ranked up, maximum two movement probabilities of numerical value are chosen from ranking results, are calculated as the first shifting
Dynamic probability and the second movement probability.Wherein, first movement probability is greater than the second movement probability, i.e. first movement probability is mobile general
Maximum value in rate, the second movement probability are the second largest value in movement probability.
In order to more accurately determine the location information of pedestrian, the ratio of first movement probability and the second movement probability can be calculated
Value.When ratio is greater than third predetermined threshold value, illustrate that pedestrian is moved to the corresponding adjacent default node of first movement probability can
Can property, a possibility that being significantly larger than moved to the second movement probability corresponding adjacent default node.Thus, it is believed that pedestrian is mobile
When adjacent default node corresponding to first movement probability, confidence level is higher.At this point, can be corresponding adjacent by first movement probability
Final position information of the location information of default node as pedestrian, and abandon the position predicted in above-mentioned steps 201 to 202
Confidence breath.
In addition, when the ratio of first movement probability and the second movement probability is not more than third predetermined threshold value, it can will be above-mentioned
Final position information of the step 201 to the pedestrian position information predicted in 202 as pedestrian.
Method provided in an embodiment of the present invention, by determining pedestrian from the total step number begun to move between stopping movement.
It is mobile that pedestrian is calculated according to the location information of the step-length of pedestrian, steering angle and pedestrian before mobile for the mobile each step of pedestrian
Location information after one step, until calculation times reach total step number, using final calculation result as the location information of pedestrian.It is based on
Indoor sport model obtains the indoor sport state of pedestrian.Based on indoor environment cartographic model, according to indoor sport state and room
The location information of interior default node, the pedestrian position information obtained to prediction are calibrated, and the final position information of pedestrian is obtained.
Due to not having to installation external equipment, to can also reduce hardware cost while avoiding design complexities higher system and disappear
Consumption, so that expending when indoor positioning, cost is relatively low.
In addition, having chosen accelerometer, barometer and gyro data when due to carrying out motion feature classification, improving
Accuracy when motion feature is classified, while can be avoided the appearance of long-time accumulated error.Since position fixing process navigates pedestrian
Position speculates that algorithm, indoor pedestrian movement's feature and Hidden Markov Model matching process are combined togather, thus guaranteeing
While located higher accuracy rate, the robustness of indoor positioning can also be promoted.
The embodiment of the invention provides a kind of indoor positioning device, the device is for executing the corresponding reality of above-mentioned Fig. 1 or Fig. 2
Apply indoor orientation method provided in example.Referring to Fig. 3, which includes:
Prediction module 301, for predicting the location information of pedestrian according to the collected data of multiple sensor;
Module 302 is obtained, for being based on indoor sport model, obtains the indoor sport state of pedestrian;
Calibration module 303, for being based on indoor environment cartographic model, according to the default node of indoor sport state and interior
Location information, the pedestrian position information obtained to prediction are calibrated, and the final position information of pedestrian is obtained.
As a kind of alternative embodiment, prediction module 301, comprising:
Determination unit, for determining pedestrian from the total step number begun to move between stopping movement;
First computing unit, it is mobile according to the step-length of pedestrian, steering angle and pedestrian for each step mobile for pedestrian
Preceding location information, knot will finally be calculated until calculation times reach total step number by calculating the location information that pedestrian moves after moving a step
Location information of the fruit as pedestrian.
As a kind of alternative embodiment, prediction module 301, further includes:
Detection unit starts mobile and stops mobile detecting for the acceleration according to each sampled point to pedestrian.
As a kind of alternative embodiment, detection unit, for for any sampled point, when detect any sampled point plus
When speed is not less than the first preset threshold, determine that pedestrian starts to move on any sampled point.
As a kind of alternative embodiment, detection unit, for for any sampled point, when detect any sampled point plus
When speed is less than the first preset threshold, lights to from any sampling, continuously carried out less than the sampled point quantity of the first preset threshold
Statistics;When statistical result reaches preset quantity, the last one sampled point when statistical result reaches preset quantity is obtained, is determined
Pedestrian stops movement on the last one sampled point.
As a kind of alternative embodiment, determination unit, for moving this period from stopping is begun to move into for pedestrian
Interior any sampled point, when the acceleration for detecting any sampled point is greater than the second preset threshold, and any sampled point is next
When the corresponding acceleration of sampled point is less than the second preset threshold, by a upper sampled point for any sampled point, any sampled point and appoint
Next sampled point of one sampled point adds one as a paces period, and by the total step number of pedestrian.
As a kind of alternative embodiment, prediction module 301, further includes:
Second computing unit calculates the second preset threshold for cutting down the acceleration correlation in the period according to previous step.
As a kind of alternative embodiment, prediction module 301, further includes:
Acquiring unit obtains the angular acceleration of current pedestrian in three directions for being based on space coordinates;
Third computing unit, for based on the projection relation between space coordinates and earth axes, according to current line
The angular acceleration and acceleration of people in three directions, calculate the steering angle of pedestrian.
As a kind of alternative embodiment, prediction module 301 is determined for the air pressure value according to pedestrian present position
Floor locating for pedestrian.
As a kind of alternative embodiment, module 302 is obtained, for the acceleration according to each sampled point in first time window
Degree calculates the corresponding characteristic value of first time window;Based on motion state classifier, according to the corresponding feature of first time window
Value, determines the motion feature of pedestrian.
As a kind of alternative embodiment, module 302 is obtained, for the steering angle according to the second time window one skilled in the art, meter
Calculate the angle of turn of pedestrian;According to the angle of turn of pedestrian, the turning feature of pedestrian is determined.
As a kind of alternative embodiment, calibration module 303, comprising:
Determination unit determines that pedestrian is moved to phase in indoor environment cartographic model for being based on indoor environment cartographic model
The movement probability of the default node of neighbour;
Selection unit chooses numerical value in ranking results for the movement probability of each adjacent default node to be ranked up
Maximum two movement probabilities, respectively first movement probability and the second movement probability, it is mobile that first movement probability is greater than second
Probability;
First comparing unit is greater than third predetermined threshold value for the ratio when first movement probability and the second movement probability
When, using the location information of the corresponding adjacent default node of first movement probability as the final position information of pedestrian.
As a kind of alternative embodiment, determination unit, for for adjacent default section any in indoor environment cartographic model
Point calculates the emission probability that pedestrian is moved to any adjacent default node according to the location information of any adjacent default node;Root
According to movement identification probability matrix, determine that the motion state of any adjacent default node shows as the state recognition of indoor sport state
Probability;The movement for being moved to any adjacent default node using the product between emission probability and state recognition probability as pedestrian is general
Rate.
As a kind of alternative embodiment, calibration module 303, further includes:
Second comparing unit is not more than third predetermined threshold value for the ratio when first movement probability and the second movement probability
When, the pedestrian position information that prediction is obtained is as the final position information of pedestrian.
Device provided in an embodiment of the present invention, by predicting the position of pedestrian according to the collected data of multiple sensor
Information.Based on indoor sport model, the indoor sport state of pedestrian is obtained.Based on indoor environment cartographic model, according to indoor fortune
The location information of dynamic state and indoor default node, the pedestrian position information obtained to prediction are calibrated, and obtain pedestrian most
Whole location information.Due to not having to installation external equipment, to can also be reduced while avoiding design complexities higher system
Hardware cost consumption, so that expending when indoor positioning, cost is relatively low.
In addition, having chosen accelerometer, barometer and gyro data when due to carrying out motion feature classification, improving
Accuracy when motion feature is classified, while can be avoided the appearance of long-time accumulated error.Since position fixing process navigates pedestrian
Position speculates that algorithm, indoor pedestrian movement's feature and Hidden Markov Model matching process are combined togather, thus guaranteeing
While located higher accuracy rate, the robustness of indoor positioning can also be promoted.
Finally, the present processes are only preferable embodiment, it is not intended to limit the scope of the present invention.It is all
Within the spirit and principles in the present invention, any modification, equivalent replacement, improvement and so on should be included in protection of the invention
Within the scope of.
Claims (6)
1. a kind of indoor orientation method, which is characterized in that the described method includes:
According to the collected data of multiple sensor, the location information of pedestrian is predicted;
Based on indoor sport model, the indoor sport state of the pedestrian is obtained;
Based on indoor environment cartographic model, according to the indoor sport state and the location information of indoor default node, to prediction
Obtained pedestrian position information is calibrated, and the final position information of the pedestrian is obtained;
It is described to be based on indoor environment cartographic model, it is right according to the indoor sport state and the location information of indoor default node
It predicts that obtained pedestrian position information is calibrated, obtains the final position information of the pedestrian, comprising:
Based on indoor environment cartographic model, determine that the pedestrian is moved to adjacent default node in the indoor environment cartographic model
Movement probability;
The movement probability of each adjacent default node is ranked up, it is general to choose maximum two movements of numerical value in ranking results
Rate, respectively first movement probability and the second movement probability, the first movement probability are greater than second movement probability;
It is when the ratio of the first movement probability and the second movement probability is greater than third predetermined threshold value, the first movement is general
Final position information of the location information of the corresponding adjacent default node of rate as the pedestrian.
2. obtaining the pedestrian's the method according to claim 1, wherein described be based on indoor sport model
Indoor sport state, comprising:
According to the acceleration of each sampled point in first time window, the corresponding characteristic value of the first time window is calculated;
Based on motion state classifier, according to the corresponding characteristic value of the first time window, determine that the movement of the pedestrian is special
Sign.
3. obtaining the pedestrian's the method according to claim 1, wherein described be based on indoor sport model
Indoor sport state, comprising:
According to the steering angle of the pedestrian in the second time window, the angle of turn of the pedestrian is calculated;
According to the angle of turn of the pedestrian, the turning feature of the pedestrian is determined.
4. determining the row the method according to claim 1, wherein described be based on indoor environment cartographic model
People is moved to the movement probability of adjacent default node in the indoor environment cartographic model, comprising:
For adjacent default node any in the indoor environment cartographic model, according to the position of any adjacent default node
Information, calculates the emission probability that the pedestrian is moved to any adjacent default node, and the emission probability is hidden by one
Hiding state generates the probability of an observation state, and the location information before the pedestrian is mobile is a hidden state, described any
The location information of adjacent default node is an observation state;
According to movement identification probability matrix, determine that the motion state of any adjacent default node shows as the indoor sport
The state recognition probability of state, the movement identification probability matrix are identified pair of every kind of motion feature of pedestrian and are identified
For the probability matrix of other motion features;
It is moved to using the product between the emission probability and the state recognition probability as the pedestrian described any adjacent
The movement probability of default node.
5. the method according to claim 1, wherein the movement probability by each adjacent default node carries out
Sequence, choose ranking results in maximum two movement probabilities of numerical value, respectively first movement probability and the second movement probability it
Afterwards, further includes:
When the ratio of the first movement probability and the second movement probability is not more than third predetermined threshold value, obtained row will be predicted
Final position information of people's location information as the pedestrian.
6. a kind of indoor positioning device, which is characterized in that described device includes:
Prediction module, for predicting the location information of pedestrian according to the collected data of multiple sensor;
Module is obtained, for being based on indoor sport model, obtains the indoor sport state of the pedestrian;
Calibration module, for being based on indoor environment cartographic model, according to the indoor sport state and the position of indoor default node
Confidence breath, the pedestrian position information obtained to prediction are calibrated, and the final position information of the pedestrian is obtained;
The calibration module includes:
Determination unit determines that the pedestrian is moved to the indoor environment cartographic model for being based on indoor environment cartographic model
In adjacent default node movement probability;
It is maximum to choose numerical value in ranking results for the movement probability of each adjacent default node to be ranked up for selection unit
Two movement probabilities, respectively first movement probability and the second movement probability, the first movement probability is greater than described second
Movement probability;
First comparing unit is greater than third predetermined threshold value for the ratio when the first movement probability and the second movement probability
When, using the location information of the corresponding adjacent default node of the first movement probability as the final position information of the pedestrian.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201611070529.6A CN106595633B (en) | 2016-11-25 | 2016-11-25 | Indoor orientation method and device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201611070529.6A CN106595633B (en) | 2016-11-25 | 2016-11-25 | Indoor orientation method and device |
Publications (2)
Publication Number | Publication Date |
---|---|
CN106595633A CN106595633A (en) | 2017-04-26 |
CN106595633B true CN106595633B (en) | 2019-07-19 |
Family
ID=58595552
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201611070529.6A Active CN106595633B (en) | 2016-11-25 | 2016-11-25 | Indoor orientation method and device |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106595633B (en) |
Families Citing this family (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107063268A (en) * | 2017-06-05 | 2017-08-18 | 合肥展游软件开发有限公司 | A kind of museum's navigation system and application method |
CN110692260B (en) | 2017-06-30 | 2021-08-17 | 北京嘀嘀无限科技发展有限公司 | Terminal equipment positioning system and method |
CN107356229B (en) * | 2017-07-07 | 2021-01-05 | 中国电子科技集团公司电子科学研究院 | Indoor positioning method and device |
CN109413683B (en) * | 2017-08-15 | 2021-09-21 | 华为技术有限公司 | Method and device for acquiring emission probability, transition probability and sequence positioning |
CN108645406A (en) * | 2018-04-19 | 2018-10-12 | 北京理工大学 | A kind of indoor autonomic positioning method based on score field pedestrian movement perception |
CN109991641A (en) * | 2019-04-10 | 2019-07-09 | 广东工业大学 | A kind of acquisition methods, the apparatus and system of position of mobile equipment information |
CN111854753B (en) * | 2020-06-02 | 2023-05-23 | 深圳全景空间工业有限公司 | Modeling method for indoor space |
CN115336936A (en) * | 2021-05-12 | 2022-11-15 | 尚科宁家(中国)科技有限公司 | Floor-crossing control method for cleaning robot and cleaning robot |
CN115223442B (en) * | 2021-07-22 | 2024-04-09 | 上海数川数据科技有限公司 | Automatic generation method of indoor pedestrian map |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102821464A (en) * | 2012-08-13 | 2012-12-12 | 北京邮电大学 | Indoor storey positioning method and device |
CN103701991A (en) * | 2013-12-20 | 2014-04-02 | 百度在线网络技术(北京)有限公司 | Mobile terminal state recognition method and mobile terminal |
CN104215238A (en) * | 2014-08-21 | 2014-12-17 | 北京空间飞行器总体设计部 | Indoor positioning method of intelligent mobile phone |
CN104977006A (en) * | 2015-08-11 | 2015-10-14 | 北京纳尔信通科技有限公司 | Indoor positioning method based on fuzzy theory and multi-sensor fusion |
CN105004343A (en) * | 2015-07-27 | 2015-10-28 | 上海美琦浦悦通讯科技有限公司 | Indoor wireless navigation system and method |
CN105509736A (en) * | 2015-12-03 | 2016-04-20 | 北京机械设备研究所 | Indoor composite locating method for fire rescue |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
TWI397671B (en) * | 2009-12-16 | 2013-06-01 | Ind Tech Res Inst | System and method for locating carrier, estimating carrier posture and building map |
-
2016
- 2016-11-25 CN CN201611070529.6A patent/CN106595633B/en active Active
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102821464A (en) * | 2012-08-13 | 2012-12-12 | 北京邮电大学 | Indoor storey positioning method and device |
CN103701991A (en) * | 2013-12-20 | 2014-04-02 | 百度在线网络技术(北京)有限公司 | Mobile terminal state recognition method and mobile terminal |
CN104215238A (en) * | 2014-08-21 | 2014-12-17 | 北京空间飞行器总体设计部 | Indoor positioning method of intelligent mobile phone |
CN105004343A (en) * | 2015-07-27 | 2015-10-28 | 上海美琦浦悦通讯科技有限公司 | Indoor wireless navigation system and method |
CN104977006A (en) * | 2015-08-11 | 2015-10-14 | 北京纳尔信通科技有限公司 | Indoor positioning method based on fuzzy theory and multi-sensor fusion |
CN105509736A (en) * | 2015-12-03 | 2016-04-20 | 北京机械设备研究所 | Indoor composite locating method for fire rescue |
Non-Patent Citations (2)
Title |
---|
Smartphone based Indoor Position and Orientation Tracking Fusing Inertial and Magnetic Sensing;Chengkai Huang 等;《17th International Symposium on Wireless Personal Multimedia Communications》;20141231;第205-220页 |
基于粒子滤波的智能移动终端室内定位方法研究;闫亭亭;《中国优秀硕士学位论文全文数据库 信息科技辑》;20140815(第08期);I136-293 |
Also Published As
Publication number | Publication date |
---|---|
CN106595633A (en) | 2017-04-26 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106595633B (en) | Indoor orientation method and device | |
Kang et al. | SmartPDR: Smartphone-based pedestrian dead reckoning for indoor localization | |
Lan et al. | Using smart-phones and floor plans for indoor location tracking-Withdrawn | |
CN104501796B (en) | A kind of interior WLAN/MEMS merges across floor 3-dimensional localization method | |
US8965688B2 (en) | System and method for locating, tracking, and/or monitoring the status of personnel and/or assets both indoors and outdoors | |
US9146113B1 (en) | System and method for localizing a trackee at a location and mapping the location using transitions | |
KR101851836B1 (en) | Systems and methods for estimating the motion of an object | |
US8731827B2 (en) | Method and apparatus for estimating path for indoor localization | |
US9807725B1 (en) | Determining a spatial relationship between different user contexts | |
Shi et al. | A robust pedestrian dead reckoning system using low-cost magnetic and inertial sensors | |
JP5754051B2 (en) | Apparatus, system and method for determining the characteristics of a path formed by successive positions of a three-axis accelerometer rigidly coupled to a moving element | |
Shin et al. | Indoor 3D pedestrian tracking algorithm based on PDR using smarthphone | |
Pei et al. | Motion recognition assisted indoor wireless navigation on a mobile phone | |
CN104977006A (en) | Indoor positioning method based on fuzzy theory and multi-sensor fusion | |
US20140019082A1 (en) | Method of calculating step length | |
JP2013531781A (en) | Method and system for detecting zero speed state of object | |
Shih et al. | Estimating walking distance with a smart phone | |
CN109164411A (en) | A kind of personnel positioning method based on multi-data fusion | |
Zhao et al. | Pseudo-zero velocity re-detection double threshold zero-velocity update (ZUPT) for inertial sensor-based pedestrian navigation | |
JP2002139340A (en) | Walking navigation device and navigation system using the same | |
Pei et al. | Using motion-awareness for the 3D indoor personal navigation on a Smartphone | |
KR101523147B1 (en) | Indoor Positioning Device and Method | |
KR20160004084A (en) | Pedestrian Tracking apparatus and method at indoors | |
Lan et al. | An indoor locationtracking system for smart parking | |
KR101997967B1 (en) | Stair gait measuring device and method using the same |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |