CN104546391B - Gyro stabilizer for tactile sticks and complementary filtering method thereof - Google Patents
Gyro stabilizer for tactile sticks and complementary filtering method thereof Download PDFInfo
- Publication number
- CN104546391B CN104546391B CN201510049666.0A CN201510049666A CN104546391B CN 104546391 B CN104546391 B CN 104546391B CN 201510049666 A CN201510049666 A CN 201510049666A CN 104546391 B CN104546391 B CN 104546391B
- Authority
- CN
- China
- Prior art keywords
- angle
- gyrosensor
- acceleration
- stick
- angular acceleration
- 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
Landscapes
- Gyroscopes (AREA)
Abstract
Disclosed are a gyro stabilizer for tactile sticks and a complementary filtering method thereof. The gyro stabilizer comprises a microcontroller unit, a posture sensor and a motor and carries a ranging module which is connected with a mobile terminal through a wireless module. The complementary filtering method includes: 1, preprocessing data of the sensors; 2, estimating an iterative process based on the angle of an acceleration sensor; 3, compensating measuring results of the acceleration sensor through angular acceleration of a gyro sensor. The gyro stabilizer and the complementary filtering method thereof have the advantages that shaking caused by movement of a blind person is avoided, a preset direction can be effectively locked for ranging, the danger detecting range of a tactile stick is widened, obstacles in air can be accurately recognized, practicality of the tactile stick is effectively improved, great convenience is brought to the blind person, stability is good, response speed is high, an algorithm is simple, noise and drift are significantly suppressed, data are smoother, and high response and high accuracy are ensured in case of great angular change.
Description
Technical field
The present invention relates to gyrostabilization unit and multi-sensor fusion technology field, it is specifically related to a kind of for blind man's stick
Gyrostabilization unit and its complementary filter method.
Background technology
Visually impaired people is a colony by social extensive concern.Traditional blind man's stick has perceived distance limited, it is impossible to side
Just the distinct disadvantages such as the barrier of the tactile upper part of the body are visited.New blind man's stick increased distance measurement function, to detect aerial barrier
And danger.But, the motion of the hand-held blind man's stick of visually impaired people causes range finding unstable, will greatly limit the function of blind man's stick, and with
Effective solution is had no toward document and patent.
Gyrostabilization unit is widely used in fields such as military affairs, navigation, detections.By some high-precision sensors,
And some data fusion methods, it is possible to quickly and stably tracking control system, at the same with motion servo system and stable control
System processed etc. coordinates, it is possible to achieve stable to follow the trail of and correct.Traditional data fusion method for being applied to gyrostabilization unit has
Complementary filter and Kalman filtering.Complementary filter method is difficult to solve the influence of noise of acceleration transducer itself.Kalman filters
Wave method needs to set up multiple system equations, and for the compact stabilized platform for being applied to blind man's stick, amount of calculation seems excessively huge.It is open
Number be given for the patent " carrier speed and corner compensating measurement method based on accelerometer and gyroscope " of CN103913171A
The conventional method of angular surveying, but the corner of one degree of freedom can only be measured, and response speed still can not meet will
Ask.
The content of the invention
Present invention aims to above-mentioned existing problems and deficiency, there is provided one kind can avoid because of blind-person sports band
That what is come rocks, and can effectively lock pre-configured orientation and be found range, and has expanded the scope that blind man's stick surveys danger so that the barrier in space
Can be accurately identified, effectively increase the practicality of blind man's stick, bring blind person the gyrocontrol for blind man's stick of great convenience
Device.On this basis, further aim of the present invention is to provide one kind with preferable stability and response speed, and calculates
Method is relatively simple, suppresses obvious to noise and drift, and smoothness is not only had more in data, while becoming in significantly angle
In the case of change, also there is the complementary filter method of the gyrostabilization unit of response speed quickly and very high accuracy.
The technical scheme is that what is be achieved in that:
Gyrostabilization unit for blind man's stick of the present invention, is characterized in including microcontroller and connects with microcontroller
The attitude transducer for connecing and motor, wherein the attitude transducer includes acceleration transducer and gyrosensor, the micro-control
Device processed is also associated with range finder module and the wireless module for being connected with mobile terminal signal, and the range finder module and motor phase
Connection, the motor can be three for steering wheel or brushless electric machine or Servo-controller, the acceleration transducer and gyrosensor
Axle sensor.
The complementary filter method of the gyrostabilization unit for blind man's stick of the present invention, is characterized in including following step
Suddenly:
The first step:Acceleration transducer and gyrosensor are placed, gyrostabilization unit is built, after pretreatment,
The angle of acceleration transducer is obtained respectivelyθ i With the angular acceleration of gyrosensorω i Data;Wherein, angleθ i By three-dimensional space
Between geometrical relationship obtain, angular accelerationω i Obtained by sensor acquisition and process;
Second step:First smooth the angle of acceleration transducerθ i , obtain Smoothing angle intermediate quantityx 1, computing formula is:
In formula,θ i For this angle value;θ’ oFor angle output valve after previous correction,K 1It is a certain proportion of smooth to take
Weighted value;That is, the angle of smooth acceleration transducerθ i Process be:Using this angle valueθ i After previous correction
Angle output valveθ’ oDifference and (1-K 1)2It is multiplied and obtains Smoothing angle intermediate quantityx 1;
x 1After computing, being converted into can characterize fast-changing class angular acceleration amountx 2, computing formula is:
In formula, ΔtFor the sampling interval;x’ 2For previous class angular acceleration amount;That is, being converted into class angular accelerationx 2Process be:By Smoothing angle intermediate quantityx 1As integrand, make it be integrated the time, be only herein between sampling
Every ΔtBe multiplied, afterwards again with previous class angular acceleration amountx’ 2It is cumulative, that is, obtain this class angular acceleration amountx 2;
3rd step:First with the angular acceleration of gyrosensorω i The angle of compensated acceleration sensor, computing formula
For:
In formula,x 3For the angle intermediate quantity of calculated acceleration transducer;x 2For class angular acceleration;ω i For gyro
The angular acceleration of sensor;x 1For Smoothing angle intermediate quantity;K 2To take a certain proportion of compensation authority weight values;That is, compensation adds
The process of the angle of velocity sensor is:Class angular accelerationx 2With the angular acceleration of gyrosensorω i Be added, and with (1-K 2) after multiplication again with angle intermediate quantityx 1The process of addition, by this process angle intermediate quantity is obtainedx 3;
Then angle estimation is carried out, this attitude angle is finally givenθ o , computing formula is:
In formula, ΔtFor the sampling interval;θ’ iFor the angle of Iterative;That is, the process of angle estimation is:
By angle intermediate quantityx 3As integrand, make it be integrated the time, be only and sampling interval Δ hereintIt is multiplied, afterwards
Again with the angle of Iterativeθ’ iIt is cumulative, finally give this angleθ o Process.
In the above-mentioned first step, described pretreatment includes sensor data acquisition and original data processing.
In the above-mentioned first step, described acceleration transducer and gyrosensor needs to be placed in parallel, in order to eliminate
The impact of Coordinate Conversion computing, and acceleration transducer and gyrosensor can arbitrarily put in the inside of gyrostabilization unit
Put.
In above-mentioned second step, described smoothing weights valueK 1Span be 0~1.
In above-mentioned 3rd step, described compensation authority weight valuesK 2Span be 0~1.
The present invention compared with prior art, has the advantage that:
On the one hand, gyrostabilization unit proposed by the present invention, is applied to blind man's stick, can provide great stability, function
Module is connected by gyrostabilization unit with blind man's stick, and under being operated in a steady environment, this application improves blind man's stick
Practicality, range finder module improved as functional module, the investigative range for making blind man's stick, and the barrier in space can be by
Blind man's stick is accurately identified, and is greatly facilitated so as to bring blind person;
On the other hand, under the premise of service precision of the blind-guiding stick using gyrostabilization unit is met, the present invention proposes one
The complementary filter method for improving is planted, traditional complementary filter is compared and only letter is done to acceleration transducer and gyrosensor data
Single weighted average, this method is processed respectively for the different qualities of both sensors, i.e., larger to dynamic error
Acceierometer sensor introduce it is " smooth " process, and for the gyrosensor with accumulated time effect only takes the acceleration of its angle
Angle value is used for algorithm compensation, so as to the accumulated time of the dynamic error and gyrosensor that effectively reduce acceleration transducer is missed
Difference, is more efficiently utilized multi-sensor data to realize the fusion of information, therefore can be disappeared to greatest extent by this method
Except effect of noise, while simplifying computation complexity, response speed and detection accuracy are drastically increased.
Below in conjunction with the accompanying drawings the present invention is further illustrated.
Description of the drawings
Fig. 1 is the frame structure schematic diagram of the gyrostabilization unit for blind man's stick of the present invention.
Fig. 2 is the system schematic of complementary filter method of the present invention.
Fig. 3 is the ratio of class angular acceleration in the angle that present invention acceleration transducer data are calculated and complementary filter method
Relatively scheme.
Fig. 4 be acceleration transducer and conventional complementary filtering method with present invention improves over complementary filter method solve respectively
The comparison diagram of the angle of calculation.
Specific embodiment
As shown in figure 1, the gyrostabilization unit for blind man's stick of the present invention, including microcontroller and and microcontroller
The attitude transducer and motor of connection, wherein the attitude transducer includes acceleration transducer and gyrosensor, and it is described
Acceleration transducer and gyrosensor are three-axis sensor, the microcontroller be also associated with range finder module and for shifting
The wireless module of dynamic terminal signaling connection, and the range finder module is connected with motor, the motor be steering wheel, brushless electric machine,
Servo-controller or other motors.For the ease of eliminating the impact of Coordinate Conversion computing, the acceleration transducer and gyro are sensed
Device is arranged in parallel, but it can arbitrarily be placed in the inside of gyrostabilization unit.
The gyrostabilization unit is the device with any attitude in two-dimensional space, and only in four sides up and down
There is movement tendency position, and azimuth motion trend can be captured.The gyrostabilization unit is connected on blind man's stick and uses, by microcontroller
Device is controlled, and to range finder module a kind of more stable working environment is provided.Acceleration transducer and gyrosensor are used for blind man's stick appearance
The identification of state, identification parameter is mainly angle, is realized by complementary filter method;Complementary filter method of the present invention can be real
When resolve the angle of blind man's stick, and rotated using the angle motor;Motor is connected with range finder module, motor rotate angle with
Blind man's stick movement angle is offseted, to realize more smoothly finding range.Microcontroller can be handed over by wireless module with mobile terminal
Mutually, the data obtained of finding range can be transmitted to mobile terminal, to realize the intelligent barrier avoiding of aerial barrage.Described range finder module is mainly
Realize distance measurement function, it is also possible to be to aid in realizing the module of intelligent barrier avoiding.
The present invention has fit closely application in the gesture recognition control of the gyrostabilization unit of blind man's stick.For blind
The gyrostabilization unit of cane, in being mainly reflected in two-dimensional environment, i.e., only has movement tendency in four orientation up and down.Blind man's stick connects
Then the arm of people, can produce erratic vibrations, thus can produce random noise.Simultaneously as the characteristics of people uses blind man's stick
Determine that the gyrostabilization unit does not require very high precision, but to embody very fast response speed.Therefore, this
Bright proposed complementary filter method is applicable in the gyrostabilization unit of gesture recognition control.
The use of gyrostabilization unit of the present invention, and its complementary filter method are illustrated below by specific example
Feasibility and the effectiveness that embodied during actual test.
As shown in Fig. 2 carrying out substep description to complementary filter method:
The first step:Gyrosensor and acceleration transducer are arranged in parallel in inside gyrostabilization unit, acceleration
Sensor and gyrosensor belong to three-axis sensor, so as to gyrosensor each axle can and acceleration sensing
The axle of device points to same direction, while sensor is connected with data acquisition module, it is ensured that gyrostabilization unit can be
Freely rotate in two-dimensional environment;
Second step:Data are carried out with pretreatment, including sensor data acquisition and original data processing;Setting gyro sensing
The sample rate of device and acceleration transducer(In the present embodiment, representative value is 125Hz)With digital low pass filtered frequency(In this reality
In applying example, representative value is 5Hz), the acceleration of two axles of acceleration transducer is obtained respectivelya x Witha y , and gyrosensor
Angular accelerationω i Data;Using the typical angle of pitch and roll angle computing formula:θ=arcsin(a /g), can obtain corresponding
The angle of the corresponding acceleration transducer of axleθ ix Withθ iy , the angle of any one axle is taken here, it is calculated asθ i , as shown in Figure 3;
3rd step:Using the angle value of interative computationθ’ oSmooth this angle valueθ i , concrete calculating process is as follows:
Wherein,x 1For Smoothing angle intermediate quantity,x 2For this class angular acceleration amount,x ’ 2Resolving for interative computation is obtained
Previous class angular acceleration amount,θ’ oFor the angle value that the resolving of interative computation is obtained,K 1For smoothing weights value, the smoothing weights value
Span be 0~1, value in this exampleK 1=0.005;ΔtFor sampling interval, value Δ in this examplet=0.008 s;By
Fig. 3 can see that the curve after smoothing effective filters out mechanical noise and shaking interference compared with primitive curve, can be accurate
The class angular acceleration amount variation tendency of ground reflection acceleration transducer;
4th step:Using the angular acceleration of gyrosensorω i The angle of compensated acceleration sensor, and carry out angle and estimate
Meter, concrete calculating process is as follows:
Wherein,x 3For angle intermediate quantity,K 2For compensation authority weight values, the span of the compensation authority weight values is 0~1, in this example
InK 2=0.2,θ ’ i For the angle that the resolving of interative computation is obtained;ΔtFor sampling interval, value Δ in this examplet=0.008 s, most
This angle value is tried to achieve eventuallyθ o , current result of calculation will be preserved, and be subsequently used for improving the precision of angle estimation, by Fig. 4
Can see, the curve after compensation improves the smoothness of angle compared with primitive curve.
Traditional complementary filter method directly allows the data of acceleration transducer and gyrosensor to give certain power respectively
Value, cannot thus be prevented effectively from the dynamic error of acceleration transducer and the time integral cumulative error of gyrosensor, and one
As be not enough to obtain accurate angle information, also to carry out offset angle measurement error with reference to Kalman filtering.It is proposed by the present invention
The complementary filter method for improving, make use of the angle accuracy that acceleration transducer has in static measurement, and gyro to pass
The angular acceleration accuracy that sensor has on transient measurement is avoiding problem above.It is not directly to adopt acceleration transducer
Intermittent angle angle value and gyrosensor integration angle value directly merging, but by smoothing and compensating the method for combining
The fusion of information is realized, the time integral accumulation of the dynamic error and gyrosensor that efficiently reduce acceleration transducer is missed
Difference.
From actual application, algorithm provided by the present invention has preferably when blind man's stick gyrostabilization unit is applied to
Stability and response speed, and algorithm complex is not high, suppresses obvious to noise and drift.It can be seen from figure 4 that
Compared with the angle value that traditional complementary filter method is calculated, complementary filter method proposed by the present invention, not only in data more
With smoothness, while in the case of significantly angle change, also there is response speed quickly and accuracy.
The present invention is described by embodiment, but is not limited the invention, with reference to description of the invention, institute
Other changes of disclosed embodiment, such as the professional person of this area is readily apparent that, such change should belong to
Within the scope of the claims in the present invention are limited.
Claims (4)
1. a kind of complementary filter method of gyrostabilization unit, the gyrostabilization unit for blind man's stick in the method, including it is micro-
Controller and the attitude transducer being connected with microcontroller and motor, wherein the attitude transducer include acceleration transducer and
Gyrosensor, the microcontroller is also associated with range finder module and the wireless module for being connected with mobile terminal signal, and
The range finder module is connected with motor, and the motor is steering wheel, and the acceleration transducer and gyrosensor are used for blind man's stick
The identification of attitude, identification parameter is mainly angle, and the angle that the motor is rotated is offseted with blind man's stick movement angle, to realize more
For smoothly range finding, it is characterised in that comprise the following steps:
The first step:Acceleration transducer and gyrosensor are placed, gyrostabilization unit is built, after pretreatment, respectively
Obtain the angle of acceleration transducerθ i With the angular acceleration of gyrosensorω i Data;Wherein, angleθ i It is several by three dimensions
What relation is obtained, angular accelerationω i Obtained by sensor acquisition and process;
Second step:First smooth the angle of acceleration transducerθ i , obtain Smoothing angle intermediate quantityx 1, computing formula is:
In formula,θ i For this angle value;θ’ oFor angle output valve after previous correction,K 1To take a certain proportion of smoothing weights
Value;
x 1After computing, being converted into can characterize fast-changing class angular acceleration amountx 2, computing formula is:
In formula, ΔtFor the sampling interval;x’ 2For previous class angular acceleration amount;
3rd step:First with the angular acceleration of gyrosensorω i The angle of compensated acceleration sensor, computing formula is:
In formula,x 3For the angle intermediate quantity of calculated acceleration transducer;x 2For class angular acceleration;ω i For gyro sensing
The angular acceleration of device;x 1For Smoothing angle intermediate quantity;K 2To take a certain proportion of compensation authority weight values;
Then angle estimation is carried out, this attitude angle is finally givenθ o , computing formula is:
In formula, ΔtFor the sampling interval;θ’ iFor the angle of Iterative.
2. the complementary filter method of gyrostabilization unit according to claim 1, it is characterised in that in the above-mentioned first step,
Described acceleration transducer and gyrosensor needs to be placed in parallel, and in order to eliminate the impact of Coordinate Conversion computing, and adds
Velocity sensor and gyrosensor can be placed arbitrarily in the inside of gyrostabilization unit.
3. the complementary filter method of gyrostabilization unit according to claim 1, it is characterised in that in above-mentioned second step,
Described smoothing weights valueK 1Span be 0~1.
4. the complementary filter method of gyrostabilization unit according to claim 1, it is characterised in that in above-mentioned 3rd step,
Described compensation authority weight valuesK 2Span be 0~1.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510049666.0A CN104546391B (en) | 2015-01-31 | 2015-01-31 | Gyro stabilizer for tactile sticks and complementary filtering method thereof |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510049666.0A CN104546391B (en) | 2015-01-31 | 2015-01-31 | Gyro stabilizer for tactile sticks and complementary filtering method thereof |
Publications (2)
Publication Number | Publication Date |
---|---|
CN104546391A CN104546391A (en) | 2015-04-29 |
CN104546391B true CN104546391B (en) | 2017-05-17 |
Family
ID=53064188
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201510049666.0A Active CN104546391B (en) | 2015-01-31 | 2015-01-31 | Gyro stabilizer for tactile sticks and complementary filtering method thereof |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN104546391B (en) |
Families Citing this family (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105651242B (en) * | 2016-04-05 | 2018-08-24 | 清华大学深圳研究生院 | A method of fusion attitude angle is calculated based on complementary Kalman filtering algorithm |
CN107485540B (en) * | 2017-07-18 | 2019-12-13 | 浙江大学城市学院 | Energy injection system for intelligent walking stick |
CN109583511B (en) * | 2018-12-14 | 2023-06-30 | 北京百度网讯科技有限公司 | Speed fusion method and device |
CN112473097B (en) * | 2019-09-11 | 2022-04-01 | Tcl科技集团股份有限公司 | Mountain climbing assisting method, server, system and storage medium |
CN110974641A (en) * | 2019-12-24 | 2020-04-10 | 中南民族大学 | Intelligent walking stick system integrating machine learning and Internet of things technology for blind people |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102519425A (en) * | 2011-10-24 | 2012-06-27 | 哈尔滨工程大学 | Laser range finder stabilized platform used for vessel with single degree of freedom and its control method |
CN203519011U (en) * | 2013-10-15 | 2014-04-02 | 顾捷 | Attitude sensor |
CN103913171A (en) * | 2014-04-28 | 2014-07-09 | 北京理工大学 | Compensation measurement method for rotary speed and rotary angle of carrier based on accelerometer and gyroscope |
CN104068997A (en) * | 2014-06-30 | 2014-10-01 | 北京信息科技大学 | Multifunctional walking stick for blind |
CN203898662U (en) * | 2014-04-11 | 2014-10-29 | 上海工程技术大学 | Guide walking stick system for blind person |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6668846B2 (en) * | 2001-11-15 | 2003-12-30 | Edward L. Meador | Gyroscopically balanced walking cane |
-
2015
- 2015-01-31 CN CN201510049666.0A patent/CN104546391B/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102519425A (en) * | 2011-10-24 | 2012-06-27 | 哈尔滨工程大学 | Laser range finder stabilized platform used for vessel with single degree of freedom and its control method |
CN203519011U (en) * | 2013-10-15 | 2014-04-02 | 顾捷 | Attitude sensor |
CN203898662U (en) * | 2014-04-11 | 2014-10-29 | 上海工程技术大学 | Guide walking stick system for blind person |
CN103913171A (en) * | 2014-04-28 | 2014-07-09 | 北京理工大学 | Compensation measurement method for rotary speed and rotary angle of carrier based on accelerometer and gyroscope |
CN104068997A (en) * | 2014-06-30 | 2014-10-01 | 北京信息科技大学 | Multifunctional walking stick for blind |
Also Published As
Publication number | Publication date |
---|---|
CN104546391A (en) | 2015-04-29 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106705968B (en) | Indoor inertial navigation algorithm based on attitude identification and step size model | |
CN108362282B (en) | Inertial pedestrian positioning method based on self-adaptive zero-speed interval adjustment | |
CN104546391B (en) | Gyro stabilizer for tactile sticks and complementary filtering method thereof | |
CN104121905B (en) | Course angle obtaining method based on inertial sensor | |
CN110398245B (en) | Indoor pedestrian navigation attitude estimation method based on foot-worn inertial measurement unit | |
CN107490378B (en) | Indoor positioning and navigation method based on MPU6050 and smart phone | |
CN106908060A (en) | A kind of high accuracy indoor orientation method based on MEMS inertial sensor | |
CN105698822B (en) | Initial Alignment Method between autonomous type inertial navigation based on reversed Attitude Tracking is advanced | |
CN112697138B (en) | Bionic polarization synchronous positioning and composition method based on factor graph optimization | |
CN102445200A (en) | Microminiature personal combined navigation system as well as navigating and positioning method thereof | |
CN104061934A (en) | Pedestrian indoor position tracking method based on inertial sensor | |
CN107941217A (en) | A kind of robot localization method, electronic equipment, storage medium, device | |
EP2951533A1 (en) | Method and system for varying step length estimation using nonlinear system identification | |
CN108132053B (en) | Pedestrian track construction method and system and inertia measurement device | |
CN106370178B (en) | Attitude measurement method and device of mobile terminal equipment | |
CN110057356B (en) | Method and device for positioning vehicles in tunnel | |
CN107255474A (en) | A kind of PDR course angles of fusion electronic compass and gyroscope determine method | |
CN109459028A (en) | A kind of adaptive step estimation method based on gradient decline | |
CN106403952A (en) | Method for measuring combined attitudes of Satcom on the move with low cost | |
CN110672095A (en) | Pedestrian indoor autonomous positioning algorithm based on micro inertial navigation | |
US10466054B2 (en) | Method and system for estimating relative angle between headings | |
CN106643713B (en) | Estimation method and device for zero-speed correction pedestrian locus for smooth and adaptive adjustment of threshold | |
CN112362057A (en) | Inertial pedestrian navigation algorithm based on zero-speed correction and attitude self-observation | |
CN108444468A (en) | The bearing compass of vision and inertial navigation information is regarded under a kind of fusion | |
CN109781096A (en) | A kind of integrated navigation and location system and method for intelligent agricultural machinery |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |