CN114185358A - Adaptive signal tracking method of satellite navigation receiver - Google Patents

Adaptive signal tracking method of satellite navigation receiver Download PDF

Info

Publication number
CN114185358A
CN114185358A CN202111448090.7A CN202111448090A CN114185358A CN 114185358 A CN114185358 A CN 114185358A CN 202111448090 A CN202111448090 A CN 202111448090A CN 114185358 A CN114185358 A CN 114185358A
Authority
CN
China
Prior art keywords
neural network
output
rbf neural
follows
equation
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.)
Granted
Application number
CN202111448090.7A
Other languages
Chinese (zh)
Other versions
CN114185358B (en
Inventor
孙克文
杨建雷
李硕
易卿武
蔚保国
赵军
邓志鑫
王振华
刘亮
赵精博
贾诗雨
郝硕
郎兴康
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
CETC 54 Research Institute
Original Assignee
CETC 54 Research Institute
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by CETC 54 Research Institute filed Critical CETC 54 Research Institute
Priority to CN202111448090.7A priority Critical patent/CN114185358B/en
Publication of CN114185358A publication Critical patent/CN114185358A/en
Application granted granted Critical
Publication of CN114185358B publication Critical patent/CN114185358B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D1/00—Control of position, course or altitude of land, water, air, or space vehicles, e.g. automatic pilot
    • G05D1/08—Control of attitude, i.e. control of roll, pitch, or yaw
    • G05D1/0808—Control of attitude, i.e. control of roll, pitch, or yaw specially adapted for aircraft
    • G05D1/0816—Control of attitude, i.e. control of roll, pitch, or yaw specially adapted for aircraft to ensure stability
    • G05D1/0833—Control of attitude, i.e. control of roll, pitch, or yaw specially adapted for aircraft to ensure stability using limited authority control

Abstract

The invention discloses a self-adaptive signal tracking method of a satellite navigation receiver, and relates to the field of signal processing. The invention combines the traditional PID control strategy with the neural network algorithm to generate a novel PID controller with robustness and adaptability. The method utilizes the self-learning capability of the neural network, and can automatically correct online and modify robust PID parameters automatically by using a gradient descent method. The tracking result of the satellite navigation receiver shows that the code tracking loop technology has the advantages of strong adaptability and strong robustness.

Description

Adaptive signal tracking method of satellite navigation receiver
Technical Field
The invention relates to the technical field of signal processing, in particular to a self-adaptive signal tracking method of a satellite navigation receiver.
Background
The baseband signal processing is always an important component of a satellite navigation receiver, and the core content of the baseband signal processing is signal acquisition and tracking. The main purpose of tracking is, among other things, to track the code phase and the doppler frequency of the carrier, which consists of two interoperable feedback loops, a Delay Locked Loop (DLL) for code tracking and a Phase Locked Loop (PLL) for carrier tracking. In code tracking loops, a linear PLL model is often employed.
Typically, the code tracking loop includes a first or second order loop filter and a Voltage Controlled Oscillator (VCO). These systems typically have large overshoot, long settling times and are unstable due to the non-linear and time delay characteristics in the actual code tracking loop. In classical control theory, the smith method can be used to construct a controller if the transfer function of the system is known.
However, the transfer function of a real system is not easily measured or accomplished. It is well known that proportional-integral-derivative (PID) control is one of the most important control strategies. PID controllers are widely used in closed loop process control systems due to their simplicity and robustness, especially in systems where accurate mathematical models can be built.
Although PID controllers are powerful, they are not suitable for control of long-delay and non-linear complex systems where P, I and D parameters are difficult to select and adapt to a wide range of time-varying characteristics. Because the parameters are not suitable, the traditional PID controller can not achieve the ideal control effect. Thus, the application of conventional PID control is more limited and challenging.
Disclosure of Invention
In view of the above, the invention provides a self-adaptive signal tracking method for a satellite navigation receiver, which combines the traditional PID control with a neural network algorithm, has the self-learning capability of the neural network, and can perform self-tuning and automatic modification on robust PID parameters in the network through a gradient descent method.
In order to achieve the purpose, the invention adopts the technical scheme that:
a satellite navigation receiver self-adaptive signal tracking method comprises the following steps:
step 1: inputting signals, and performing closed-loop control through a PID controller;
step 2: sending a signal output by the PID controller into an RBF neural network recognizer, recognizing a nonlinear system through the RBF neural network, and predicting Jacobian matrix information of a controlled object in the system;
and step 3: and the signal output by the RBF neural network recognizer is sent to a neural network controller, and the neural network controller adopts a self-adaptive PID controller based on a single neuron and carries out self-regulation according to the system state to achieve the optimal control effect.
Further, the specific manner of step 1 is as follows:
step 1.1: calculating three independent parameters of the input signal, namely a proportional term P, an integral term I and a differential term D, defining u (t) as the output of the controller, and then the output equation in the time domain is as follows:
Figure BDA0003384618520000021
in equation (1), the input is e (t) ═ r (t) -y (t), and is the difference between the measured process output and the reference inputI.e. error; kpRepresents a proportional gain; t isiRepresents the integration time; t isdRepresents the integration time; ki＝Kpthe/T is the integral gain; kd＝KpTdIs the differential gain; t is the current time;
when the backward difference method is adopted, the output of the incremental digital PID algorithm is as follows:
u[k]＝u[k-1]+KP{e[k]-e[k-1]+Kie[k]}+Kd{e[k]-2e[k-1]+e[k-2]} (2)
step 1.2: the following transfer function is obtained according to equation (2):
Figure BDA0003384618520000031
and (3) modeling the PID controller through the transfer function of the formula (3) to realize PID control.
Further, the specific manner of step 2 is as follows:
step 2.1: let X be ═ X1，x2，…，xn]TIs the input vector of the RBF neural network, the neurons in the second layer of the RBF neural network are activated by the radial basis function, the radial vector of the RBF neural network is expressed as H ═ H1，h2，…，hm]TWherein h isjJ is 1, 2, …, m is a multivariate gaussian function, specifically:
Figure BDA0003384618520000032
in the formula (4), the symbol | | · | | represents the euclidean norm; cj＝[cj1，cj2，…，cjm]TIs a neural network jthA center vector of the node; bjIs the basic width parameter of node j in the hidden layer, bjGreater than 0, forming a base width vector B ═ B1，b2，…，bm]TIn a hidden layer of the neural network;
step 2.2:the weight vector of the RBF neural network is denoted as W ═ W1，W2，…Wm]TThe output of the RBF neural network is formed by a linear weighted sum of the radial basis functions in the hidden layer, as shown in equation (5):
Figure BDA0003384618520000033
in formula (5), the subscript I indicates that the control system adopts RBF neural network as identifier, wjRepresenting output neurons and hidden layer neurons JthThe weight value between;
step 2.3: the performance function of the identification process is defined as follows:
Figure BDA0003384618520000034
step 2.4: in order to make the RBF neural network output yI(k) Error e between actual output y (k) of controlled objectI(k) The minimum, adopting a gradient descent method to adjust the weight between the output layer and the hidden layer, the node center of the hidden layer and the radial width parameter of the node; specifically, the iterative operation is performed by equations (7), (8) and (9):
wj(k)＝wj(k-1)+η[y(k)-yI(k)]hj+α[wj(k-1)-wj(k-2)] (7)
bj(k)＝bj(k-1)+ηΔbj+α[bj(k-1)-bj(k-2)] (8)
cji(k)＝cji(k-1)+ηΔcji+α[cji(k-1)-cji(k-2)] (9)
wherein:
Figure BDA0003384618520000041
Figure BDA0003384618520000042
in the above formula, η is the tilt rate and α is the momentum factor;
step 2.5: obtaining Jacobian matrix information through an RBF neural network online identification process, namely the sensitivity of controlled object output to control input; the expression of Jacobian matrix information is as follows:
Figure BDA0003384618520000043
further, the specific manner of step 3 is as follows:
step 3.1: the output of the adaptive PID controller corresponds to the NCO output in the code tracking loop of the GNSS receiver, as shown in equation (13):
Δu(k)＝v1(k)xc1(k)+v2(k)xc2(k)+v3(k)xc3(k) (13)
in the formula (13), xci(i ═ 1, 2, 3) is input, v isi(i ═ 1, 2, 3) is the corresponding weight of a single neuron;
step 3.2: combining equation (2), the following equation is obtained:
Figure BDA0003384618520000051
Figure BDA0003384618520000052
step 3.3: in a PID controller based on a neural network, by adjusting the weight vi(i ═ 1, 2, 3) square J of the systematic error is takenC(k) Is reduced to zero; j. the design is a squareC(k) Is defined as follows:
Figure BDA0003384618520000053
in equation (16), e (k) ═ r (k) — y (k) is the error between the actual output of the control system and its currently required reference input;
step 3.4: in order to enable the adaptive PID controller of a single neuron to obtain the optimal performance, the value of the weight is adjusted and updated on line by adopting a gradient descent method according to the Hebb criterion, and the algorithm is as follows:
Figure BDA0003384618520000054
namely, the method is equivalent to:
Figure BDA0003384618520000055
in the formula
Figure BDA0003384618520000056
The Jacobian matrix information of the controlled object in the system can be obtained from the RBF neural network identification device in the formula (12).
Compared with the prior art, the invention has the following advantages:
1. the invention combines the traditional PID control strategy with the neural network algorithm to generate a novel PID controller with robustness and adaptability.
2. The method provided by the invention obviously improves the performance of a code tracking loop of the GNSS receiver under the working of a nonlinear and time-varying signal environment.
3. The method has the self-learning capability of the neural network, so that the robust PID parameters in the network can be self-adjusted and automatically modified by a gradient descent method.
Drawings
FIG. 1 is a block diagram of a generic GNSS receiver code tracking loop.
FIG. 2 is a block diagram of an online identification process of an adaptive PID control system based on an RBF neural network.
Fig. 3 shows the structure of the RBF neural network.
FIG. 4 is a structure of a single neuron PID controller.
FIG. 5 is a scatter diagram of an adaptive code tracking loop in a GNSS receiver based on RBF neural network identification.
FIG. 6 is a fast phase accumulation for an adaptive code tracking loop in a GNSS receiver based on RBF neural network identification.
FIG. 7 is a correlation envelope of early, instantaneous and late correlations for an adaptive code tracking loop in a GNSS receiver based on RBF neural network identification.
FIG. 8 shows Jacobian information identified online by the RBF neural network of the GNSS receiver adaptive code tracking loop.
FIG. 9 shows adaptive code tracking loop parameter self-tuning based on RBF neural network identification in a GNSS receiver.
Detailed Description
The following is a detailed description of the embodiments of the present invention, which is implemented on the premise of the technical solution of the present invention, and the detailed implementation manner and the specific operation process are given, but the protection scope of the present invention is not limited to the following embodiments.
A satellite navigation receiver self-adaptive signal tracking method comprises the following steps:
step 1: inputting signals into a traditional PID controller, which forms closed-loop control and ensures the stability of the whole system;
step 2: the signal from the PID is divided into several blocks and enters into the RBF Neural Network Identifier (NNI), which uses the RBF neural network to realize the nonlinear system identification function for predicting the Jacobian information of the controlled object in the system, i.e. the signal
Figure BDA0003384618520000071
And step 3: the signal from RBFNNI enters a Neural Network Controller (NNC) again, and the NNC adopts a self-adaptive PID controller based on a single neuron and carries out self-regulation capacity according to the system state so as to achieve the optimal control effect;
and 4, step 4: and (4) comprehensively optimizing the result of the GNSS receiver self-adaptive code tracking loop.
The traditional PID controller is entered in the step 1, and the specific process is as follows:
step 1.1: in the first step, three independent constant parameters of the input signal, namely proportional, integral and differential terms, are calculated, denoted as P, I and D, respectively.
These values can be interpreted in time: p depends on the current error, I depends on the accumulation of past errors, and D is a prediction of future errors based on the current rate of change.
Define u (t) as the controller output, the output equation in the time domain is:
Figure BDA0003384618520000072
in equation (1), the input is e (t) ═ r (t) — y (t), and is the difference (error) between the measured process output and the reference input; kpRepresents a proportional gain; t isiRepresents the integration time; t isdRepresents the integration time; ki＝Kpthe/T is the integral gain; kd＝KpTdIs the differential gain; t is time or instant time (current).
Step 1.2: when backward difference method is adopted, the output of the incremental digital PID algorithm:
u[k]＝u[k-1]+KP{e[k]-e[k-1]+Kie[k]}+Kd{e[k]-2e[k-1]+e[k-2]} (2)
step 1.3: secondly, modeling the PID controller through the transfer function of formula (3):
Figure BDA0003384618520000081
the process of selecting controller parameters to meet a given performance specification is referred to as controller tuning. The control loop is adjusted to adjust its control parameters (proportional band/gain, integral gain/reset, differential gain/rate) to the optimum values of the desired control response.
Stability (bounded oscillation) is a basic requirement, but in addition to this, different systems have different behavior, different applications have different requirements, and conflicts between requirements may occur.
In order to obtain better PID control performance, the proportional, integral and differential parameters of the PID controller need to be well adjusted to form a cooperative constraint relation of the control quantity. Because the neural network has the arbitrary approximation capability to the nonlinear function, the control system based on the neural network has strong self-adaption capability, and the optimal combined PID control effect can be achieved through the self-learning process.
Entering an RBF Neural Network Identifier (NNI) in the step 2, and predicting Jacobian information of the controlled object in the system specifically comprises the following steps:
step 2.1: let X be ═ X1，x2，…，xn]TIs the input vector of the neural network, the neurons in the second layer (hidden layer) are activated by radial basis functions, and the radial vector of the neural network is denoted as H ═ H1，h2，…，hm]TWherein h isj(j ═ 1, 2, …, m) should be a multivariate gaussian function, which can be written specifically as:
Figure BDA0003384618520000082
the symbol | · | |, of formula (4) represents the euclidean norm; cj＝[cj1，cj2，…，cjm]TIs a neural network jthA center vector of the node; bjIs a basic width parameter of node j in the hidden layer, whose value is greater than zero (b)j> 0), a basic width vector B ═ B) is formed1，b2，…，bm]TIn the hidden layer of the neural network.
Step 2.2: the weight vector of the neural network is denoted as W ═ W1，W2，…Wm]TThe output of the RBF neural network is formed by a linear weighted sum of the number of radial basis functions in the hidden layer, as shown in formula (5):
Figure BDA0003384618520000091
wherein the output subscript I indicates that the control system adopts RBF neural network as identifier, wjRepresenting output neurons and hidden layer neurons JthThe weight value in between.
Step 2.3: in this section, the performance function of the recognition process may be defined as follows:
Figure BDA0003384618520000092
step 2.4: in order to make the RBF neural network output yI(k) Error e between actual output y (k) of controlled objectI(k) And at the minimum, adopting a gradient descent method to adjust the weight between the output layer and the hidden layer, and the node center and the node radial width parameters of the hidden layer. The following iterative algorithm was applied:
wj(k)＝wj(k-1)+η[y(k)-yI(k)]hj+α[wj(k-1)-wj(k-2)] (7)
bj(k)＝bj(k-1)+ηΔbj+α[bj(k-1)-bj(k-2)] (8)
cji(k)＝cji(k-1)+ηΔcji+α[cji(k-1)-cji(k-2)] (9)
Figure BDA0003384618520000093
Figure BDA0003384618520000094
where η is the tilt rate and α is the momentum gene.
Step 2.5: jacobian information, namely the sensitivity of controlled object output to control input, can be obtained through an RBF neural network online identification process. The expression of Jacobian information is as follows:
Figure BDA0003384618520000101
step 2.6: the obtained Jacobian information will be sent to the single neuron based PID controller described below to achieve the best control effect.
The specific process of re-entering the Neural Network Controller (NNC) from the RBFNNI in step 3 is as follows:
step 3.1: the output of a single neuron (i.e., the output of the adaptive PID controller) corresponds to the NCO output in the GNSS receiver code tracking loop, as shown in equation (13):
Δu(k)＝v1(k)xc1(k)+v2(k)xc2(k)+v3(k)xc3(k) (13)
in the above formula, xci(i ═ 1, 2, 3) is an input; v. ofi(i ═ 1, 2, 3) are the respective weights of the individual neurons.
Compared to equation (2), the following equation can be obtained:
Figure BDA0003384618520000102
Figure BDA0003384618520000103
in a neural network based PID controller, one of the basic tasks is by adjusting the weight vi(i ═ 1, 2, 3) square J of the systematic error is takenC(k) The reduction is zero. J. the design is a squareC(k) Is defined as follows:
Figure BDA0003384618520000104
in the above equation, e (k) ═ r (k) — y (k) is the error between the actual output of the control system and its currently required reference input.
In order to obtain the optimal performance of the adaptive PID controller of a single neuron, the value of the weight is adjusted and updated on line by adopting a gradient descent method according to the Hebb criterion, and the algorithm is as follows:
Figure BDA0003384618520000111
the method is equivalent to the following steps:
Figure BDA0003384618520000112
in the formula
Figure BDA0003384618520000113
Jacobian information of the controlled object in the system can be obtained from the RBF neural network identification device in the formula (12).
The comprehensive optimization process of the GNSS signal tracking result in the step 4 comprises the following steps: simulation is carried out on Galileo EIOS BOC (1, 1) signals, and simulation results show that the method has good performance in a code tracking loop of a GNSS receiver, and particularly, a PID control strategy based on a neural network has strong robustness to nonlinear and time-varying characteristics.
FIG. 1 is a general block diagram of a GNSS code tracking loop, including a programmable pre-detection integrator, a code loop discriminator, and a loop filter.
These three functions play a key role in determining the two most important performance characteristics of the receiver code tracking loop design: code loop thermal noise error and maximum LOS dynamic stress threshold.
The idea of a DLL is to correlate the input signal with the three codes shown in fig. 1. The method comprises the following specific steps:
step 1: the input signal is multiplied by a perfectly aligned local replica of the carrier, thereby allowing the PRN code to be converted to baseband.
Step 2: the signal obtained in step 1 is multiplied with three code replicas (early, prompt and late) respectively. The three replica codes are generated at intervals of ± 0.5 chips in principle.
And step 3: after the second multiplication, the six outputs enter an I-D (integrator) circuit. The output of these integrals is a numerical value indicating the degree of correlation of the particular code replica with the code in the input signal.
The receiver has a total of six correlator outputs, namely:
Figure BDA0003384618520000121
in the formula IES，QES，ILS，QLS，IPSAnd QPSIs the six correlator outputs shown in figure 1.
This has the advantage that it is independent of the phase on the local carrier. If the local carrier is in phase with the input signal, all energy will be in the in-phase arm.
FIG. 2 is a block diagram of an online identification process of an adaptive PID control system based on an RBF neural network. It consists of three units:
conventional PID controllers: the closed-loop control is formed, and the stability of the whole control system is ensured;
RBF Neural Network Identifier (NNI): it uses RBF neural network to realize non-linear system identification function for predicting Jacobian information of controlled object in system, i.e. using RBF neural network
Figure BDA0003384618520000122
Neural Network Controller (NNC): the NNC uses an adaptive PID controller based on a single neuron, and self-regulates capacity according to the system state so as to achieve the optimal control effect.
Fig. 3 is a structure of a typical RBF neural network, which is composed of three layers: an input layer, a hidden layer and an output layer.
In an RBF neural network, the mapping from input to output is non-linear, but linear from the hidden layer to the output layer. Therefore, the learning speed of such a neural network can be greatly improved.
It has been demonstrated that RBF neural networks have the ability to approximate any continuous function with arbitrary precision.
In an RBF neural network, each input neuron corresponds to an element of an input vector and is fully connected to neurons in the hidden layer. Similarly, each neuron of the hidden layer is also connected to a neuron of the output layer.
Fig. 4 shows the structure of a single neuron, which is a multi-input single-output information processing unit constituting the basic component of a neural network, and has self-learning and self-adaptive capabilities. Simple structure and easy realization.
Therefore, the single neuron is combined with the traditional PID controller to form the single neuron self-adaptive PID controller with simple design, strong adaptability and strong robustness.
The controller utilizes Jacobian information obtained by the RBF neural network identification unit to carry out online adjustment on control parameters, and realizes the self-adaptive setting of the PID controller under different real-time conditions.
Fig. 5 is a scatter diagram of an adaptive code tracking loop identified by an RBF neural network in a GNSS receiver, which is a simulation of Galileo EI OS BOC (1, 1) signals according to an embodiment of the present invention.
The input layer consists of 4 neurons, so the input vector of the RBF neural network can be selected as X ═ u (k), u (k-1), y (k-1)]TWhere u is the NCO output of the code tracking loop, the actual output y of the code tracking system can be selected as
Figure BDA0003384618520000131
Figure BDA0003384618520000132
In the hidden layer, 8 neurons were used. Considering equation (5), the output layer consists of only one neuron, the output of which corresponds to y in equation (5)I。
Its carrier to noise ratio (C/N)0) Is 46dB-Hz, and the sampling frequency fsAt 16.3676MHz, intermediate frequency fIFAt 4.1304MHz, the front-end quantization level is 4 bits.
In the simulation, the pre-detection time of the integration circuit block of the code tracking loop is 8 ms.
In the Jacobian information online identification device (namely NNI) based on the RBF neural network, the learning rate eta of the gradient descent algorithm is 0.44, and the momentum factor alpha is 0.12.
In a single neuron based adaptive PID controller (i.e., NNC), the learning rate is set to η1＝η2＝η30.2, the momentum factor is set to α1＝α2＝α3＝0.1。
In fig. 5, a scatter plot of the in-phase (I instant) and quadrature (Q instant) components is shown, with two bubbles appearing due to the navigation bit transitions.
FIG. 6 shows the accumulation of the I-prompt indicating the navigation position in the input signal.
In FIG. 7, the early, instantaneous and late (E-P-L) correlation envelopes of an adaptive code tracking loop based on the on-line system identification of the RBF neural network are depicted.
FIG. 8 shows Jacobian information for the RBF neural network identifier in the GNSS receiver adaptive code tracking loop, which will be sent to the NNCs for weight adjustment to achieve the best control effect.
FIG. 9 is a diagram of adaptive code tracking loop parameter self-tuning based on RBF neural network identification in a GNSS receiver, in a single-neuron based adaptive PID controller, the PID tuning parameter K can be adaptively tuned by using Jacobian information obtained from an RBF neural network identification device in a code tracking systemp，KiAnd Kd。
These setting parameters become almost constant when the time instant reaches about 0.48 s.
Simulation results show that the method has good performance in a code tracking loop of a GNSS receiver, and particularly, a PID control strategy based on a neural network has strong robustness to nonlinear and time-varying characteristics. The self-adaptive code tracking technology provided by the invention enables the GNSS receiver to have robustness, self-adaptability and self-learning, and can adapt to work under different signal conditions.
In conclusion, the invention combines the traditional PID control strategy with the neural network algorithm to generate a novel PID controller with robustness and adaptability. The method utilizes the self-learning capability of the neural network, and can automatically correct online and modify robust PID parameters automatically by using a gradient descent method. The tracking result of the satellite navigation receiver shows that the code tracking loop technology has the advantages of strong adaptability and strong robustness.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents and improvements made within the spirit and principle of the present invention are intended to be included within the scope of the present invention.

Claims (4)

1. An adaptive signal tracking method for a satellite navigation receiver, comprising the steps of:
step 1: inputting signals, and performing closed-loop control through a PID controller;
step 2: sending a signal output by the PID controller into an RBF neural network recognizer, recognizing a nonlinear system through the RBF neural network, and predicting Jacobian matrix information of a controlled object in the system;
and step 3: and the signal output by the RBF neural network recognizer is sent to a neural network controller, and the neural network controller adopts a self-adaptive PID controller based on a single neuron and carries out self-regulation according to the system state to achieve the optimal control effect.
2. The adaptive signal tracking method of a satellite navigation receiver according to claim 1, wherein the specific manner of step 1 is as follows:
step 1.1: calculating three independent parameters of the input signal, namely a proportional term P, an integral term I and a differential term D, defining u (t) as the output of the controller, and then the output equation in the time domain is as follows:
Figure FDA0003384618510000011
in equation (1), the input is e (t) ═ r (t) — y (t), and is the difference, i.e., the error, between the measured process output and the reference input; kpIndicating proportional gain；TiRepresents the integration time; t isdRepresents the integration time; ki＝Kpthe/T is the integral gain; kd＝KpTdIs the differential gain; t is the current time;
when the backward difference method is adopted, the output of the incremental digital PID algorithm is as follows:
u[k]＝u[k-1]+KP{e[k]-e[k-1]+Kie[k]}+Kd{e[k]-2e[k-1]+e[k-2]} (2)
step 1.2: the following transfer function is obtained according to equation (2):
Figure FDA0003384618510000021
and (3) modeling the PID controller through the transfer function of the formula (3) to realize PID control.
3. The adaptive signal tracking method for the satellite navigation receiver according to claim 2, wherein the specific manner of step 2 is as follows:
step 2.1: let X be ═ X1，x2，…，xn]TIs the input vector of the RBF neural network, the neurons in the second layer of the RBF neural network are activated by the radial basis function, the radial vector of the RBF neural network is expressed as H ═ H1，h2，…，hm]TWherein h isjJ is 1, 2, …, m is a multivariate gaussian function, specifically:
Figure FDA0003384618510000022
in the formula (4), the symbol | | · | | represents the euclidean norm; cj＝[cj1，cj2，…，cjm]TIs a neural network jthA center vector of the node; bjIs the basic width parameter of node j in the hidden layer, bjGreater than 0, forming a base width vector B ═ B1，b2，…，bm]TIn a hidden layer of the neural network;
step 2.2: the weight vector of the RBF neural network is denoted as W ═ W1，W2，…Wm]TThe output of the RBF neural network is formed by a linear weighted sum of the radial basis functions in the hidden layer, as shown in equation (5):
Figure FDA0003384618510000023
in formula (5), the subscript I indicates that the control system adopts RBF neural network as identifier, wjRepresenting output neurons and hidden layer neurons JthThe weight value between;
step 2.3: the performance function of the identification process is defined as follows:
Figure FDA0003384618510000024
step 2.4: in order to make the RBF neural network output yI(k) Error e between actual output y (k) of controlled objectI(k) The minimum, adopting a gradient descent method to adjust the weight between the output layer and the hidden layer, the node center of the hidden layer and the radial width parameter of the node; specifically, the iterative operation is performed by equations (7), (8) and (9):
wj(k)＝wj(k-1)+η[y(k)-yI(k)]hj+α[wj(k-1)-wj(k-2)] (7)
bj(k)＝bj(k-1)+ηΔbj+α[bj(k-1)-bj(k-2)] (8)
cji(k)＝cji(k-1)+ηΔcji+α[cji(k-1)-cji(k-2)] (9)
wherein:
Figure FDA0003384618510000031
Figure FDA0003384618510000032
in the above formula, η is the tilt rate and α is the momentum factor;
step 2.5: obtaining Jacobian matrix information through an RBF neural network online identification process, namely the sensitivity of controlled object output to control input; the expression of Jacobian matrix information is as follows:
Figure FDA0003384618510000033
4. the adaptive signal tracking method of a satellite navigation receiver according to claim 3, wherein the specific manner of step 3 is as follows:
step 3.1: the output of the adaptive PID controller corresponds to the NCO output in the code tracking loop of the GNSS receiver, as shown in equation (13):
Δu(k)＝v1(k)xc1(k)+v2(k)xc2(k)+v3(k)xc3(k) (13)
in the formula (13), xci(i ═ 1, 2, 3) is input, v isi(i ═ 1, 2, 3) is the corresponding weight of a single neuron;
step 3.2: combining equation (2), the following equation is obtained:
Figure FDA0003384618510000041
Figure FDA0003384618510000042
step 3.3: in a PID controller based on a neural network, by adjusting the weight vi(i ═ 1, 2, 3) square J of the systematic error is takenC(k) Is reduced to zero; j. the design is a squareC(k) Is defined as follows:
Figure FDA0003384618510000043
in equation (16), e (k) ═ r (k) — y (k) is the error between the actual output of the control system and its currently required reference input;
step 3.4: in order to enable the adaptive PID controller of a single neuron to obtain the optimal performance, the value of the weight is adjusted and updated on line by adopting a gradient descent method according to the Hebb criterion, and the algorithm is as follows:
Figure FDA0003384618510000044
namely, the method is equivalent to:
Figure FDA0003384618510000045
in the formula
Figure FDA0003384618510000046
The Jacobian matrix information of the controlled object in the system can be obtained from the RBF neural network identification device in the formula (12).
CN202111448090.7A 2021-11-30 2021-11-30 Self-adaptive signal tracking method for satellite navigation receiver Active CN114185358B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202111448090.7A CN114185358B (en) 2021-11-30 2021-11-30 Self-adaptive signal tracking method for satellite navigation receiver

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202111448090.7A CN114185358B (en) 2021-11-30 2021-11-30 Self-adaptive signal tracking method for satellite navigation receiver

Publications (2)

Publication Number Publication Date
CN114185358A true CN114185358A (en) 2022-03-15
CN114185358B CN114185358B (en) 2024-04-12

Family

ID=80603164

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202111448090.7A Active CN114185358B (en) 2021-11-30 2021-11-30 Self-adaptive signal tracking method for satellite navigation receiver

Country Status (1)

Country Link
CN (1) CN114185358B (en)

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5396415A (en) * 1992-01-31 1995-03-07 Honeywell Inc. Neruo-pid controller
US6353766B1 (en) * 1996-02-09 2002-03-05 Siemens Aktiengesellschaft Method for generating control parameters from a response signal of a controlled system and system for adaptive setting of a PID controller
CN107255923A (en) * 2017-06-14 2017-10-17 哈尔滨工程大学 Drive lacking unmanned boat Track In Track control method based on the RBF ICA CMAC neutral nets recognized
CN108415042A (en) * 2018-01-19 2018-08-17 武汉大学 Improve the successional Phase Prediction method and system of GNSS receiver carrier phase
CN108717258A (en) * 2018-05-23 2018-10-30 中国电子科技集团公司第五十四研究所 Onboard satellite antenna servo system control method based on RBF neural
CN112180729A (en) * 2020-10-09 2021-01-05 电子科技大学中山学院 Self-adaptive trajectory tracking control method for mechanical arm

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5396415A (en) * 1992-01-31 1995-03-07 Honeywell Inc. Neruo-pid controller
US6353766B1 (en) * 1996-02-09 2002-03-05 Siemens Aktiengesellschaft Method for generating control parameters from a response signal of a controlled system and system for adaptive setting of a PID controller
CN107255923A (en) * 2017-06-14 2017-10-17 哈尔滨工程大学 Drive lacking unmanned boat Track In Track control method based on the RBF ICA CMAC neutral nets recognized
CN108415042A (en) * 2018-01-19 2018-08-17 武汉大学 Improve the successional Phase Prediction method and system of GNSS receiver carrier phase
CN108717258A (en) * 2018-05-23 2018-10-30 中国电子科技集团公司第五十四研究所 Onboard satellite antenna servo system control method based on RBF neural
CN112180729A (en) * 2020-10-09 2021-01-05 电子科技大学中山学院 Self-adaptive trajectory tracking control method for mechanical arm

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
马萍;王松艳;秦莉;杨明;: "小脑模型神经网络控制器在天线跟踪指向控制中的应用", 吉林大学学报(工学版), vol. 40, no. 4, 15 July 2010 (2010-07-15), pages 1091 - 1095 *

Also Published As

Publication number Publication date
CN114185358B (en) 2024-04-12

Similar Documents

Publication Publication Date Title
Shi et al. Support vector echo-state machine for chaotic time-series prediction
Lin et al. Event-triggered fault detection filter design for discrete-time memristive neural networks with time delays
Wu et al. Optimal sensor querying: General Markovian and LQG models with controlled observations
CN102141776A (en) Particle filter and RBF identification-based neural network PID control parameter self-setting method
Haber et al. Nonlinear internal model control using neural networks: An application for machining processes
CN106249590B (en) The method that integrated self-adaptive Nano satellite posture determines
Jwo et al. Neural network aided adaptive extended Kalman filtering approach for DGPS positioning
Berenguel et al. Modelling the free response of a solar plant for predictive control
Wu et al. Adaptive fuzzy output feedback quantized control for uncertain nonlinear hysteretic systems using a new feedback-based quantizer
Iplikci Online trained support vector machines‐based generalized predictive control of non‐linear systems
CN114185358A (en) Adaptive signal tracking method of satellite navigation receiver
CN111709496A (en) Modulation mode recognition and model training method and device based on neural network
Dah-Jing et al. Neural network aided adaptive Kalman filter for GPS/INS navigation system design
CN114567288A (en) Distributed cooperative nonlinear system state estimation method based on variational Bayes
Asano et al. A design of self-tuning predictive PID controllers
Amin et al. System identification via artificial neural networks-applications to on-line aircraft parameter estimation
Brouillon et al. Regularization for distributionally robust state estimation and prediction
Maiworm et al. Online gaussian process learningbased model predictive control with stability guarantees
Landicheff et al. Continuous-discrete time observer design for multi-output linear continuous systems with aperiodic asynchronous measurements
Chen et al. Insufficient initial condition of fractional order derivative definitions
Chen et al. Adaptive bandwidth PLL design based on fuzzy logic control
Zamani et al. Minimum-energy filtering on the unit circle
Sun Adaptive code tracking loop design for GNSS receivers
Zribi et al. Models’ bank selection of nonlinear systems by integrating gap metric, margin stability, and MOPSO algorithm
Chen et al. Deep representation-based packetized predictive compensation for networked nonlinear systems

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