CN103412682B - Electronic whiteboard experimental system based on infrared ultrasonic co-located and localization method - Google Patents

Abstract

A kind of electronic whiteboard experimental system based on infrared ultrasonic co-located and localization method, belong to wireless positioning field.This experimental system mainly includes being integrated with infrared transmitter and the signal emission module of ultrasonic transmitter, being integrated with infrared remote receiver and the signal receiving module of six ultrasonic receivers, signal processing module and host computer.Sign pen can produce periodically infrared, ultrasonic signal, the infrared remote receiver of receiving terminal detects infrared signal, six ultrasonic receivers start simultaneously at work, detection is respectively received the time used by ultrasonic signal respectively, send signal processing module to, use Chan algorithm that the data through BP corrected neural network are solved and obtain signal source.The present invention is greatly saved the cost of sensor, and small volume, it is possible to the detection data obtaining sensor promptly and accurately, can simultaneously serve as the experiment porch of multiprecision arithmetic, and provides convenience condition for the research of algorithm validity etc..

Description

Electronic whiteboard experiment system based on infrared and ultrasonic combined positioning and positioning method

Technical Field

The invention discloses an electronic whiteboard experiment system based on infrared and ultrasonic combined positioning and a positioning method, and belongs to the field of wireless positioning.

Background

The electronic whiteboard is a dry-wiping whiteboard which can write on the board surface at will and can be captured electronically, integrates a blackboard, multimedia and a projector into a whole, has far more functions than the sum of the blackboard, the multimedia and the projector, and is a high-tech product internationally promoted in recent years. The electronic whiteboard is a breakthrough to the traditional blackboard writing and multimedia classroom, is used on line with a projector, a computer and the like, and can easily realize the functions of information demonstration, editing and the like by utilizing the data processing function and the information transmission function of the computer. The electronic whiteboard teaching platform compares two teaching modes of a blackboard and an electrified education platform, keeps the interactivity of the blackboard and the richness of the electrified education, overcomes the defects of monotonicity of the blackboard and unidirectionality of the electrified education, fundamentally solves the problems and the defects existing in the previous teaching mode, really realizes the interaction of teaching and learning, and realizes a high-quality and high-efficiency teaching mode.

However, electronic whiteboards in the market are large in board surface and highly integrated due to the fact that the electronic whiteboards face customers, internal sensor information is difficult to obtain, and an electronic whiteboard experiment system suitable for experimental research is not provided. Therefore, when researchers want to research on the core technology (such as positioning accuracy, identification accuracy, etc.) of the electronic whiteboard, the difficulty that feedback information such as accuracy, response speed, accuracy, etc. necessary for verifying the validity of the research result cannot be obtained due to the inability to obtain the sampling information of each sensor integrated in the whiteboard hardware must be faced, which is a great obstacle in the research work, and the update of the electronic whiteboard must be delayed.

Currently, basic positioning algorithms of electronic whiteboards include a field strength indication (RSSI) positioning method, a time of arrival (TOA) or Time Difference (TDOA) based positioning method, an AOA based positioning method, and various hybrid positioning methods. The TDOA-based positioning method has no strict time synchronization requirement, can be suitable for various systems, and has the advantages of low application cost, high positioning precision and wide attention. However, due to various interferences existing in the environment, data obtained by actual measurement often contains errors, which leads to the reduction of positioning accuracy and even failure of positioning, so that it is necessary to research a high-accuracy positioning algorithm.

Disclosure of Invention

The invention provides an electronic whiteboard experiment system based on infrared and ultrasonic combined positioning and a positioning method. The experimental system generates infrared and ultrasonic signals by utilizing the transmitting principle of the infrared transmitting sensor and the ultrasonic transmitting sensor to simulate the signal pen, the signal receiving end obtains the propagation time between the signal sent from the signal pen and the signal received by the receiving sensor, and then the positioning algorithm solves the data to obtain the position information of the signal pen, so that the purpose of positioning the signal pen is achieved. The invention can accurately acquire the detection data of the sensor in time, can make up the defects that the existing electronic whiteboard product has high integration level and is not suitable for experimental research, and has the characteristics of small volume and low cost; meanwhile, the method can be used as an experimental platform of a high-precision algorithm, and provides convenient conditions for researches on verification of algorithm effectiveness and the like. Meanwhile, the electronic whiteboard experiment system is used as a hardware platform, and a combined positioning algorithm based on a BP neural network and a Chan algorithm is provided.

The main idea of the invention is as follows: when a signal pen starts to write in a writing area of the electronic whiteboard experiment system by pressing a switch, the signal pen can generate periodic infrared and ultrasonic signals, an infrared receiving sensor at a receiving end detects the infrared signals (the transmission speed of the infrared is close to the light speed, and the transmission time of the infrared is negligible), a DSP (digital signal processor) is triggered to interrupt, six ultrasonic receiving sensors simultaneously start to work (considered as instantaneous trigger), the time for respectively receiving the ultrasonic signals is respectively detected and transmitted to a signal processing module, and sampling data corrected by a BP (back propagation) neural network is solved by utilizing a Chan algorithm to obtain position information of a signal source (namely the signal pen) on the electronic whiteboard experiment system, so that the purpose of positioning the signal pen is achieved.

The invention adopts the following specific technical scheme: including signal emission module (signal pen for short), signal reception module, signal processing module, host computer software, its characterized in that: still include infrared emission sensor, ultrasonic emission sensor, infrared receiving sensor, six ultrasonic receiving sensor, infrared emission sensor and ultrasonic emission sensor are fixed in the signal pen, send periodic infrared ray signal and ultrasonic signal by single chip microcomputer control, infrared receiving sensor and six ultrasonic receiving sensor are fixed at signal reception module, receive infrared ray signal and ultrasonic signal, the output of signal reception sensor module links to each other with signal processing module's input, signal processing module adopts DSP as main control chip, signal processing module's output links to each other with host computer software.

The positioning algorithm of the electronic whiteboard experiment system comprises the following steps:

1) when the signal transmitting module switch is pressed down, the infrared transmitting sensor and the ultrasonic transmitting sensor send out periodic infrared signals and ultrasonic signals;

2) the infrared receiving sensor at the signal receiving end detects the signal, triggers six ultrasonic receiving sensors to start working at the same time, and respectively detects the time t for respectively receiving the ultrasonic signal_i,i=1,2,3,4,5,6；

3) Using the data of the first ultrasonic receiving sensor as reference, and adopting TDOA algorithm to calculate and obtain 5 TDOA values t_i1,i=2,3,4,5,6；

4) Training BP neural network by using error-free theoretical data, and using the trained BP neural network to perform the training on 5 TDOA values t_i1I =2,3,4,5, 6;

5) and performing position estimation by using the corrected data and adopting a Chan algorithm to obtain the position information of the signal pen on the electronic whiteboard experiment system, thereby achieving the purpose of positioning the signal pen.

The invention adopts the infrared sensor and the ultrasonic sensor as the signal input for acquiring the positioning information, greatly saves the cost of the sensor, reduces the volume and can acquire the information of the sensor timely and accurately.

Drawings

FIG. 1 is a schematic diagram of an electronic whiteboard experiment system

FIG. 2 is a schematic block diagram of an electronic whiteboard experiment system

Fig. 3 is a block diagram of a system structure of the electronic whiteboard experiment system

FIG. 4 is a software flow chart of signal transmission of the electronic whiteboard experiment system

FIG. 5 is a schematic diagram of a positioning algorithm of the electronic whiteboard experiment system

FIG. 6 is a schematic block diagram of a positioning algorithm of the electronic whiteboard experiment system

FIG. 7 BP neural network model with TDOA value correction

FIG. 8 is a software flow chart of a positioning algorithm of the electronic whiteboard experiment system

Detailed Description

The experimental system is further described below with reference to the accompanying drawings:

as shown in fig. 1 and 2, the hardware structure of the present invention is relatively simple, low in cost and easy to implement.

In order to reduce the volume of a signal transmitting end, the singlechip PIC12F629 which is small in volume and is used for controlling and generating pulse waveforms is selected from the singlechip PIC which is small in volume and is provided with 8 bits and 8 pins. The waveform circuit is mainly generated by a GP0 port and a GP1 port of the singlechip, because a resonant circuit in the driving circuit resonates a frequency component of 80KHZ contained in the pulse, the pulse width is directly related to the power of each frequency component contained in the pulse, and the wider the pulse is, the higher the power is, and the shorter the service life of the battery is; the pulse width is too narrow, and the 80KHZ signal energy contained in the pulse is small, so that the system emission power is insufficient, therefore, the pulse width selected by the experimental system is 400uS, and the infrared diode can be completely driven by adopting a pulse signal with the width of 200 uS. The period of the pulse signal is set to 10mS, so that 100 signals can be sent out every second, and the receiving requirement and the visual requirement are completely met. A GP0 port of the singlechip PIC12F629 sends out a pulse signal with the pulse width of 400uS and the period of 10mS as a driving signal of the ultrasonic sensor; the GP1 port sends out a pulse signal with a pulse width of 200uS and a period of 10mS as a driving signal of the infrared diode. The software implementation flow of the PIC single-chip microcomputer control process is shown in fig. 4.

The ultrasonic emission sensor of the experimental system requires 360-degree omnidirectional transmission and has good transmission characteristics in an omnidirectional range, so that the piezoelectric film ultrasonic sensor PT80 KHZ-01 produced by EMAS sensor company in America is selected. The sensor has the characteristics of high frequency response, good dynamic range, high output voltage, good stability, impact resistance, difficult aging and the like while meeting the requirements, and meets various requirements of the experimental system.

The infrared emission sensor required by the experimental system also requires 360-degree omnidirectional transmission, but no infrared emission sensor meeting the conditions is available in the market, so that the experimental system selects the infrared emission diode SFH426, the emission angle of the infrared emission diode SFH426 is 120 degrees, and the infrared emission diode SFH426 is uniformly distributed for one circle to realize 360-degree omnidirectional emission.

The ultrasonic receiving sensor adopts SR80 KHZ-01 of EMAS sensor company, which is a piezoelectric film sensor, and has the advantages of large range of received signals, larger direction angle and bandwidth, and the like.

The infrared receiving sensor is SFH 506-38, which is a small-outline infrared remote control system receiving sensor, and the PIN diode and the preamplifier are arranged in a welding shell and packaged by epoxy materials. The modulated signal output by the receiving sensor can be directly demodulated by the microprocessor. Its main advantages are high reliability in interference environment and shielding function to prevent electromagnetic interference.

The signal processing module of the experimental system, including the power management module and the DSP main control circuit, is described below with reference to fig. 3. The infrared receiving sensor is connected with an I/O port of the DSP, the ultrasonic receiving sensor is connected with the I/O port with the capturing function in the DSP, and the capturing interruption is utilized to acquire the ultrasonic propagation time. And (3) carrying out algorithm realization in the DSP, solving the sampling data, outputting the position information of the signal pen, and transmitting the position information to the upper computer software through an RS232 serial port.

In order to realize an ultrasonic time delay accurate extraction algorithm, a core chip of a signal processing circuit of the experimental system adopts a TMS320C28X series floating point DSP controller TMS320F28335 produced by TI company, and the TMS320F28335 has the high-speed processing capacity of 150MHz and completely meets the real-time requirement of the experimental system. Compared with the prior fixed-point DSP, the device has the advantages of high precision, low cost, low power consumption, high performance, high peripheral integration level, large data and program memory space, more accurate and faster A/D conversion and the like. TMS320F28335 is provided with a 32-bit floating point processing unit, 6 DMA channels supporting ADC, McBSP and EMIF, has up to 18 PWM outputs, of which 6 are PWM outputs of higher precision unique to TI, 12-bit 16-channel ADC. Thanks to the floating point arithmetic unit, a user can quickly write a control algorithm without consuming excessive time and energy on decimal processing operation, and compared with the prior DSC, the average performance is improved by 50 percent, thereby simplifying software development, shortening the development period and reducing the development cost.

The power management module adopts an integrated DC-DC voltage transformation circuit, and the main function of the power management module is to convert the battery voltage (12 v) into the voltages (5 v and 3.3 v) required by the DSP control board, the infrared receiving sensor and the ultrasonic receiving sensor.

The electronic whiteboard experiment system is used as a hardware platform, and a combined positioning algorithm based on a BP neural network and a Chan algorithm is provided. The method is as shown in fig. 5-7, and a plane rectangular coordinate system is established with the upper left corner of the writing area as the origin, and the horizontal and vertical directions as the X and Y axes, respectively. The specific implementation of the algorithm is as follows:

1) when the signal pen switch is pressed down, the infrared emission sensor and the ultrasonic emission sensor send out periodic infrared signals and ultrasonic signals;

2) the infrared receiving sensor at the signal receiving end detects that the infrared signal is interrupted and triggers six sensorsThe ultrasonic receiving sensors start to operate simultaneously, and the time t taken for the ultrasonic signals to be received is respectively detected_i,i=1,2,3,4,5,6；

3) Using the first ultrasonic receiving sensor as reference, and adopting TDOA algorithm to calculate and obtain 5 TDOA values t_i1=t_i-t₁,i=2,3,4,5,6；

4) And (3) carrying out data correction by using a BP neural network, specifically as follows:

a. designing a BP neural network: the device consists of an input layer, a hidden layer and an output layer. The input layer consists of 5 TDOA measurements provided by 6 ultrasonic receiving sensors; the number of neurons in the hidden layer can be obtained by an empirical formula, and a Sigmoid function is selected as a transfer function of the hidden layer; the output layer is composed of 5 neurons, adopts a linear transfer function, and has a modified TDOA value

b. Training a BP neural network: the measured value of the input layer adopts error-free data obtained by fixed-point measurement; the output value of the output layer is the same as the input layer.

c. With 5 TDOA values t in the trained BP neural network pair (3)_i1I =2,3,4,5,6 for correction

5) And performing position estimation by using the corrected distance value and adopting a Chan algorithm to obtain the position information of the ultrasonic signal source (namely, the signal pen) on the electronic whiteboard experiment system, thereby achieving the purpose of positioning the signal pen. The method comprises the following specific steps:

let (x, y) be the position of the signal pen to be determined, (x)_i,y_i) Is the known position of the ith ultrasonic receiving transducer, C is the ultrasonic propagation velocity, R_iIndicating the distance, R, from the stylus to the ith ultrasonic receiving transducer_i1Representing the difference between the distances from the signal pen to the ith ultrasonic receiving sensor and the distances from the signal pen to the 1 st ultrasonic receiving sensor, then:

$R_{i1}=c{t}_{i1}=R_i-R_1=\sqrt{{(x_i-x)}^2+{(y_i-y)}^2}-\sqrt{{(x_1-x)}^2+{(y_1-y)}^2},i=\mathrm{2,3,4,5,6}---\left(1\right)$

the two sides are squared and arranged to obtain:

$R_{i1}^2+2R_{i1}{R}_1=K_i-2x_{i1}x-2y_{i1}y-K_1---\left(2\right)$

wherein x_i1=x_i-x₁,y_i1=y_i-y₁,,i=2,3,4,5,6

Is provided with Z_a=[x,y,R₁]^TFor unknown vectors, equation (2) is now a one-dimensional equation of three with respect to these three variables, and thus an error vector can be obtained:

$\mathrm{\ψ}=h-G_a{Z}_a^0---\left(3\right)$

wherein, $h=\frac{1}{2}\left[\begin{array}{c}{R}_{\mathrm{2,1}}^2-K_2+K_1\\ R_{\mathrm{3,1}}^2-K_3+K_1\\ ...\\ R_{\mathrm{6,1}}^2-K_6+K_1\end{array}\right]$ ， $G_a=\frac{1}{2}\left[\begin{array}{ccc}{x}_{\mathrm{2,1}}& y_{\mathrm{2,1}}& R_{\mathrm{2,1}}\\ x_{\mathrm{3,1}}& y_{\mathrm{3,1}}& R_{\mathrm{3,1}}\\ ...& ...& ...\\ x_{\mathrm{6,1}}& y_{\mathrm{6,1}}& R_{\mathrm{6,1}}\end{array}\right]$

the expression form of { } when no noise is defined is { }⁰And then:

$t_{i,1}=t_{i,1}^0+n_{i,1}$ ， $R_{i,1}=R_{i,1}^0+{\mathrm{Cn}}_{i,1}$

wherein n is_i1=n_i-n₁，n_iThe systematic error of the time is detected for the ith ultrasonic wave reception sensor.

The error vector thus obtained is:

$\mathrm{\ψ}=\mathrm{CBn}+0.5C^{2}n\⊗n---\left(4\right)$

wherein, $B=\mathrm{dig}\{R_2^0,R_3^0,...,R_6^0\}$ ， $n={[n_{\mathrm{2,1}},n_{\mathrm{3,1}},...,n_{\mathrm{6,1}}]}^T$ ，indicating dot multiplication operation, in practical casesSo a term behind the above equation is negligible, the error vector ψ can be approximated as a random vector following a normal distribution, the covariance of ψ is:

Ψ=E[ψψ^T]=C²BQB (5)

wherein Q = E [ nn ] is the covariance matrix of TDOA, and WLS is used to find an approximate solution of equation (4) as:

$Z_a=\arg \min \left\{{(h-G_a{Z}_a)}^T{\mathrm{\Ψ}}^{-1}(h-G_a{Z}_a)\right\}={\left(G_a^T{\mathrm{\Ψ}}^{-1}{G}_a\right)}^{-1}{G}_a^T{\mathrm{\Ψ}}^{-1}h---\left(6\right)$

when the distance between the signal pen and the ultrasonic receiving sensor is far, an R is defined⁰To make it and eachApproximately, when B = R⁰I，Z_aThe substitution can be as follows:

$Z_a={\left(G_a^T{Q}^{-1}{G}_a\right)}^{-1}{G}_a^T{Q}^{-1}h---\left(7\right)$

when the distance between the signal pen and the ultrasonic receiving sensor is short, the value of B is firstly calculated, and then the position of the signal pen is obtained by the formula (6). The first WLS is now over.

And performing second WLS by using the first WLS estimation value and the constraint relation:

$Z_a^{\′}={\left(G_a^{\′T}{B}^{\′-1}{G}_a^T{Q}^{-1}{G}_a^{\′}{B}^{\′-1}{G}_a^{\′}\right)}^{-1}\×\left(G_a^{\′T}{B}^{\′-1}{G}_a^T{Q}^{-1}{G}_a^{\′}{B}^{\′-1}\right)h^{\′}---\left(8\right)$

wherein, $h^{\′}=\left[\begin{array}{c}{(Z_a\left(1\right)-x_1)}^2\\ {(Z_a\left(2\right)-x2)}^2\\ Z_a{\left(3\right)}^2\end{array}\right]$ ， $G_a^{\′}=\left[\begin{array}{cc}1& 0\\ 0& 1\\ 1& 1\end{array}\right]$ ， $B^{\′}=\left[\begin{array}{ccc}{Z}_a\left(1\right)-x_1& 0& 0\\ 0& Z_a\left(2\right)-y_1& 0\\ 0& 0& Z_a\left(3\right)\end{array}\right]$

the final solution is to estimate the location of the stylus as:

$Z=\±\sqrt{Z_a^{\′}}+\left[\begin{array}{c}{x}_1\\ y_1\end{array}\right]---\left(9\right)$

in the experimental system, the signal pen only moves in the first quadrant, so the result obtained by the formula (9) only takes a positive value, thereby realizing positioning.

Claims (1)

1. A positioning method for an electronic whiteboard experiment system based on infrared and ultrasonic combined positioning comprises a signal transmitting module, a signal receiving module, a signal processing module and an upper computer, and further comprises an infrared transmitter, an ultrasonic transmitter, an infrared receiver and six ultrasonic receivers, wherein the infrared transmitter and the ultrasonic transmitter are fixed on the signal transmitting module and send periodic infrared signals and ultrasonic signals, the infrared receiver and the six ultrasonic receivers are fixed on the signal receiving module and receive the infrared signals and the ultrasonic signals, the output end of the signal receiving module is connected with the input end of the signal processing module, and the output end of the signal processing module is connected with the upper computer;

the method is characterized by comprising the following steps:

1) the infrared emitter and the ultrasonic emitter emit periodic infrared signals and ultrasonic signals;

2) the infrared receiver at the receiving end detects the signal, triggers six ultrasonic receivers to start working at the same time, and respectively detects the time t for respectively receiving the ultrasonic signals_i,i＝1,2,3,4,5,6；

3) Using the data of the first ultrasonic receiving sensor as reference, and adopting TDOA algorithm to calculate and obtain 5 TDOA values t_i1,i＝2,3,4,5,6；

4) Training BP neural network by using error-free theoretical data, and using the trained BP neural network to perform the training on 5 TDOA values t_i1I is corrected to 2,3,4,5, 6;

5) and based on the corrected TDOA value, position estimation is carried out by adopting a Chan algorithm to obtain the position information of the signal pen on the electronic whiteboard experiment system, so that the aim of positioning the signal pen is fulfilled.