CN105716604A - Mobile robot indoor positioning method and system based on geomagnetic sequences - Google Patents
Mobile robot indoor positioning method and system based on geomagnetic sequences Download PDFInfo
- Publication number
- CN105716604A CN105716604A CN201610102312.2A CN201610102312A CN105716604A CN 105716604 A CN105716604 A CN 105716604A CN 201610102312 A CN201610102312 A CN 201610102312A CN 105716604 A CN105716604 A CN 105716604A
- Authority
- CN
- China
- Prior art keywords
- earth magnetism
- sequence
- sub
- mobile apparatus
- particle
- 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
- 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
- G01C21/04—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 by terrestrial means
- G01C21/08—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 by terrestrial means involving use of the magnetic field of the earth
Abstract
The invention discloses a mobile robot indoor positioning method and system based on geomagnetic sequences. A geomagnetic sequence fingerprint collecting method of mobile robots and an improved particle filter positioning algorithm are involved in the mobile robot indoor positioning method based on the geomagnetic sequences. The system comprises a mobile robot side and a cloud server side, wherein the mobile robot side is mainly responsible for collecting geomagnetic sequence fingerprints, controlling motion, collecting motion information and applying for visualization of positioning services and positioning information to a server, and the cloud server side is mainly responsible for storing the geomagnetic sequence fingerprints, processing geomagnetic sequence fingerprint data, carrying out similarity calculation on the geomagnetic sequences and providing the poisoning services for the multiple mobile robots at the same time. The mobile robot indoor positioning method and system based on the geomagnetic sequences are hardly disturbed by the external environment, do not need other additional infrastructures and are suitable for positioning of the mobile robots in an indoor one-dimensional passage, and use and operation are easy. The practical running effect proves that the method has high positioning precision on the mobile robots.
Description
Technical field
The invention belongs to indoor positioning technologies field, specifically, particularly to based on mobile apparatus people's indoor orientation method of geomagnetic field information and system.
Background technology
Mobile apparatus people's indoor positioning refer to robot indoor environment by airborne sensor perception self and around environmental information, process and conversion obtains the pose estimation that robot is current through certain data, thus realizing the indoor positioning of mobile apparatus people.Autonomous location is the premise that mobile apparatus people completes intelligent task in indoor environment, robot realizes positioning accurately not only have great theory value and realistic meaning, and lays a good foundation for completing the follow-up mission planning of robot, raising intelligent robot level and reliability.
Traditional method for positioning mobile robot has dead reckoning to position.Dead reckoning localization method does not need extraneous reference information, as long as speedometer or inertial navigation system by robot achieve that location, also referred to as relative localization.But, the shortcoming of reckoning location is to need to obtain robot initial position, and there is the cumulative error increased in time.Be not suitable for the realization of precise positioning.
Based on mobile apparatus people's indoor orientation method of distance measuring sensor, realize mainly through laser, ultrasound wave and the airborne sensor such as infrared.But in order to obtain high-precision location, the sensor such as laser is expensive.In the complex scene that crowd is relatively more, locating effect is a greater impact.In special scenes, for instance Glazed fence or railing, owing to distance measuring sensor cannot record actual distance, this localization method is also inapplicable.
Based in the method for positioning mobile robot of computer vision, owing to visual information is computationally intensive, the change at visual angle, illumination the reason such as change, the extraction of visual signature and coupling are with certain difficulty.Especially when the indoor environment that reply is complicated, in order to obtain higher positioning accuracy, algorithm is complicated, and amount of calculation is relatively big, and current locating effect is not good.
The simultaneous localization and mapping problem of the robot in map match localization method is the important research direction of localization for Mobile Robot (SLAM).In SLAM problem, environmental map robot is unknown in advance, and robot obtains environmental information thus setting up indoor environment map by airborne sensor, calculates position simultaneously, and both interdepend, inseparable.Still suffering from observation error, association error and cumulative error in whole process, researcher is it is also proposed that a lot of method and theory solve this problem.
The beacons such as indoor orientation method many employings Wi-Fi, bluetooth, RFID in recent years position, and are scanned detecting to beacon by airborne sensor, then carry out location estimation by trigonometry.Although wherein the mode such as bluetooth and RFID can obtain relatively high positioning precision, but is required for extra deployment base facility, and is disturbed by indoor arrangement and external signal, limit the large-scale use of location technology.Although as WiFi network at present large scale deployment reduce system cost, but factor impacts such as WiFi is vulnerable to that receiving sensitivity is relatively low, multipath effect, signal time variation and equipment diversity, positioning performance is always barely satisfactory.
Summary of the invention
For overcoming the shortcoming and defect of prior art, the present invention proposes a kind of Localization Approach for Indoor Mobile based on earth magnetism sequence and system.This localization method is little by external environment condition interference, it is not necessary to other additional infrastructure, uses simple to operate, it is adaptable to mobile apparatus people is in the location of indoor one-dimensional passage way.Actual motion effect proves that mobile apparatus people is had good positioning precision by the method.
It is an object of the invention to be achieved through the following technical solutions.
A kind of mobile apparatus people's indoor orientation method based on earth magnetism sequence, including the particle filtering algorithm of the earth magnetism sub-sequence fingerprint collection method of mobile apparatus people, improvement, the airborne geomagnetic sensor of mobile apparatus people, mileage gauge and the display screen etc. that the present invention uses.
Described earth magnetism sub-sequence fingerprint, refers to that mobile apparatus people is while uniform motion, utilizes the absolute force sequence that geomagnetic sensor gathers with fixed frequency.Earth magnetism sub-sequence fingerprint F={M1,M2,...,Mn, wherein, MiIt is the absolute force (1≤i≤n) of i-th, Mi=(Mix,Miy,Miz), Mix,Miy,MizRespectively geomagnetic field intensity MiComponent on x, y, z direction.
The earth magnetism sub-sequence fingerprint collection method of described mobile apparatus people, including following 3 steps:
Step 1, is divided into some straight lines by the corridor of indoor plane map, and with the starting point of each bar straight line of capitalization sequentially labelling, terminal, cross point；
Step 2, the artificial each bar corridor of mobile apparatus arrives terminal locality magnetic sub-sequence fingerprint from the off, and by the length of mileage gauge every corridor of record.
The data such as the earth magnetism sub-sequence fingerprint of collection and corridor length are uploaded to cloud server end, are stored in data base by step 3.
The particle filtering algorithm of described improvement is the improvement to particle filter algorithm, particle filter algorithm refers to by finding the expression probability density function that one group of random sample propagated in state space is similar to, integral operation is replaced by sample average, and then the process of the minimum variance estimate of acquisition system mode, these samples are called particle by image.In alignment system, each particle has position and two attributes of weights, and weights are the degrees of approximation of position and the mobile apparatus people's actual position assessing this particle.The particle filtering algorithm of described improvement comprises the following steps:
Step 1, the corridor gathering earth magnetism sub-sequence fingerprint in indoor map is uniformly distributed multiple particle, and each particle has positional information and weight information, and the more high explanation robot of weight is more likely in this position；
Step 2, mobile apparatus people obtains movable information during location by mileage gauge, and updates the positional information of each particle；
Step 3, mobile apparatus people obtains earth magnetism sub-sequence fingerprint collected in motor process during location, and is filtered this earth magnetism sub-sequence fingerprint processing.
Step 4, using the earth magnetism sub-sequence fingerprint after Filtering Processing as observed value, calculates itself and the similarity of the earth magnetism sub-sequence fingerprint of particle correspondence position in data base the weight information of each particle of reappraising；
Step 5, is normalized the weight information of all particles, chooses the optimal solution that current position of mobile robot is assessed；
Step 6, it may be judged whether lost efficacy in location.If losing efficacy in location, then return step 1 and reorientate；If not losing efficacy in location, then continue executing with step 7；
All particles are carried out resampling by step 7, and return execution step 2 and enter next round iteration.
In above-mentioned location algorithm, movable information described in step 2 includes direction Θ and displacement X, and direction Θ value only has+1 and-1, represents the both direction of corridor respectively.When updating particle location information, displacement X is added Gaussian noise.If particle is in outside border, corridor after motion updates, then it is 0 this particle weights information assignment, does not reappraise this particle weights information in step 4.In above-mentioned location algorithm, step 3 has the earth magnetism sub-sequence fingerprint collected during the storage location, relief area of a fixed size.When the earth magnetism sub-sequence fingerprint length collected when location is less than buffer size, then being not enough to position, mobile apparatus people continues motion pick earth magnetism sequence；When the earth magnetism sub-sequence fingerprint length collected when location is more than buffer size, relief area only stores the earth magnetism sub-sequence fingerprint data of up-to-date collection, and positions with the earth magnetism sub-sequence fingerprint in relief area.
In above-mentioned location algorithm, being filtered processing to the earth magnetism sub-sequence fingerprint gathered in step 3, in order to the variation tendency making earth magnetism sub-sequence fingerprint is smoother, eliminate noise jamming, present invention magnetic sub-sequence fingerprint over the ground employs low-pass filtering treatment；Its variation tendency is retained, it is to avoid the impact of the overall offset that the earth magnetism sub-sequence fingerprint that differently Magnetic Sensor gathers exists, present invention magnetic sub-sequence fingerprint over the ground employs high-pass filtering and processes in order to make earth magnetism sub-sequence fingerprint move to sustained height position.
In above-mentioned location algorithm, in step 4, the Similarity Measure of earth magnetism sub-sequence fingerprint present invention uses DTW(DynamicTimeWarping, dynamic time consolidation) algorithm, DTW algorithm is a kind of method weighing two different time series similarities of length, and being applied in process better has on the Similarity Measure of sequence of Similar trend.
In above-mentioned location algorithm, the strategy choosing optimal solution in step 5 is not simply take t to take turns the position of the highest particle of iteration weight as optimal solution, consider that earth magnetism sub-sequence fingerprint exists similar probability partially, the invention comprehensively utilizes historical location data to choose optimal solution.
In above-mentioned location algorithm, step 6 is be distributed, by aggregation of particles degree and weighted value, the judgement positioning inefficacy.If being judged as, lost efficacy in location, then restart location.
In above-mentioned location algorithm, along with the continuous iteration of system in step 7, particle is restrained gradually and is lacked multiformity, and most of particle set can cause the waste of calculating resource at regional area, comprehensively have employed three kinds of resampling strategies for solving the problem above present invention: 1) sampling of roulette dish；2) random distribution sampling；3) sampling near optimal solution.The sampling of described roulette dish refers to and carries out resampling under existing particle weights distribution situation, retains the particle of high weighted value.The sampling of described random distribution refer to mobile apparatus people up to collection the region random distribution particle of earth magnetism sub-sequence fingerprint.Near described optimal solution, sampling refers to that the particle position in optimal solution is distributed around particle.
In above-mentioned location algorithm, it is also possible to Global localization and local positioning can be included.Described Global localization requires to be uniformly distributed particle on the passage all gathering earth magnetism sub-sequence fingerprint, determines user's initial position；Described local positioning is after Global localization determines reliable initial position, chooses this initial position near zone distribution of particle and continues location.
Described mobile apparatus people's indoor locating system based on earth magnetism sequence, including mobile apparatus people's end and cloud server end, as follows:
(1) mobile apparatus people end.The collection of the collection of mobile apparatus people's end primary responsibility earth magnetism sub-sequence fingerprint, motor control and movable information, to server application positioning service and location information visuallization etc..
(2) cloud server end.Cloud server end primary responsibility storage earth magnetism sub-sequence fingerprint, the process of earth magnetism sub-sequence fingerprint data, the Similarity Measure of earth magnetism sequence and provide positioning service etc. for multiple stage mobile apparatus people simultaneously.
Compared with prior art, the inventive method is in that relative to the advantage of other technologies:
(1) adopt indoor magnetic-field vector of the earth as the fingerprint of location, signal stabilization, be susceptible to the environmental effects such as artificial abortion.
(2) need not additionally dispose the equipment such as beacon again, save great amount of cost, large-scale promotion application can be carried out.
(3) the coupling location of earth magnetism sub-sequence fingerprint, it is achieved complexity and positioning precision be better than the coupling of grid point.
(4) mobile apparatus people end equipment is simple, only need to arrange in pairs or groups earth magnetism and mileage gauge sensor just may be used.
(5) cloud server end provides positioning service, workable, it is simple to for multiple stage, mobile apparatus people provides positioning service simultaneously.
Accompanying drawing explanation
Fig. 1 is the description figure of the earth magnetism sub-sequence fingerprint collection method of the mobile apparatus people using the present invention；
Fig. 2 is the particle filtering algorithm frame figure of a kind of improvement provided by the invention；
Fig. 3 is the sub-sequence fingerprint schematic diagram of magnetic primitively that on the same passage that mobile apparatus people gathers, differently magnetic sensor device gathers；
Fig. 4 be on the same passage that mobile apparatus people gathers the sub-sequence fingerprint of magnetic primitively of differently magnetic sensor device collection process after filtering after schematic diagram；
Fig. 5 is the mobile apparatus people's indoor locating system Organization Chart based on earth magnetism sequence provided by the invention.
Detailed description of the invention
Below in conjunction with accompanying drawing, by specific embodiment, the present invention will be described in detail.Should be appreciated that embodiment herein is only for helping to understand the present invention, but do not limit present disclosure in any form, if there being the process of not detailed description especially below, being all that those skilled in the art can refer to existing techniques in realizing.
As it is shown in figure 1, in one embodiment of the invention, it is provided that the description figure of the earth magnetism sub-sequence fingerprint collection method of mobile apparatus people.
In the step 1 of collection method, the corridor of indoor plane map has been divided into 4 straight lines, is AB, CD, EH and FG successively.Mobile apparatus people's range coverage (WC, 301 ~ 310) is limited on the corridor representated by 4 straight lines.
In the step 2 of collection method, mobile apparatus people is followed successively by each bar linear pick-up earth magnetism sub-sequence fingerprint.Mobile apparatus people's uniform motion, and with certain frequency locality magnetic data, moved in such as mobile apparatus people 1s 0.25m acquire the geomagnetic data of 5 points.Uniform motion is also conducive to improving the matching precision of earth magnetism sub-sequence fingerprint with fixed frequency locality magnetic data.The earth magnetism sub-sequence fingerprint of the straight line AB gathered is expressed as FAB={M1,M2,...,Mn, wherein Mi=(Mix,Miy,Miz), (1≤i≤n), three component Mix,Miy,MizRespectively geomagnetic field intensity MiComponent on x, y, z direction.
In the step 3 of collection method, after the earth magnetism sub-sequence fingerprint gathered uploads to cloud server end, straight line in data base such as AB is likely to there is the earth magnetism sub-sequence fingerprint that a plurality of different time is uploaded, select most representational earth magnetism sub-sequence fingerprint in cluster as the earth magnetism sub-sequence fingerprint of straight line AB, the earth magnetism sub-sequence fingerprint selecting certain maloperation collection of user so can be avoided to position by AffinityPropagation (AP) clustering algorithm.
As in figure 2 it is shown, in another embodiment of the present invention, it is provided that the particle filtering algorithm frame figure of a kind of improvement, wherein:
In the step 1 of location algorithm, the positional information of particle represents mobile apparatus people and is likely to the position at place, and weight information represents the probability to this position and assesses, and weight more high explanation mobile apparatus people be more likely in this position.When Global localization, uniform particle is distributed to all regions gathering earth magnetism and is estimated, after moveable robot movement certain distance, if the particle weights information in certain region is persistently all high than other regions, then robot physical location is in this region, next enters local positioning；During local positioning, in the securing position near zone distribution of particle that Global localization is determined, by the earth magnetism sub-sequence fingerprint of the ensuing movable information of mobile apparatus people and collection, the physical location of robot is precisely located.
In the step 2 of location algorithm, according to the moveable robot movement information updating particle position obtained, due to many influence factors such as sensor acquisition equipment and natures, adding Gaussian noise, more new formula is as follows herein:
Xt=Xt-1+N(ΔXt,σ2)
XtThe position of iteration, X is taken turns for particle tt-1The position of iteration, N (Δ X is taken turns for particle t-1t,σ2) for the function of Gauss distribution, standard deviation sigma determines the amplitude of distribution.
In the step 3 of location algorithm, mobile apparatus people's uniform motion also collects field pulses fingerprint with fixed frequency.The earth magnetism sub-sequence fingerprint collected during the storage location, relief area of a fixed size is had during mobile apparatus people's end location.The earth magnetism sub-sequence fingerprint collected during location is stored in relief area, when the earth magnetism sub-sequence fingerprint length collected when location is less than buffer size, then is not enough to position, and mobile apparatus people continues motion pick earth magnetism sequence；When the earth magnetism sub-sequence fingerprint length collected when location is more than buffer size, relief area only stores the earth magnetism sub-sequence fingerprint data of up-to-date collection, and positions with the earth magnetism sub-sequence fingerprint in relief area.
In the step 3 of location algorithm, the earth magnetism sub-sequence fingerprint gathered during localization for Mobile Robot, need to carry out data prediction denoising before similarity mode, the present invention uses low-pass filtering and high-pass filtering.
Low-pass filtering is a kind of filter type, and rule can be normal through for low frequency signal, and the high-frequency signal exceeding setting marginal value is then blocked, weakens.Low-pass filtering treatment formula is as follows:
Yi=αXi+(1-α)Yi-1
Wherein, YiThe ground magnetic value of the earth magnetism sub-sequence fingerprint i-th that expression low-pass filtering treatment is crossed, XiRepresenting the ground magnetic value of magnetic sub-sequence fingerprint i-th primitively, α is low pass smoothing parameter, 0≤α≤1.
Relatively, high-pass filtering is also a kind of filter type, and rule can be normal through for high-frequency signal, is then blocked lower than the low frequency signal of setting marginal value, weakens.It is as follows that high-pass filtering processes formula:
Zi=βZi-1+β(Yi–Yi-1)
Wherein, ZiThe ground magnetic value of the earth magnetism sub-sequence fingerprint i-th that expression high-pass filtering processed, YiThe ground magnetic value of the earth magnetism sub-sequence fingerprint i-th that expression low-pass filtering treatment is crossed, β is high pass smoothing parameter (0≤β≤1).
Low-pass filtering treatment makes earth magnetism sub-sequence fingerprint variation tendency smoother, and high-pass filtering processes and makes earth magnetism sub-sequence fingerprint reservation variation tendency but move to sustained height position, it is simple to the matching primitives of similarity.
If Fig. 3 is the sub-sequence fingerprint schematic diagram of magnetic primitively that on the same passage of mobile apparatus people collection, differently magnetic sensor device gathers, as Fig. 4 be on the same passage that gathers of mobile apparatus people the sub-sequence fingerprint of magnetic primitively of differently magnetic sensor device collection process after filtering after schematic diagram, article three, after the earth magnetism sub-sequence fingerprint of the same straight line of distinct device collection processes after filtering, substantially overlapping together.The earth magnetism sub-sequence fingerprint and the earth magnetism sub-sequence fingerprint gathered that gather in position fixing process will first carry out low-pass filtering treatment, then carry out high-pass filtering process.
In the step 4 of location algorithm, the calculating of similarity uses DTW(DynamicTimeWarping, dynamic time consolidation) algorithm, DTW algorithm is a kind of method weighing two different time series similarities of length, and being applied in process better has on the Similarity Measure of sequence of Similar trend.When each particle earth magnetism sub-sequence fingerprint in relevant position and location, the earth magnetism sub-sequence fingerprint of relief area carries out DTW Similarity Measure, and result similarity diff represents.
In the step 4 of location algorithm, more the formula of the weight information of new particle is as follows:
P.Wt=P.Wt-1*e-λ*diff
Wherein, P represents particle, and P.W is the weighted value of particle, and t is iterations, and e is natural number, and λ is adjustable parameter (λ > 0), and diff is similarity.Can drawing according to formula, similarity is more high, and diff is closer to 0, P.Wt=P.Wt-1；Similarity is more low, and diff is more big, P.WtIt is closer to 0.So, each take turns iteration after, the particle weights value that similarity is high remains unchanged, and can be retained when in resampling；The particle weights value that similarity is low diminishes, and can be abandoned in resampling.
In the step 5 of location algorithm, owing in actual environment, earth magnetism sub-sequence fingerprint exists similar probability partially, so the optimal solution of position of mobile robot is not take the position of each particle taking turns iteration highest weight value.If t takes turns the highest weight weight values of all particles in iteration and takes turns the iterative position optimal solution particle weighted value in new round iteration more than t-1, and difference exceedes the threshold value of setting, then choose t and take turns the position optimal solution as t wheel iterative position of the iteration the highest particle of weight, otherwise, continue to choose t-1 take turns the particle of iterative position optimal solution in the position of a new round iteration optimal solution as position.
In the step 6 of location algorithm, the foundation that location inefficacy judges is to be distributed by aggregation of particles degree and weighted value.If the optimal solution ambient particles concentration class of position and Weight meansigma methods are lower than threshold value, then it is assumed that losing efficacy in location, restarts Global localization；Otherwise then think and position the continuation local positioning that do not lose efficacy.
In the step 7 of location algorithm, for avoiding convergence degradation phenomena and the computing resource waste of particle, the resampling strategy of particle comprehensively have employed following three kinds of strategies:
1) particle of 60% carries out roulette dish sampling, resampling on existing particle.This strategy remains the particle that similarity is high, has abandoned the particle that similarity is low；
2) particle of 20% carries out random distribution sampling.This strategy remains other position and there is the probability of optimal solution；
3) particle of 20% is sampled near optimal solution.This strategy makes optimal solution more accurate partially.
Above sampling policy is along with the continuous iteration of system, near the sampling of roulette dish and optimal solution, the population of sampling gradually decreases, the population of random distribution sampling remains unchanged, and the while of so keeping particle multifarious, also avoids multiparticle to restrain the phenomenon of the computing resource waste caused.
As it is shown in figure 5, in yet another embodiment of the present invention, it is provided that based on mobile apparatus people's indoor locating system Organization Chart of earth magnetism sequence.This system includes mobile apparatus people's end and cloud server end, as follows:
(1) mobile apparatus people end.Mobile apparatus people's terminal set sensor includes geomagnetic sensor, mileage gauge etc., and can carry out man-machine interaction with UI interface.Mobile apparatus people's end includes map building module, sensor assembly, motion-control module, navigation module.Described map building module is responsible for collecting the earth magnetism sub-sequence fingerprint of indoor map and being uploaded to cloud server end；Described sensor assembly is responsible for the collection of earth magnetism sub-sequence fingerprint and the acquisition of movable information；Described motion-control module is responsible for the motor control of mobile apparatus people；Described navigation module is responsible for mobile apparatus people navigation feature from starting point to destination.Mobile apparatus people, after cloud server end application positioning service, can visualize current location information on UI interface.
(2) cloud server end.Cloud server end primary responsibility storage earth magnetism sub-sequence fingerprint, the process of earth magnetism sub-sequence fingerprint data, earth magnetism sub-sequence fingerprint similarity mode calculate and provide positioning service etc. for multiple stage robot.Cloud server end comprises map service module, positioning service module, navigation Service module.Described map service module is responsible for storage and the renewal in earth magnetism sub-sequence fingerprint storehouse；Described positioning service module is responsible for processing the Location Request of multiple stage mobile apparatus people restoring to normal position result；Described navigation Service module is responsible for the path planning of mobile apparatus people.
Mobile apparatus people is by network service to cloud server end application positioning service, and the content of communication includes earth magnetism sub-sequence fingerprint, movable information, navigation information etc..Multiple stage mobile apparatus people can simultaneously to cloud server end application positioning service.
Earth magnetism location used by the present invention directly utilizes geomagnetic sensor locality signal magnetic field and positions, it is not necessary to extra deployment base facility again.The armored concrete of modern architecture or steel construction, produce metastable interference to the earth's magnetic field of subrange so that the indoor different spaces magnetic field property of there are differences is mainly manifested in intensity and the direction in magnetic field.Earth magnetism location technology carries out geomagnetic matching and location mainly by this difference.Integrated comparative, based on earth magnetism indoor positioning technologies the most prominent, no matter put into from software and hardware, implement difficulty, controllability, or locating effect aspect is investigated, and all has and has great advantage.It addition, when positioning with traditional particle filter algorithm, along with the continuous iteration of system, particle converges on minority state gradually, lacks multiformity, there is degradation phenomena, and computational efficiency depends on a number of subsets.
Although the present invention has been described in detail already by above example, but the present invention is not limited to above example, also includes change or amendment that technical staff makes within the scope of the present invention.
Claims (10)
1. based on mobile apparatus people's indoor orientation method of earth magnetism sequence, it is characterised in that include the earth magnetism sub-sequence fingerprint collection method of mobile apparatus people and the particle filtering algorithm of improvement；
Described earth magnetism sub-sequence fingerprint, refers to that mobile apparatus people is while uniform motion, utilizes the absolute force sequence that geomagnetic sensor gathers with fixed frequency；Earth magnetism sub-sequence fingerprint F={M1,M2,...,Mn, wherein, MiIt is the absolute force (1≤i≤n) of i-th, Mi=(Mix,Miy,Miz), Mix,Miy,MizRespectively geomagnetic field intensity MiComponent on x, y, z direction；
The earth magnetism sub-sequence fingerprint collection method of described mobile apparatus people comprises the following steps:
Step a, is divided into some straight lines by the corridor of indoor plane map, and with the starting point of each bar straight line of capitalization sequentially labelling, terminal, cross point；
Step b, the artificial each bar corridor of mobile apparatus arrives terminal locality magnetic sub-sequence fingerprint from the off, and by the length of mileage gauge every corridor of record；
The data such as the earth magnetism sub-sequence fingerprint of collection and corridor length are uploaded to cloud server end, are stored in data base by step c；
The particle filtering algorithm of described improvement comprises the following steps:
Step 1, the corridor gathering earth magnetism sub-sequence fingerprint in indoor map is uniformly distributed multiple particle, and each particle has positional information and weight information, and the more high explanation robot of weight is more likely in this position；
Step 2, mobile apparatus people obtains movable information during location by mileage gauge, and updates the positional information of each particle；
Step 3, mobile apparatus people obtains earth magnetism sub-sequence fingerprint collected in motor process during location, and is filtered this earth magnetism sub-sequence fingerprint processing；
Step 4, using the earth magnetism sub-sequence fingerprint after Filtering Processing as observed value, calculates itself and the similarity of the earth magnetism sub-sequence fingerprint of particle correspondence position in data base the weight information of each particle of reappraising；
Step 5, is normalized the weight information of all particles, chooses the optimal solution that current position of mobile robot is assessed；
Step 6, it may be judged whether lost efficacy in location；If losing efficacy in location, then return step 1 and reorientate；If not losing efficacy in location, then continue executing with step 7；
All particles are carried out resampling by step 7, and return execution step 2 and enter next round iteration.
2. the mobile apparatus people's indoor orientation method based on earth magnetism sequence according to claim 1, it is characterised in that movable information described in step 2 includes direction Θ and displacement X, direction Θ value only has+1 and-1, represents the both direction of corridor respectively；When updating particle location information, displacement X is added Gaussian noise；If particle is in outside border, corridor after motion updates, then it is 0 this particle weights information assignment, does not reappraise this particle weights information in step 4.
3. the mobile apparatus people's indoor orientation method based on earth magnetism sequence according to claim 1, it is characterised in that have the earth magnetism sub-sequence fingerprint collected during the storage location, relief area of a fixed size in step 3；When the earth magnetism sub-sequence fingerprint length collected when location is less than buffer size, then being not enough to position, mobile apparatus people continues motion pick earth magnetism sequence；When the earth magnetism sub-sequence fingerprint length collected when location is more than buffer size, relief area only stores the earth magnetism sub-sequence fingerprint data of up-to-date collection, and positions with the earth magnetism sub-sequence fingerprint in relief area.
4. the mobile apparatus people's indoor orientation method based on earth magnetism sequence according to claim 1, it is characterized in that the earth magnetism sub-sequence fingerprint gathered is filtered processing by step 3, in order to the variation tendency making earth magnetism sub-sequence fingerprint is smoother, eliminating noise jamming, magnetic sub-sequence fingerprint employs low-pass filtering treatment over the ground；Retaining its variation tendency to make magnetic sub-sequence fingerprint move to sustained height position, it is to avoid the impact of the overall offset that the earth magnetism sub-sequence fingerprint that differently Magnetic Sensor gathers exists, magnetic sub-sequence fingerprint employs high-pass filtering process over the ground.
5. according to the mobile apparatus people's indoor orientation method based on earth magnetism sequence described in claim 1, it is characterised in that in step 4, the Similarity Measure of earth magnetism sub-sequence fingerprint employs dynamic time consolidation and DTW algorithm.
6. according to the mobile apparatus people's indoor orientation method based on earth magnetism sequence described in claim 1, it is characterised in that the strategy choosing optimal solution in step 5 make use of historical location data to choose optimal solution.
7. according to the mobile apparatus people's indoor orientation method based on earth magnetism sequence described in claim 1, it is characterised in that step 6 is be distributed, by aggregation of particles degree and weighted value, the judgement positioning inefficacy；If being judged as, lost efficacy in location, then restart location.
8. according to the mobile apparatus people's indoor orientation method based on earth magnetism sequence described in claim 1, it is characterised in that step 7 have employed three kinds of resampling strategies: 1) sampling of roulette dish；2) random distribution sampling；3) sampling near optimal solution；The sampling of described roulette dish refers to and carries out resampling under existing particle weights distribution situation, retains the particle of high weighted value；The sampling of described random distribution refer to mobile apparatus people up to collection the region random distribution particle of earth magnetism sub-sequence fingerprint；Near described optimal solution, sampling refers to that the particle position in optimal solution is distributed around particle.
9. according to the mobile apparatus people's indoor orientation method based on earth magnetism sequence described in claim 1, it is characterised in that include Global localization and local positioning；Described Global localization requires to be uniformly distributed particle on the passage all gathering earth magnetism sub-sequence fingerprint, determines user's initial position；Described local positioning is after Global localization determines reliable initial position, chooses this initial position near zone distribution of particle and continues location.
10. realize described in any one of claim 1 ~ 9 system of mobile apparatus people's indoor orientation method based on earth magnetism sequence, it is characterized in that including mobile apparatus people's end and cloud server end, the collection of the collection of mobile apparatus people's end primary responsibility earth magnetism sub-sequence fingerprint, motor control and movable information, to server application positioning service and location information visuallization；Cloud server end primary responsibility storage earth magnetism sub-sequence fingerprint, the process of earth magnetism sub-sequence fingerprint data, the Similarity Measure of earth magnetism sequence and provide positioning service for multiple stage mobile apparatus people simultaneously.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610102312.2A CN105716604A (en) | 2016-02-25 | 2016-02-25 | Mobile robot indoor positioning method and system based on geomagnetic sequences |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610102312.2A CN105716604A (en) | 2016-02-25 | 2016-02-25 | Mobile robot indoor positioning method and system based on geomagnetic sequences |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN105716604A true CN105716604A (en) | 2016-06-29 |
Family
ID=56157028
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610102312.2A Pending CN105716604A (en) | 2016-02-25 | 2016-02-25 | Mobile robot indoor positioning method and system based on geomagnetic sequences |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105716604A (en) |
Cited By (26)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105979442A (en) * | 2016-07-22 | 2016-09-28 | 北京地平线机器人技术研发有限公司 | Noise suppression method and device and mobile device |
| CN106054748A (en) * | 2016-07-30 | 2016-10-26 | 重庆峰创科技有限公司 | Intelligent robot indoor composite positioning system based on Internet of things and cloud computing |
| CN106289257A (en) * | 2016-07-27 | 2017-01-04 | 无锡知谷网络科技有限公司 | Indoor orientation method and alignment system |
| CN106289242A (en) * | 2016-07-18 | 2017-01-04 | 北京方位捷讯科技有限公司 | Particle filtering method and device based on earth magnetism |
| CN106767828A (en) * | 2016-12-29 | 2017-05-31 | 南京邮电大学 | A kind of mobile phone indoor positioning solution |
| CN107504965A (en) * | 2017-08-15 | 2017-12-22 | 华北理工大学 | Underground location method based on magnetic characteristic parameter joint distance |
| CN107562058A (en) * | 2017-09-14 | 2018-01-09 | 重庆理工大学 | WiFi fingerprint acquisition systems and acquisition method based on location tags identification |
| CN107576325A (en) * | 2017-08-25 | 2018-01-12 | 北京麦钉艾特科技有限公司 | A kind of indoor positioning terminal for merging visual odometry and Magnetic Sensor |
| CN107907134A (en) * | 2017-11-13 | 2018-04-13 | 中国科学院光电研究院 | A kind of mileage information aids in the matched Vehicle positioning system of earth magnetism and method |
| CN108240808A (en) * | 2017-12-18 | 2018-07-03 | 苏州点格信息科技有限公司 | Indoor earth magnetism map automatic collector |
| CN108614980A (en) * | 2018-04-16 | 2018-10-02 | 西南科技大学 | A kind of the dynamic object positioning system and method for combining RFID and laser intelligence |
| CN109556600A (en) * | 2017-09-27 | 2019-04-02 | 腾讯科技（深圳）有限公司 | A kind of particular space localization method, device and storage equipment |
| WO2019095301A1 (en) * | 2017-11-17 | 2019-05-23 | 苏州点格信息科技有限公司 | Cloud service-based geomagnetic signal positioning navigation system |
| CN109839640A (en) * | 2019-03-08 | 2019-06-04 | 深圳市沃特沃德股份有限公司 | Sound localization method, system, computer equipment and storage medium |
| CN110057354A (en) * | 2019-03-11 | 2019-07-26 | 杭州电子科技大学 | One kind being based on the modified geomagnetic matching navigation method of magnetic declination |
| CN110057355A (en) * | 2019-04-18 | 2019-07-26 | 吉林大学 | A kind of indoor orientation method, device, system and calculate equipment |
| CN110554353A (en) * | 2019-08-29 | 2019-12-10 | 华中科技大学 | mobile robot absolute positioning method based on RFID system |
| CN110764497A (en) * | 2019-09-09 | 2020-02-07 | 深圳市无限动力发展有限公司 | Mobile robot remote control method, device, storage medium and remote control terminal |
| CN110779513A (en) * | 2019-10-28 | 2020-02-11 | 杭州十域科技有限公司 | Indoor public service place positioning system |
| CN110856112A (en) * | 2019-11-14 | 2020-02-28 | 深圳先进技术研究院 | Crowd-sourcing perception multi-source information fusion indoor positioning method and system |
| CN110955233A (en) * | 2018-09-25 | 2020-04-03 | 上海云绅智能科技有限公司 | Robot anti-collision method, robot and anti-collision system |
| CN110967013A (en) * | 2018-10-01 | 2020-04-07 | 刘佳兴 | Indoor area positioning system based on indoor geomagnetic field information and smart phone |
| CN111721292A (en) * | 2019-03-20 | 2020-09-29 | 北京智慧图科技有限责任公司 | Indoor positioning method utilizing geomagnetism |
| CN112880677A (en) * | 2019-12-01 | 2021-06-01 | 成都易书桥科技有限公司 | Indoor positioning algorithm based on geomagnetism and similarity matching |
| CN114245307A (en) * | 2021-12-21 | 2022-03-25 | 北京云迹科技股份有限公司 | Positioning method and device for robot, electronic equipment and storage medium |
| CN116793199A (en) * | 2023-08-24 | 2023-09-22 | 四川普鑫物流自动化设备工程有限公司 | Centralized multi-layer goods shelf four-way vehicle positioning system and method |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103925923A (en) * | 2014-05-07 | 2014-07-16 | 南京大学 | Geomagnetic indoor positioning system based on self-adaptive particle filter algorithm |
| CN103983266A (en) * | 2014-05-28 | 2014-08-13 | 北京天地方元科技有限公司 | Indoor locating method based on geomagnetic information and indoor locating system based on geomagnetic information |
| CN104618458A (en) * | 2015-01-13 | 2015-05-13 | 金廷岳 | Improved signal fingerprint acquiring method for indoor positioning method and system |
| CN104677352A (en) * | 2015-03-11 | 2015-06-03 | 北京建筑大学 | Data acquisition method and device for geomagnetic indoor positioning reference map |
| CN105043380A (en) * | 2015-06-29 | 2015-11-11 | 武汉大学 | Indoor navigation method based on a micro electro mechanical system, WiFi (Wireless Fidelity) positioning and magnetic field matching |
-
2016
- 2016-02-25 CN CN201610102312.2A patent/CN105716604A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103925923A (en) * | 2014-05-07 | 2014-07-16 | 南京大学 | Geomagnetic indoor positioning system based on self-adaptive particle filter algorithm |
| CN103983266A (en) * | 2014-05-28 | 2014-08-13 | 北京天地方元科技有限公司 | Indoor locating method based on geomagnetic information and indoor locating system based on geomagnetic information |
| CN104618458A (en) * | 2015-01-13 | 2015-05-13 | 金廷岳 | Improved signal fingerprint acquiring method for indoor positioning method and system |
| CN104677352A (en) * | 2015-03-11 | 2015-06-03 | 北京建筑大学 | Data acquisition method and device for geomagnetic indoor positioning reference map |
| CN105043380A (en) * | 2015-06-29 | 2015-11-11 | 武汉大学 | Indoor navigation method based on a micro electro mechanical system, WiFi (Wireless Fidelity) positioning and magnetic field matching |
Non-Patent Citations (1)
| Title |
|---|
| 张有为等: "《人机自然交互》", 30 September 2004 * |
Cited By (35)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106289242A (en) * | 2016-07-18 | 2017-01-04 | 北京方位捷讯科技有限公司 | Particle filtering method and device based on earth magnetism |
| CN105979442A (en) * | 2016-07-22 | 2016-09-28 | 北京地平线机器人技术研发有限公司 | Noise suppression method and device and mobile device |
| CN105979442B (en) * | 2016-07-22 | 2019-12-03 | 北京地平线机器人技术研发有限公司 | Noise suppressing method, device and movable equipment |
| CN106289257A (en) * | 2016-07-27 | 2017-01-04 | 无锡知谷网络科技有限公司 | Indoor orientation method and alignment system |
| CN106054748A (en) * | 2016-07-30 | 2016-10-26 | 重庆峰创科技有限公司 | Intelligent robot indoor composite positioning system based on Internet of things and cloud computing |
| CN106767828A (en) * | 2016-12-29 | 2017-05-31 | 南京邮电大学 | A kind of mobile phone indoor positioning solution |
| CN107504965A (en) * | 2017-08-15 | 2017-12-22 | 华北理工大学 | Underground location method based on magnetic characteristic parameter joint distance |
| CN107504965B (en) * | 2017-08-15 | 2020-07-31 | 华北理工大学 | Underground positioning method based on magnetic characteristic parameter combined distance |
| CN107576325B (en) * | 2017-08-25 | 2019-10-11 | 北京麦钉艾特科技有限公司 | A kind of indoor positioning terminal merging visual odometry and Magnetic Sensor |
| CN107576325A (en) * | 2017-08-25 | 2018-01-12 | 北京麦钉艾特科技有限公司 | A kind of indoor positioning terminal for merging visual odometry and Magnetic Sensor |
| CN107562058A (en) * | 2017-09-14 | 2018-01-09 | 重庆理工大学 | WiFi fingerprint acquisition systems and acquisition method based on location tags identification |
| CN107562058B (en) * | 2017-09-14 | 2020-08-04 | 重庆理工大学 | WiFi fingerprint acquisition system and acquisition method based on position tag identification |
| CN109556600A (en) * | 2017-09-27 | 2019-04-02 | 腾讯科技（深圳）有限公司 | A kind of particular space localization method, device and storage equipment |
| CN107907134A (en) * | 2017-11-13 | 2018-04-13 | 中国科学院光电研究院 | A kind of mileage information aids in the matched Vehicle positioning system of earth magnetism and method |
| WO2019095301A1 (en) * | 2017-11-17 | 2019-05-23 | 苏州点格信息科技有限公司 | Cloud service-based geomagnetic signal positioning navigation system |
| CN108240808A (en) * | 2017-12-18 | 2018-07-03 | 苏州点格信息科技有限公司 | Indoor earth magnetism map automatic collector |
| CN108614980A (en) * | 2018-04-16 | 2018-10-02 | 西南科技大学 | A kind of the dynamic object positioning system and method for combining RFID and laser intelligence |
| CN110955233A (en) * | 2018-09-25 | 2020-04-03 | 上海云绅智能科技有限公司 | Robot anti-collision method, robot and anti-collision system |
| CN110967013A (en) * | 2018-10-01 | 2020-04-07 | 刘佳兴 | Indoor area positioning system based on indoor geomagnetic field information and smart phone |
| CN109839640A (en) * | 2019-03-08 | 2019-06-04 | 深圳市沃特沃德股份有限公司 | Sound localization method, system, computer equipment and storage medium |
| CN109839640B (en) * | 2019-03-08 | 2020-10-09 | 深圳市沃特沃德股份有限公司 | Sound source positioning method, system, computer equipment and storage medium |
| CN110057354A (en) * | 2019-03-11 | 2019-07-26 | 杭州电子科技大学 | One kind being based on the modified geomagnetic matching navigation method of magnetic declination |
| CN111721292B (en) * | 2019-03-20 | 2023-06-16 | 北京智慧图科技有限责任公司 | Indoor positioning method utilizing geomagnetism |
| CN111721292A (en) * | 2019-03-20 | 2020-09-29 | 北京智慧图科技有限责任公司 | Indoor positioning method utilizing geomagnetism |
| CN110057355A (en) * | 2019-04-18 | 2019-07-26 | 吉林大学 | A kind of indoor orientation method, device, system and calculate equipment |
| CN110554353A (en) * | 2019-08-29 | 2019-12-10 | 华中科技大学 | mobile robot absolute positioning method based on RFID system |
| CN110764497A (en) * | 2019-09-09 | 2020-02-07 | 深圳市无限动力发展有限公司 | Mobile robot remote control method, device, storage medium and remote control terminal |
| CN110779513A (en) * | 2019-10-28 | 2020-02-11 | 杭州十域科技有限公司 | Indoor public service place positioning system |
| CN110779513B (en) * | 2019-10-28 | 2021-06-18 | 杭州十域科技有限公司 | Indoor public service place positioning system |
| CN110856112A (en) * | 2019-11-14 | 2020-02-28 | 深圳先进技术研究院 | Crowd-sourcing perception multi-source information fusion indoor positioning method and system |
| CN110856112B (en) * | 2019-11-14 | 2021-06-18 | 深圳先进技术研究院 | Crowd-sourcing perception multi-source information fusion indoor positioning method and system |
| CN112880677A (en) * | 2019-12-01 | 2021-06-01 | 成都易书桥科技有限公司 | Indoor positioning algorithm based on geomagnetism and similarity matching |
| CN114245307A (en) * | 2021-12-21 | 2022-03-25 | 北京云迹科技股份有限公司 | Positioning method and device for robot, electronic equipment and storage medium |
| CN116793199A (en) * | 2023-08-24 | 2023-09-22 | 四川普鑫物流自动化设备工程有限公司 | Centralized multi-layer goods shelf four-way vehicle positioning system and method |
| CN116793199B (en) * | 2023-08-24 | 2023-11-24 | 四川普鑫物流自动化设备工程有限公司 | Centralized multi-layer goods shelf four-way vehicle positioning system and method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105716604A (en) | Mobile robot indoor positioning method and system based on geomagnetic sequences | |
| CN108632761B (en) | Indoor positioning method based on particle filter algorithm | |
| CN103925923B (en) | A kind of earth magnetism indoor locating system based on adaptive particle filter device algorithm | |
| US9696165B2 (en) | Last-mile navigation using smartphones | |
| CN111491367B (en) | Indoor positioning method based on crowd sensing and multi-fusion technology | |
| CN109974712A (en) | It is a kind of that drawing method is built based on the Intelligent Mobile Robot for scheming optimization | |
| CN110146909A (en) | A kind of location data processing method | |
| US20150153151A1 (en) | Determining Location Using Magnetic Fields From AC Power Lines | |
| CN109164411B (en) | Personnel positioning method based on multi-data fusion | |
| CN106597363A (en) | Pedestrian location method in indoor WLAN environment | |
| CN106767828A (en) | A kind of mobile phone indoor positioning solution | |
| US20160195401A1 (en) | Method and system for locating an object | |
| CN103298156B (en) | Based on the passive multi-target detection tracking of wireless sensor network | |
| CN106525031A (en) | Combined indoor positioning method | |
| CN110308729A (en) | The AGV combined navigation locating method of view-based access control model and IMU or odometer | |
| CN105246153A (en) | High-density rapid collection method for indoor fingerprint positioning database | |
| CN112230243A (en) | Indoor map construction method for mobile robot | |
| CN109919862B (en) | Radar image denoising system and method and computer equipment | |
| CN106028446A (en) | Indoor parking lot location method | |
| CN108801267B (en) | Indoor anchor-free positioning method integrating multiple sensors | |
| CN110426037A (en) | A kind of pedestrian movement track real time acquiring method under enclosed environment | |
| CN108345823A (en) | A kind of barrier tracking and device based on Kalman filtering | |
| CN114119886A (en) | High-precision map point cloud reconstruction method and device, vehicle, equipment and storage medium | |
| CN102981160B (en) | Method and device for ascertaining aerial target track | |
| Kwak et al. | Magnetic field based indoor localization system: A crowdsourcing approach |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20160629 |