CN112923922B - Method, system and storage medium for counting steps and determining position information of pedestrian - Google Patents
Method, system and storage medium for counting steps and determining position information of pedestrian Download PDFInfo
- Publication number
- CN112923922B CN112923922B CN202110241308.5A CN202110241308A CN112923922B CN 112923922 B CN112923922 B CN 112923922B CN 202110241308 A CN202110241308 A CN 202110241308A CN 112923922 B CN112923922 B CN 112923922B
- Authority
- CN
- China
- Prior art keywords
- data
- output data
- pedestrian
- historical output
- inertial sensor
- 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
Images
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
- G01C21/10—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 by using measurements of speed or acceleration
- G01C21/12—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 by using measurements of speed or acceleration executed aboard the object being navigated; Dead reckoning
- G01C21/16—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 by using measurements of speed or acceleration executed aboard the object being navigated; Dead reckoning by integrating acceleration or speed, i.e. inertial navigation
- G01C21/165—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 by using measurements of speed or acceleration executed aboard the object being navigated; Dead reckoning by integrating acceleration or speed, i.e. inertial navigation combined with non-inertial navigation instruments
-
- 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
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C22/00—Measuring distance traversed on the ground by vehicles, persons, animals or other moving solid bodies, e.g. using odometers, using pedometers
- G01C22/006—Pedometers
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/044—Recurrent networks, e.g. Hopfield networks
Abstract
The invention discloses a method, a system and a storage medium for counting steps and determining position information of pedestrians. And generating a step number model according to the sample data and the truth value data. And finally, determining the total step number of the pedestrian and the position information of the pedestrian according to the step number model. The method solves the problem that the calculated position information is inaccurate due to the defects of noise interference and insufficient adaptivity when the pedestrian dead reckoning algorithm measures steps by comparing the output data of the inertial sensor with the experience threshold in the prior art.
Description
Technical Field
The invention relates to the technical field of positioning, in particular to a method, a system and a storage medium for counting steps and determining position information of pedestrians.
Background
The PDR (Pedestrian Dead Reckoning) is a simple and reliable Pedestrian positioning and navigation mode, and errors mainly come from three aspects, namely statistical errors of step numbers, step length errors and course angle errors, wherein the statistical errors of the step numbers have the largest influence on navigation accuracy. At present, the PDR algorithm mainly adopts a traditional step counting mode, namely, the step counting is carried out by comparing the output of an accelerometer or a gyroscope with an empirical threshold, and the PDR algorithm has the defects of easy noise interference, insufficient self-adaptability and the like.
Thus, there is still a need for improvement and development of the prior art.
Disclosure of Invention
The present invention is directed to a method, a system, and a storage medium for counting steps and determining position information of a pedestrian, which are provided to solve the problem in the prior art that an estimated position information is inaccurate due to the defects of noise interference and insufficient adaptivity when a pedestrian dead reckoning algorithm performs step counting by comparing output data of an inertial sensor with an empirical threshold.
The technical scheme adopted by the invention for solving the problems is as follows:
in a first aspect, an embodiment of the present invention provides a method for counting steps and determining location information of a pedestrian, where the method includes:
acquiring historical output data of an inertial sensor and historical output data of a pedometer, generating sample data according to the historical output data of the inertial sensor, and generating truth data according to the historical output data of the pedometer;
generating a step number model according to the sample data and the truth value data;
and determining the total steps of the pedestrian and the position information of the pedestrian according to the step number model.
In one embodiment, the acquiring historical output data of an inertial sensor and historical output data of a pedometer, generating sample data based on the historical output data of the inertial sensor, and generating truth data based on the historical output data of the pedometer includes:
acquiring historical output data of an inertial sensor and time information corresponding to the historical output data of the inertial sensor, and taking the time information corresponding to the historical output data of the inertial sensor as first time information;
acquiring historical output data of a pedometer and time information corresponding to the historical output data of the pedometer, and taking the time information corresponding to the historical output data of the pedometer as second time information;
performing data reconstruction operation on historical output data of the inertial sensor, and taking data obtained after the data reconstruction operation is finished as reconstruction data;
aligning the reconstructed data and the historical output data of the pedometer according to the first time information and the second time information;
and taking the reconstructed data obtained after the alignment operation as sample data, and taking the historical output data of the pedometer obtained after the alignment operation as truth data.
In one embodiment, the performing a data reconstruction operation on the historical output data of the inertial sensor, and using the modified data as the reconstructed data includes:
inputting historical output data of the inertial sensor into a preset algorithm;
when the stability of the historical output data of the inertial sensor meets a preset condition is determined through the preset algorithm, obtaining a hysteresis coefficient generated by the preset algorithm based on the historical output data of the inertial sensor, and taking the hysteresis coefficient as time step data;
and modifying the data structure of the historical output data of the inertial sensor according to the time step data, and taking the data obtained after modification as reconstruction data.
In one embodiment, said generating a step number model from said sample data and said true value data comprises:
dividing the sample data into a training data set and a test data set;
performing model fitting according to the training data set and the truth value data to obtain an initial step number model;
evaluating the judging effect of the initial step number model according to the test data set and the truth value data and generating feedback information;
and adjusting the initial step number model according to the feedback information to obtain a step number model.
In one embodiment, the determining the total number of steps of the pedestrian and the position information of the pedestrian according to the step number model includes:
acquiring output data of the inertial sensor, inputting the output data of the inertial sensor into the step number model, and acquiring a judgment value output by the step number model based on the output data of the inertial sensor;
obtaining detection information of the number of steps of the pedestrian according to the judgment value;
counting the total steps of the pedestrian according to the detection information of the steps of the pedestrian;
and acquiring course information, and determining the position information of the pedestrian according to the detection information of the step number of the pedestrian and the course information.
In one embodiment, the obtaining of the detection information of the number of steps of the pedestrian according to the determination value includes:
acquiring a preset threshold value, and comparing the judgment value with the preset threshold value;
when the judgment value is larger than the preset threshold value, determining the detection information of the number of steps of the pedestrian as the detected number of steps;
and when the judgment value is smaller than or equal to the preset threshold value, determining that the detection information of the step number of the pedestrian is the undetected step number.
In one embodiment, the counting the total number of steps of the pedestrian according to the detection information of the number of steps of the pedestrian includes:
when the detection information of the number of steps of the pedestrian is determined to be the detected number of steps, increasing the value of a preset counter by 1;
and acquiring the numerical value of the preset counter, and taking the numerical value of the preset counter as the total step number of the pedestrian.
In one embodiment, the obtaining the heading information and determining the position information of the pedestrian according to the detection information of the number of steps of the pedestrian and the heading information includes:
when the detection information of the step number of the pedestrian is determined to be the detected step number, acquiring course information;
acquiring historical position information of a pedestrian and a step value of the pedestrian at the previous moment;
and inputting the course information, the historical position information and the step value into a preset pedestrian navigation algorithm, and acquiring the position information of the pedestrian output by the pedestrian navigation algorithm based on the course information, the historical position information and the step value.
In a second aspect, an embodiment of the present invention further provides a system for counting steps and determining location information of a pedestrian, where the system includes:
the acquisition module is used for acquiring historical output data of an inertial sensor and historical output data of a pedometer, generating sample data according to the historical output data of the inertial sensor and generating true value data according to the historical output data of the pedometer;
the generating module is used for generating a step number model according to the sample data and the true value data;
and the determining module is used for determining the total step number of the pedestrian and the position information of the pedestrian according to the step number model.
In a third aspect, the present invention further provides a computer-readable storage medium, on which a plurality of instructions are stored, wherein the instructions are adapted to be loaded and executed by a processor to implement any of the steps of the method for counting steps and determining position information of a pedestrian.
The invention has the beneficial effects that: according to the embodiment of the invention, the fitting and training processes of the model are completed by deep learning the historical output data of the inertial sensor and the pedometer, and the model capable of accurately counting steps is obtained. And acquiring the step number information of the pedestrian through the model and determining the position information of the pedestrian. The method solves the problem that the calculated position information is inaccurate due to the defects of noise interference and insufficient adaptivity when the pedestrian dead reckoning algorithm measures steps by comparing the output data of the inertial sensor with the experience threshold in the prior art.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments described in the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
Fig. 1 is a schematic flow chart of a method for counting steps and determining position information of a pedestrian according to an embodiment of the present invention.
Fig. 2 is a reference diagram of a training process and a using process of the step number model provided by the embodiment of the invention.
Fig. 3 is a reference diagram of the internal structure of the LSTM model provided by the embodiment of the present invention.
FIG. 4 is a block diagram of the internal components of a system for step counting and pedestrian location information determination according to an embodiment of the present invention.
Fig. 5 is a schematic block diagram of a terminal according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention clearer and clearer, the present invention is further described in detail below with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
It should be noted that, if directional indications (such as up, down, left, right, front, and back … …) are involved in the embodiment of the present invention, the directional indications are only used to explain the relative positional relationship between the components, the movement situation, and the like in a specific posture (as shown in the drawing), and if the specific posture is changed, the directional indications are changed accordingly.
With the increasingly complex structures of buildings such as cities, crescents, malls and the like, people rely on navigation systems such as mobile phones for positioning and navigation, and pedestrian navigation becomes an indispensable technology in modern life gradually, and is a key field of navigation research. At present, many positioning technologies need to deploy a large number of beacon nodes around or inside a building in advance, and PDRs (Pedestrian Dead Reckoning) do not need to pre-install beacon nodes in advance, and only need to calculate the step length and direction of a person by using inertial sensors (such as an acceleration sensor, a gyroscope, a magnetometer and the like), the trace of the Pedestrian can be estimated, so that the positioning technologies are widely applied.
Specifically, the PDR is a relative positioning algorithm, which estimates the position information of the next time by assuming the known initial position information, and then deduces the position information of each subsequent time. There are two key factors in the PDR calculation process, one is the walking displacement and one is the heading angle. The walking displacement can be estimated through the number of steps and the step length, and the direction angle can be obtained through a heading reference system in the inertial sensor or a preset direction sensor. Therefore, the error of the PDR algorithm mainly originates from three aspects, one is the statistical error of the step number, the other is the error of the step size, and the last is the error of the heading angle. Wherein, the statistical error of the step number has the biggest influence on the navigation precision. The traditional step counting mode is mainly judged by comparing the output of an accelerometer or a gyroscope with an empirical threshold, and has the defects of easy noise interference, insufficient self-adaptability and the like.
Aiming at the defects in the prior art, the invention provides a method for counting steps and determining the position information of pedestrians, which completes the fitting and training process of a model by deep learning the historical output data of an inertial sensor and a pedometer and obtains the model capable of accurately counting steps. And acquiring the step number information of the pedestrian through the model and determining the position information of the pedestrian. The method solves the problem that the calculated position information is inaccurate due to the defects of noise interference and insufficient adaptivity when the pedestrian dead reckoning algorithm measures steps by comparing the output data of the inertial sensor with the experience threshold in the prior art.
As shown in fig. 1, the present embodiment provides a method of counting steps and determining position information of a pedestrian, the method comprising the steps of:
step S100, obtaining historical output data of an inertial sensor and historical output data of a pedometer, generating sample data according to the historical output data of the inertial sensor, and generating truth value data according to the historical output data of the pedometer.
In order to realize real-time and accurate statistics of the number of steps of the pedestrian and determination of the position information of the pedestrian, a deep learning technology is required in the embodiment. In brief, the deep learning technology is a technology that enables a machine to have an analytic learning ability like a human by learning the intrinsic rules and the expression levels of sample data. Therefore, first, the embodiment needs to acquire sample data that helps the machine perform deep learning. Specifically, in this embodiment, it is necessary to obtain historical output data of the inertial sensor and historical output data of the pedometer, generate sample data and the true value data according to the two data, and perform deep learning on the model through the sample data and the true value data, so that the model obtains a rule between the output of the inertial sensor and the number of steps generated by the pedestrian. In one implementation, the inertial sensor used in this embodiment may include an accelerometer and a gyroscope, and it is understood that this embodiment is not limited to a specific type of the inertial sensor.
In one implementation, the step S100 specifically includes the following steps:
step S110, acquiring historical output data of an inertial sensor and time information corresponding to the historical output data of the inertial sensor, and taking the time information corresponding to the historical output data of the inertial sensor as first time information;
step S120, acquiring historical output data of the pedometer and time information corresponding to the historical output data of the pedometer, and taking the time information corresponding to the historical output data of the pedometer as second time information;
step S130, performing data reconstruction operation on historical output data of the inertial sensor, and taking data obtained after the data reconstruction operation is finished as reconstruction data;
step S140, aligning the reconstructed data and the historical output data of the pedometer according to the first time information and the second time information;
and S150, taking the reconstructed data obtained after the alignment operation as sample data, and taking the historical output data of the pedometer obtained after the alignment operation as true value data.
Specifically, after acquiring the historical output data of the inertial sensor and the historical output data of the pedometer, the present embodiment needs to acquire the time information attached to each of the two types of data, and when the two types of data lack the time information, the time information may be supplemented to the data by means of time stamping. For the convenience of distinction, in the present embodiment, time information corresponding to the historical output data of the inertial sensor is named as first time information, and time information corresponding to the historical output data of the pedometer is named as second time information. Since the historical output data of the inertial sensor and the pedometer need to be input into the deep learning model at the same time, the two historical output data also need to be converted into a data format executable by the deep learning model, that is, the historical output data of the inertial sensor and the pedometer need to be preprocessed. Specifically, the present embodiment first requires a data reconstruction operation on the historical output data of the inertial sensor. The data reconstruction operation process specifically comprises the following steps: inputting the historical output data of the inertial sensor into a preset algorithm, and judging the stability of the historical output data of the inertial sensor through the preset algorithm. And when the stability of the historical output data of the inertial sensor meets a preset requirement, taking a hysteresis coefficient generated by the preset algorithm based on the time sequence data as time step data, modifying the data structure of the historical output data of the inertial sensor according to the time step data, and taking the data obtained after modification as reconstruction data.
For example, the present embodiment mainly uses ADF (extended Dickey-Fuller) to confirm the stability of the historical output data of the inertial sensor, and the specific formula is as follows:
where Δ represents the difference, yt is the time series data to be detected (i.e., the historical output data of the inertial sensor), t is the time, α is a constant, β is a time trend coefficient, p is a hysteresis coefficient, and ε t is white noise. The ADF method determines whether a unit root exists by determining whether the coefficient γ is 0, and further determines the stability of the time-series data from the unit root: if γ is 0, the unit root does not exist, and the time-series data is stable.
And when the historical output data of the inertial sensor is determined to be stable, taking the hysteresis coefficient as time step data, and modifying the data structure of the historical output data of the inertial sensor according to the time step data. Specifically, in this embodiment, the historical output data of the inertial sensor needs to be arranged into p rows and 6 columns of tensors according to the hysteresis coefficient p to obtain reconstructed data. In summary, the time step data is mainly used for intercepting the historical output data of the inertial sensor, in practical application, the historical output data of the inertial sensor is usually a long time sequence, and the function of the time step is to determine how long the sequence needs to be intercepted in the long time sequence as the processing data, for example, the length of the time sequence is 300, and the time step is 50, which means that the part with the length of 50 needs to be intercepted in the time sequence with the length of 300 as the processing data. And then, aligning the reconstructed data and the historical output data of the pedometer according to the first time information and the second time information, taking the reconstructed data obtained after the alignment operation as sample data, and taking the historical output data of the pedometer as true value data.
After generating the sample data and the truth data, as shown in fig. 1, the method further includes the following steps:
and S200, generating a step number model according to the sample data and the true value data.
As shown in fig. 2, the step number model in this embodiment is a deep learning model, and the structure and parameters of the model can be continuously optimized by learning the sample data and the true value data, so as to achieve the technical effect of predicting the step number of the pedestrian in real time. In one implementation, the step number model may adopt an LSTM model, where the LSTM model is an RNN (Recurrent Neural Networks) model with long and short term memory, a drop layer of the LSTM model is used to randomly delete some nodes in a Neural network, and a dense layer is used to output one-dimensional tensor information. Fig. 3 shows a specific structure of the LSTM model, and assuming that the inertial sensors are three-dimensional accelerometer and gyroscope, the input of the LSTM model is a 50 × 6 two-dimensional tensor, where 50 is the time step, i.e. the above-mentioned hysteresis coefficient p, and 6 is the dimension of the input data, i.e. the total dimension of the outputs of the three-dimensional accelerometer and gyroscope.
In one implementation, the step S200 specifically includes the following steps:
step S210, dividing the sample data into a training data set and a test data set;
step S220, performing model fitting according to the training data set and the truth value data to obtain an initial step number model;
step S230, evaluating the judgment effect of the initial step number model according to the test data set and the truth value data and generating feedback information;
and S240, adjusting the initial step number model according to the feedback information to obtain a step number model.
Specifically, the present embodiment divides the sample data into two types, one type is data in the training data set, the other type is data in the test data set, and both types of data have corresponding true value data. Firstly, model fitting is carried out through data in a training data set and true value data corresponding to the data to obtain an initial step number model. In order to optimize the model and avoid the overfitting of the model, after the initial step number model is fitted, the judgment effect of the initial step number model needs to be evaluated according to the data in the test data set and the corresponding truth value data. In one implementation, the criterion for evaluation may employ a P/R curve, where P represents the accuracy, i.e., the proportion of the correct objects detected to the total objects detected; r represents recall or recall, i.e. the proportion of detected objects to all objects that need to be detected. When the P and the R are closer to 1, it indicates that the prediction effect of the initial step number model is better, otherwise, it indicates that the prediction effect of the initial step number model is worse, so as to generate feedback information, and then return to the step of model fitting according to the feedback information to adjust the initial step number model, for example, adjust the parameters of the initial step number model or adjust the structure of the initial step number model, so as to obtain the final step number model. In one implementation, the initial step model may also be optimally adjusted by changing the value of the time step. In addition, when the output data of the inertial sensor is unstable, the time step can be adjusted by evaluating the initial step number model by the test data set.
After obtaining the trained step number model, as shown in fig. 1, the method further includes the following steps:
and step S300, determining the total step number of the pedestrian and the position information of the pedestrian according to the step number model.
Since the step number model in this embodiment has learned the distributed feature representation of a large amount of historical output data of the inertial sensors and the pedometer, when the current output data of the inertial sensors is input into the step number model again, the step number model can judge whether the pedestrian generates a new step number according to the current output data of the inertial sensors, and then count the step number of the pedestrian to obtain the total step number of the pedestrian. Furthermore, the output of the step number model may be combined with a pedestrian navigation algorithm to determine the current position of the pedestrian.
In one implementation, the step S300 specifically includes the following steps:
step S310, acquiring output data of the inertial sensor, inputting the output data of the inertial sensor into the step number model, and acquiring a judgment value output by the step number model based on the output data of the inertial sensor;
step S320, obtaining the detection information of the step number of the pedestrian according to the judgment value;
step S330, counting the total steps of the pedestrian according to the detection information of the steps of the pedestrian;
and step S340, acquiring course information, and determining the position information of the pedestrian according to the detection information of the step number of the pedestrian and the course information.
In order to determine the latest step information of the pedestrian, the present embodiment needs to acquire the output data of the inertial sensor in real time, input the output data of the inertial sensor into the step number model, and then acquire the determination value calculated by the step number model based on the input data, which may indicate whether or not the newly generated step number of the pedestrian is detected, and generate the detection information of the step number of the pedestrian. Specifically, a threshold is preset in this embodiment, after a judgment value output by the step number model is obtained, the judgment value is compared with the threshold, and when the judgment value is greater than the preset threshold, the detection information of the step number of the pedestrian is determined as the detected step number; and when the judgment value is smaller than or equal to the preset threshold value, determining that the detection information of the step number of the pedestrian is the undetected step number. Therefore, according to the detection information of the number of steps of the pedestrian, the total number of steps of the pedestrian can be counted step by step: in this embodiment, a counter is preset, and when it is determined that the detection information of the number of steps of the pedestrian is the detected number of steps, the value of the preset counter is increased by 1, and finally the value of the preset counter is obtained, which is the total number of steps of the pedestrian.
In addition, as shown in fig. 2, the present embodiment may further acquire heading information, and determine the position information of the pedestrian according to the detection information of the number of steps of the pedestrian and the heading information. Specifically, when the detection information of the number of steps of the pedestrian is the detected number of steps, the course information is acquired. In an implementation manner, the heading information may be obtained by a heading reference system of the inertial sensor, and the embodiment does not limit a specific manner of obtaining the heading information. And then acquiring historical position information of the pedestrian and the step value of the pedestrian at the previous moment, inputting the course information, the historical position information and the step value into a preset pedestrian navigation algorithm, and finally acquiring the position information of the pedestrian, which is output by the pedestrian navigation algorithm based on the course information, the historical position information and the step value.
For example, assuming that the position of the pedestrian at time 0 is (0,0), i.e. the initial position, if the step number is determined to be detected according to the determination value output by the step number model at time k (k ≧ 1), the position information at time k is updated by using a pedestrian navigation algorithm, specifically, the following formula is as follows:
wherein, (xk-1, yk-1) is two-dimensional position information of x and y axes at the time of k-1, ψ k is heading information (heading angle) at the time of k, and dk is step length at the time of k (dk can use fixed value, also can use adaptive step length, if using fixed step length, can use general adult step length, about 0.5m, also can make concrete measurement).
Based on the above embodiment, the present invention further provides a system for counting steps and determining location information of a pedestrian, as shown in fig. 4, the system comprising:
the acquisition module 01 is used for acquiring historical output data of an inertial sensor and historical output data of a pedometer, generating sample data according to the historical output data of the inertial sensor, and generating truth value data according to the historical output data of the pedometer;
a generating module 02, configured to generate a step number model according to the sample data and the true value data;
and the determining module 03 is configured to determine the total number of steps of the pedestrian and the position information of the pedestrian according to the number of steps model.
Based on the above embodiments, the present invention further provides a terminal, and a schematic block diagram thereof may be as shown in fig. 5. The terminal comprises a processor, a memory, a network interface and a display screen which are connected through a system bus. Wherein the processor of the terminal is configured to provide computing and control capabilities. The memory of the terminal comprises a nonvolatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of an operating system and computer programs in the non-volatile storage medium. The network interface of the terminal is used for connecting and communicating with an external terminal through a network. The computer program is executed by a processor to implement a method of counting steps and determining location information of a pedestrian. The display screen of the terminal can be a liquid crystal display screen or an electronic ink display screen.
It will be appreciated by those skilled in the art that the block diagram of fig. 5 is only a block diagram of a portion of the structure associated with the inventive arrangements and does not constitute a limitation of the terminal to which the inventive arrangements are applied, and that a particular terminal may include more or less components than those shown, or may combine certain components, or have a different arrangement of components.
In one implementation, one or more programs are stored in a memory of the terminal and configured to be executed by one or more processors include instructions for performing a method of counting steps and determining location information of pedestrians.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by hardware instructions of a computer program, which can be stored in a non-volatile computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. Any reference to memory, storage, databases, or other media used in embodiments provided herein may include non-volatile and/or volatile memory. Non-volatile memory can include read-only memory (ROM), Programmable ROM (PROM), Electrically Programmable ROM (EPROM), Electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), Dynamic RAM (DRAM), Synchronous DRAM (SDRAM), Double Data Rate SDRAM (DDRSDRAM), Enhanced SDRAM (ESDRAM), Synchronous Link DRAM (SLDRAM), Rambus Direct RAM (RDRAM), direct bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).
In summary, the present invention discloses a method, a system and a storage medium for step counting and pedestrian position information determination, wherein historical output data of an inertial sensor and historical output data of a pedometer are obtained, sample data is generated according to the historical output data of the inertial sensor, and true value data is generated according to the historical output data of the pedometer. And generating a step number model according to the sample data and the truth value data. And finally, determining the total step number of the pedestrian and the position information of the pedestrian according to the step number model. The method solves the problem that the calculated position information is inaccurate due to the defects of noise interference and insufficient adaptivity when the pedestrian dead reckoning algorithm measures steps by comparing the output data of the inertial sensor with the experience threshold in the prior art.
It is to be understood that the invention is not limited to the examples described above, but that modifications and variations may be effected thereto by those of ordinary skill in the art in light of the foregoing description, and that all such modifications and variations are intended to be within the scope of the invention as defined by the appended claims.
Claims (8)
1. A method of step counting and determining location information of a pedestrian, the method comprising:
acquiring historical output data of an inertial sensor and historical output data of a pedometer, generating sample data according to the historical output data of the inertial sensor, and generating truth data according to the historical output data of the pedometer;
generating a step number model according to the sample data and the truth value data;
determining the total steps of the pedestrians and the position information of the pedestrians according to the step number model;
the acquiring historical output data of the inertial sensor and historical output data of the pedometer, generating sample data according to the historical output data of the inertial sensor, and generating true value data according to the historical output data of the pedometer comprises:
acquiring historical output data of an inertial sensor and time information corresponding to the historical output data of the inertial sensor, and taking the time information corresponding to the historical output data of the inertial sensor as first time information;
acquiring historical output data of a pedometer and time information corresponding to the historical output data of the pedometer, and taking the time information corresponding to the historical output data of the pedometer as second time information;
performing data reconstruction operation on historical output data of the inertial sensor, and taking data obtained after the data reconstruction operation is finished as reconstruction data;
aligning the reconstructed data and the historical output data of the pedometer according to the first time information and the second time information;
taking the reconstructed data obtained after the alignment operation as sample data, and taking the historical output data of the pedometer obtained after the alignment operation as truth data;
the performing data reconstruction operation on the historical output data of the inertial sensor, and taking the data obtained after modification as reconstruction data includes:
inputting historical output data of the inertial sensor into a preset algorithm;
when the stability of the historical output data of the inertial sensor meets a preset condition is determined through the preset algorithm, obtaining a hysteresis coefficient generated by the preset algorithm based on the historical output data of the inertial sensor, and taking the hysteresis coefficient as time step data;
and modifying the data structure of the historical output data of the inertial sensor according to the time step data, and taking the data obtained after modification as reconstruction data.
2. The method of claim 1, wherein the generating a step count model from the sample data and the truth data comprises:
dividing the sample data into a training data set and a test data set;
performing model fitting according to the training data set and the truth value data to obtain an initial step number model;
evaluating the judging effect of the initial step number model according to the test data set and the truth value data and generating feedback information;
and adjusting the initial step number model according to the feedback information to obtain a step number model.
3. The method of claim 1, wherein determining the total number of steps and the position information of the pedestrian according to the step number model comprises:
acquiring output data of the inertial sensor, inputting the output data of the inertial sensor into the step number model, and acquiring a judgment value output by the step number model based on the output data of the inertial sensor;
obtaining detection information of the number of steps of the pedestrian according to the judgment value;
counting the total steps of the pedestrian according to the detection information of the steps of the pedestrian;
and acquiring course information, and determining the position information of the pedestrian according to the detection information of the step number of the pedestrian and the course information.
4. The method of claim 3, wherein the obtaining of the detection information of the number of steps of the pedestrian according to the determination value comprises:
acquiring a preset threshold value, and comparing the judgment value with the preset threshold value;
when the judgment value is larger than the preset threshold value, determining the detection information of the number of steps of the pedestrian as the detected number of steps;
and when the judgment value is smaller than or equal to the preset threshold value, determining that the detection information of the step number of the pedestrian is the undetected step number.
5. The method of claim 4, wherein the step counting and pedestrian position information determining method according to the detection information of the pedestrian step count comprises:
when the detection information of the number of steps of the pedestrian is determined to be the detected number of steps, increasing the value of a preset counter by 1;
and acquiring the numerical value of the preset counter, and taking the numerical value of the preset counter as the total step number of the pedestrian.
6. The method of claim 4, wherein the step counting and pedestrian position information determining step count and course information comprises:
when the detection information of the step number of the pedestrian is determined to be the detected step number, acquiring course information;
acquiring historical position information of a pedestrian and a step value of the pedestrian at the previous moment;
and inputting the course information, the historical position information and the step value into a preset pedestrian navigation algorithm, and acquiring the position information of the pedestrian output by the pedestrian navigation algorithm based on the course information, the historical position information and the step value.
7. A system for counting steps and determining location information of a pedestrian, the system comprising:
the acquisition module is used for acquiring historical output data of an inertial sensor and historical output data of a pedometer, generating sample data according to the historical output data of the inertial sensor and generating true value data according to the historical output data of the pedometer;
the acquiring historical output data of the inertial sensor and historical output data of the pedometer, generating sample data according to the historical output data of the inertial sensor, and generating true value data according to the historical output data of the pedometer comprises:
acquiring historical output data of an inertial sensor and time information corresponding to the historical output data of the inertial sensor, and taking the time information corresponding to the historical output data of the inertial sensor as first time information;
acquiring historical output data of a pedometer and time information corresponding to the historical output data of the pedometer, and taking the time information corresponding to the historical output data of the pedometer as second time information;
performing data reconstruction operation on historical output data of the inertial sensor, and taking data obtained after the data reconstruction operation is finished as reconstruction data;
aligning the reconstructed data and the historical output data of the pedometer according to the first time information and the second time information;
taking the reconstructed data obtained after the alignment operation as sample data, and taking the historical output data of the pedometer obtained after the alignment operation as truth data;
the performing data reconstruction operation on the historical output data of the inertial sensor, and taking the data obtained after modification as reconstruction data includes:
inputting historical output data of the inertial sensor into a preset algorithm;
when the stability of the historical output data of the inertial sensor meets a preset condition is determined through the preset algorithm, obtaining a hysteresis coefficient generated by the preset algorithm based on the historical output data of the inertial sensor, and taking the hysteresis coefficient as time step data;
modifying the data structure of the historical output data of the inertial sensor according to the time step data, and taking the data obtained after modification as reconstruction data;
the generating module is used for generating a step number model according to the sample data and the true value data;
and the determining module is used for determining the total step number of the pedestrian and the position information of the pedestrian according to the step number model.
8. A computer readable storage medium having stored thereon a plurality of instructions adapted to be loaded and executed by a processor to perform the steps of a method of step counting and determining location information of a pedestrian according to any one of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110241308.5A CN112923922B (en) | 2021-03-04 | 2021-03-04 | Method, system and storage medium for counting steps and determining position information of pedestrian |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110241308.5A CN112923922B (en) | 2021-03-04 | 2021-03-04 | Method, system and storage medium for counting steps and determining position information of pedestrian |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112923922A CN112923922A (en) | 2021-06-08 |
| CN112923922B true CN112923922B (en) | 2022-01-25 |
Family
ID=76173391
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110241308.5A Active CN112923922B (en) | 2021-03-04 | 2021-03-04 | Method, system and storage medium for counting steps and determining position information of pedestrian |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112923922B (en) |
Family Cites Families (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106525068A (en) * | 2016-11-08 | 2017-03-22 | 深圳市金立通信设备有限公司 | Step-counting method and terminal |
| CN107462258B (en) * | 2017-07-13 | 2020-06-16 | 河海大学 | Step counting method based on mobile phone three-axis acceleration sensor |
| CN107727110A (en) * | 2017-09-11 | 2018-02-23 | 上海斐讯数据通信技术有限公司 | The statistical method and device of a kind of step number |
| CN107909677B (en) * | 2017-11-28 | 2020-08-07 | 中国船舶工业系统工程研究院 | Method for extracting and judging stability of operation data of ship electromechanical equipment |
| CN108399248A (en) * | 2018-03-02 | 2018-08-14 | 郑州云海信息技术有限公司 | A kind of time series data prediction technique, device and equipment |
| CN110811578A (en) * | 2019-11-27 | 2020-02-21 | 青岛歌尔智能传感器有限公司 | Step counting device and step counting method thereof, controller and readable storage medium |
| CN111623797B (en) * | 2020-06-10 | 2022-05-20 | 电子科技大学 | Step number measuring method based on deep learning |
| CN112067015B (en) * | 2020-09-03 | 2022-11-22 | 青岛歌尔智能传感器有限公司 | Step counting method and device based on convolutional neural network and readable storage medium |
-
2021
- 2021-03-04 CN CN202110241308.5A patent/CN112923922B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN112923922A (en) | 2021-06-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11796320B2 (en) | Positioning method, apparatus and device, and computer-readable storage medium | |
| KR20190116397A (en) | Face detection training methods, devices and electronic devices | |
| CN111060125B (en) | Collision detection method and device, computer equipment and storage medium | |
| CN110824587A (en) | Image prediction method, image prediction device, computer equipment and storage medium | |
| KR101576424B1 (en) | Automatic calibration method of magnetometer for indoor positioning | |
| CN113240756B (en) | Pose change detection method and device for vehicle-mounted BSD camera and storage medium | |
| CN112945227A (en) | Positioning method and device | |
| CN112362044A (en) | Indoor positioning method, device, equipment and system | |
| CN111678513A (en) | Ultra-wideband/inertial navigation tight coupling indoor positioning device and system | |
| CN115600044A (en) | River section flow calculation method, device, equipment and storage medium | |
| CN112923922B (en) | Method, system and storage medium for counting steps and determining position information of pedestrian | |
| US10677881B2 (en) | Map assisted inertial navigation | |
| CN111178126A (en) | Target detection method, target detection device, computer equipment and storage medium | |
| CN112468959B (en) | Position determining method, position determining device, computer equipment and storage medium | |
| CN113310505A (en) | External parameter calibration method and device of sensor system and electronic equipment | |
| CN110824496B (en) | Motion estimation method, motion estimation device, computer equipment and storage medium | |
| CN111582378A (en) | Training generation method, position detection method and device of positioning recognition model | |
| KR101462007B1 (en) | Apparatus for estimating attitude and method for estimating attitude | |
| CN114019182B (en) | Zero-speed state detection method and device and electronic equipment | |
| WO2018099089A1 (en) | Method and device for recognizing stationary state | |
| CN112135245B (en) | Pedestrian positioning correction method based on distance approximation discrimination Bluetooth | |
| EP3499419A1 (en) | Information processing method, information processing apparatus, and program | |
| US10001505B2 (en) | Method and electronic device for improving accuracy of measurement of motion sensor | |
| CN111401123A (en) | S L AM loop detection method and system based on deep learning | |
| KR102109374B1 (en) | Apparatus and method for estimating position |
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 |