Data processing: vehicles – navigation – and relative location – Navigation – Employing position determining equipment
Reexamination Certificate
2001-11-02
2003-03-25
Cuchlinski, Jr., William A. (Department: 3661)
Data processing: vehicles, navigation, and relative location
Navigation
Employing position determining equipment
C701S200000, C073S17800T, C342S457000
Reexamination Certificate
active
06539304
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The invention relates generally to GPS navigation systems and more particularly, to a GPS navigation system utilizing neural network technology as an alternative to traditional methods of solving GPS navigation equations.
2. Description of Related Art
Traditional GPS receivers solve GPS navigation equations utilizing recursive least squares and extended Kalman filter techniques. Each of these well known techniques, however, tend to exhibit a high sensitivity to measurement errors. Accordingly, alternative solution techniques are desirable.
Neural networks are information processing systems composed of a large number of simple processing elements called neurons. Many kinds of neural network models have been proposed and extensively used in various signal processing applications. The key features of neural networks are asynchronous parallel and distributed processing, nonlinear dynamics, global interconnection of network elements, self-organization, and high-speed computational capability.
Neural networks have been designed to mimic the human brain as much as possible in the realm of mathematics and to try and duplicate the intelligence of the human brain. The design of a neural network follows the model of the human brain in terms of a large collection of individual nodes (neurons in biological terminology) interconnected to one another in a fashion similar to the neuron connections in a human brain. Neural network systems are capable of solving virtually any problem that involves mapping input data to output data. Generally, a neural network can be classified into various categories varying from supervised learning to unsupervised self learning networks.
It has been shown that a neural network can approximate any continuous and differentiable function to any degree of accuracy. This property leads to many potential applications of neural networks in signal processing such as blind signal separation, image registration, blind deconvolution and many others.
Generally, and in the simplest case, the topology of a neural network consists of a single layer of fully interconnected processors or neurons. More often, the network is divided into multiple layers with full interconnection between layers but not necessarily with every neuron. One layer of neurons may be used to input values into the network, while another layer may output the final results. In between the input and output layer, there may be intermediate or hidden layers. The hidden layers are necessary to allow solutions of nonlinear problems. When the weighting matrices and various parameters associated with the activation functions have been set to correct levels, a complex stimulus pattern at the input layer successively propagates between hidden layers, to result in an often simpler output pattern.
Typically, a neural network is “taught” by feeding a succession of input patterns and corresponding expected output patterns. The network “learns” by measuring, directly or indirectly, the difference or error, at each output unit, between the desired output pattern and the pattern that it just produced. Having done this, the internal weights and activation parameters of the hidden layer(s) are modified by a learning algorithm to provide an output pattern which more closely approximates the desired output pattern, while minimizing the error over the spectrum of input patterns. Neural network learning is an iterative process, involving multiple “lessons”. Neural networks have the ability to process information in the presence of noisy or incomplete data and yet still generalize to the correct solution.
Hence, those skilled in the art have recognized a need for alternative methods of solving GPS navigation equations having less sensitivity to measurement errors. The invention fulfills these needs and others.
SUMMARY OF THE INVENTION
Briefly, and in general terms, the present invention is directed to a method of and a device for solving GPS navigation equations. The invention is further directed to systems incorporating such devices.
In a first aspect, the invention relates to a method of determining navigation data for a GPS receiver. The method includes receiving input signals via the GPS receiver from at least one GPS satellite. The input signals include satellite-related navigation information. The method further includes applying the input signals to a neural network to obtain an output signal representative of receiver-related navigation information.
In a detailed facet of the invention, the input signals include at least one of a position vector of the GPS satellite and a pseudorange measurement of the GPS satellite. In another detailed aspect, the neural network comprises one of either a discrete-time Hopfield neural network, a continuous-time Hopfield neural network, a cellular neural network, a multilayer perception networks, a self-organizing system, a radial basis function network or a high-order neural network. In yet another detailed aspect, the neural network comprises an first node layer connected to a second node layer through a first connection layer and a third node layer connected to the second node layer through a second connection layer. In this aspect, applying the input signals to a neural network to obtain an output signal includes connecting the input signals to the second node layer through the first node layer and the first connection layer; connecting the outputs of the second node layer to the third node layer through the second connection layer; and combining the outputs of the second node layer to provide receiver position data.
In a second aspect, the invention relates to a GPS receiver comprising a satellite receiver/processor having an input receiving input signals from at least one GPS satellite and an output providing satellite-related navigation information. The receiver further includes a neural network having an input receiving the satellite-related information to obtain an output signal representative of receiver-related navigation information.
In a detailed aspect of the invention, the neural network includes a first node layer connected to a second node layer through a first connection layer and a third node layer connected to the second node layer through a second connection layer. In another detailed aspect, the first node layer comprises a plurality of input neurons receiving the input signals, the second node layer comprises a plurality of hidden neurons connected to the plurality of input neurons through the first connection layer and the third node layer comprises a plurality of output neurons connected to the hidden neurons through the second connection layer. In still another facet of the invention, the second connection layer is weighted.
In a third aspect, the invention relates to a navigation system for tracking the position of an object within a geographic area. The system includes a satellite receiver/processor having an input receiving input signals from at least one GPS satellite and an output providing satellite-related navigation information. The system also includes a neural network having an input receiving the satellite-related information to obtain an output signal representative of receiver-related navigation information. Also included in the system are complimentary devices such as a presentation device for presenting the receiver-related navigation information to a system user, processing devices such as a computer system, or a transmission/communication device such as a cellular telephone for transmitting the receiver-related navigation information across a communication link.
In a detailed aspect of the invention, the presentation device includes a map data base that stores information related to the geographic area. The device also includes a map matching processor that receives as input the receiver-related navigation information and combines the geographic-area information with the navigation information to provide display information. The device also includes a display responsive to the display informa
Fulwider Patton Lee & Utecht LLP
SiRF Technology Inc.
LandOfFree
GPS navigation system using neural networks does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with GPS navigation system using neural networks, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and GPS navigation system using neural networks will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3057623