AU672348B2 - Integrated vehicle positioning and navigation system, apparatus and method - Google Patents

Description

-1- R Fgu00ti11 Regulation 32

AUSTRALIA

Patents Act 1990 COMPLETE SPECIFICATION FOR A STANDARD PATENT

ORIGINAL

a.

a ee. eQ Name of Applicant: Actual Inventors: CATERPILLAR INC.

Adam J. GUDAT, Walter J. BRADBURY, Dana A. CHRISTENSEN, Richard G. CLOW, Lonnie J. DEVIER, Carl A. KEMNER, Karl W. KLEIMENHAGEN, Craig L. KOEHRSEN, Christos N. KYRTSOS, Norman K. LAY, Joel L. PETERSON, Prithvi N. RAO, Larry E. SCHMIDT, James W. SENNOTT, Gary K. SHAFFER, WenFan SHI, Dong Hun SHIN, Sanjiv J. SINGH, Darrell E. STAFFORD, Louis J. WEINBECK, Jay H. WEST, William L. WHITTAKER, BaoXin WU CARTER SMITH BEADLE 2 Railway Parade Camberwell Victoria 3124 Australia Integrated Vehicle Positioning and Navigation System, Apparatus and Method Address for service in Australia: Invention Title: The following statement is a full description of this invention, including the best method of performing it known to us -2- INTEGRATED VEHICLE POSITIONING AND NAVIGATION SYSTEM, APPARATUS AND METHOD BACKGROUND OF THE INVENTION 1. Field of the Invention The present application is a divisional application of Australian Patent No.

46043/93 which was divided from Australian Patent No. 642638. The present application relates to a system for improving the accuracy of a position determination system. Other inventions relating to systems/methods for navigating an autonomous vehicle are disclosed in other co-pending applications e.g. 46045/93.

For an overview of the interrelated navigation and positioning systems/apparatus/methods, the reader should refer to the specification accompanying Australian Patent No. 642638.

2. Related Art There is presently under development a terrestrial position determining system, referred to as the global positioning system (GPS), designated NAVSTAR by the U.S. Government. In this system, a multitude of orbiting satellites will be used to determine the terrestrial position of receivers on the Earth. In the planned system, there will be eight orbiting satellites in each of three sets of orbits, 21 satellites on line and three spares, for a total of 24 satellites. The three sets of orbits 20 will have mutually orthogonal planes relative to the Earth. The orbits are neither polar orbits nor equatorial orbits, however. The satellites will be in 12-hour orbits.

The position of each satellite at all times will be precisely known. The longitude, latitude, and altitude with respect to the center of the Earth, of a i&eiver at any point close to Earth at the time of transmission, will be calculated by determining the propagation time of transmissions from at least four of the satellites to the receiver. The more satellites used the better. A current constraint on the number of satellites is that the currently available receiver has only five channels.

SUMMARY OF THE INVENTION In this specification, the term "integrated vehicle positioning and navigation system" as used throughout, means apparatus, method or a combination of both DCC:TJI11:TG:#17679.SPC 5 AraU 1995 -3apparatus and method. The present invention overcomes many of the limitations present in conventional technology in the fields of positioning and navigation, and thereby provides for highly accurate and autonomous positioning and navigation of a vehicle.

As disclosed, the conventional capabilities of the IRU (Inertial Reference Unit) and the GPS (Global Positioning System) have been both combined, and greatly enhanced, in a cost-effective manner to provide autonomous vehicle capability. In doing so, many novel and inventive systems, apparatus, methods and techniques have been developed which allow for a superior positioning capability, and flexible autonomous navigational capability.

The vehicle positioning and navigation system achieves some of its advantages by using a novel and enhanced combination of three independent S• subsystems to determine position. These are: GPS (Global Positioning System) satellites, an inertial reference system, and the vehicle odometer. These three 15 independent subsystems produce positioning data which may then be combined by a novel Kalman filter so that an output stream of positioning data of improved accuracy may be produced at a much faster rate (approximately 20 iHz).

The present invention also provides for increased positioning accuracy by 0*° *0 .providing novel systems for compensating for noise and errors in positioning data.

20 In accordance with a first aspect of the present invention, there is disclosed a system for determining an estimate of the position of a vehicle based on .ooo.i pseudoranges derived from a vehicle positioning system and prior position estimates, the system comprising: first means for receiving signals from a plurality of satellites, responsively determining respective pseudoranges from said satellites to the vehicle, and deriving a first position estimate of the vehicle's position as it function of said pseudoranges; second means responsive to said first means for receiving the prior position estimates and responsively determining a vehicle path model x;wng a best fit algorithm; DCC:TJHI:TG #17679 RS I 31 luly 1996 aCIII -4third means responsive to said second means, said third means for extrapolating a second position estimate from said vehicle path model; and fourth means responsive to said third means, said fourth means for setting said first position estimate equal to said second position estimate if said second position estimate is more accurate.

In accordance with a second aspect, there is disclosed a method for determining an estimate of the position of a vehicle based on a vehicle position system and prior position estimates via a best fit algorithm, the method comprising the steps of: receiving signals from a plurality of satellites, responsively determining respective pseudoranges fiom said satellites to said vehicle, and computing a first position estimate using the vehicle positioning system as a function of said pseudoranges; determining a vehicle path model using the best fit algorithm as a 15 function of the prior position estimates; extrapolating a second position estimate from said vehicle path model; and setting said first position estimate equal to said second position estimate if said second position estimate is more accurate.

In accordance with a third aspect, there is disclosed a method for improving the accuracy of position estimates of a vehicle based on a vehicle positioning system by considering prior position estimates via a best fit algorithm, the method comprising the steps of: computing positions using the vehicle positioning system; incorporating said positions into a best fit algorithm; extrapolating an estimated subsequent position from said best fit algorithm; and substituting a subsequent position computed by said vehicle positioning system with said estimated subsequent position extrapolated from said best fit algorithm.

DCC:TIH:TG #17679 RS I 31 July 1996 4a- A better appreciation of these and other advantages and features of the present invention, as well as how the present invention realizes them, will be gained from the following detailed description and drawings of the various embodiments, and from the claims.

e a oe 0 0 o* DCC:TJH:TG:#I 7679.RS I 31 July 1996 I I- BRIEF DESCRIPTION OF THE DRAWINGS The present invention is better understood by reference to the following drawings in conjunction with the accompanying text.

Fig. 1 is a diagrammatic view of the GPS satellites in 6 orbital planes.

Fig. 2 is a diagrammatic view illustrating the GPS navigation equations.

Fig. 3 is a diagrammatic view of a typical autonomous work site in which the present invention is practiced.

Fig. 4 is a diagrammatic representation of the interrelationships between the navigator, the vehicle positioning system (VPS), and the vehicle controls.

Fig. 5 is a context diagram, illustrating the various elements and their interrelationship in an autonomous control system according to the present invention.

Fig. 6 is a diagrammatic representation of the operation of a Global Positioning System (GPS), including a satellite constellation, A pseudolite, the base station and a vehicle.

Fig. 7 is a block diagram representing the GPS Processing System.

Fig. 8 is a flow diagram illustrating communications in one embodiment of the GPS Processing System of Fig. 7.

Fig. 9 is a block diagram illustrating the Motion Positioning System (MPS) including the Inertial Reference Unit and the Odometer.

20 Fig. 10 is a block diagram illustrating one embodiment of the VPS System Architecture.

Fig. 11 is a detailed block diagram of the embodiment of the VPS System Architecture of Fig. Fig. 12 is a block diagram of one embodiment of the VPS Main Processor of 25 Fig. 10 showing the VPS Kalman Filter and the Weighted Combiner.

Fig. 13 is a flowchart of the Constellation Effects Method for improving position accuracy.

Fig. 14 is a polar plot illustrating the computed pseudo ranges from a four satellite constellation.

Fig. 15 is a flowchart of the original bias technique for Differential DCC:TJl:TG: 17679.SPC 5 ArU 1995 -6- Corrections.

Fig. 16 is a flowchart of the Parabolic Bias Method for Differential Corrections.

Fig. 17 is a flowchart of the BASE residuals as BIAS Method for Differential Corrections.

Fig. 18 is a flowchart of the method for Satellite Position Prediction.

Fig. 19 is a flowchart of the Weighted Path History Method.

Fig. 20 is a diagrammatic representation of the Weighted Path History Method.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENT TABLE OF CONTENTS I. General Overview II. Vehicle Positioning System (VPS) A. Overview B. Global Positioning System (GPS) 1. GPS Space Segment 2. GPS System Operation C. Motion Positioning System (MPS) D. VPS System Architecture 20 E. Base Station F. Satellite Based Accuracy Improvements 1. Constellation Effects 2. Differential Correction Techniques a. Original Bias Technique b. Parabolic Bias Technique c. Base Residuals as Bias Technique G. Weighted Path History I. General Overview In the following description of the present invention, and throughout the specification, the term "system" is used for shorthand purposes to mean apparatus, e oo oo *o 10': *o ft

O

DCC:TJII:TG:#17679.SPC 5 1pri 1995 -7method, or a combination of both apparatus and method.

Autonomous is used herein in its conventional sense. It indicates operation which is either completely automatic or substantially automatic, that is, without significant human involvement in the operation. An autonomous vehicle will generally be unmanned, that is without a human pilot or co-pilot. However, an autonomous vehicle may be driven or otherwise operated automatically, and have one or more human passengers.

The task of guiding an autonomous vehicle along a prescribed path requires that one accurately knows the current position of the vehicle. Once the current position is known, one can command the vehicle to proceed to its next destination.

For a truly autonomous vehicle to be practical, very accurate position information is required. In the past, it was not believed possible to obtain this nearly absolute position information without using a prohibitively large number of reference points. Ali positioning was done relative to the last reference point using inertial navigation or dead reckoning.

Using the systems disclosed herein we are able to obtain the nearly absolute position information required for a truly autonomously navigated vehicle. This led us to the development of simpler, more reliable vehicle controllers and related systems as well.

The development and implementation of the GPS (global positioning system) was a necessary and vital link which allowed us to develop our inventive systems ,•ooo to obtain more accurate position information. The GPS satellite position information, significantly enhanced through our inventive techniques, is combined and filtered with IRU (Inertial Reference Unit) inertial navigation information, and vehicle odometer information, resulting in a highly accurate system for determining position and effecting navigation.

Integration of both positioning and navigation systems, apparatus, methods and techniques provide for the highly accurate transportation of unmanned vehicles.

The navigation system more fully disclosed in Australian Patent No. 642638, using the highly accurate position information obtained from the positioning system, DCC:TJII:T :#17679.SPC 5 April 1995 -8is able to provide for accurate navigation between points along pre-established or dynamically generated paths.

Use is made of various models or conceptual representations to generate and follows these paths. For example, lines and arcs may be used to establish paths between objective points the vehicle is to obtain. B-splines or clothoid curves may be used to model the actual path the vehicle is to navigate. Using modelling or representational techniques provides for enhanced data communication, storage and handling efficiencies. It allows for simplification of supervisory tasks by providing a hierarchy of control and communication such that the higher the level of control, the simpler the task, and the more compact the commands.

At the base station, where GPS positioning data is also received from the satellites, a host processing system provides for the highest level of control of the navigation system. The host handles scheduling and dispatching of vehicles with much the same results as a human dispatcher would achieve. The host thereby determines the work cycle of each vehicle.

The host commands each of the vehicles to proceed from a current position to another position taking a specified route in order that they effect their work goals.

The host specifies the routes by name, and not by listing each point along the route.

Each vehicle has an on-board system which looks up the named route and translates the named route into a set of nodes and segments along the route, using, for example, the models or representations discussed above to connect the nodes.

The on-board system also provides for controlling the vehicle systems, such as brakes, steering, and engine and transmission, to effect the necessary physical acts required to move, stop, and steer the vehicle. These may be designed to be retrofitted onto existing vehicles, such as the Caterpillar, Inc. 785 off-highway truck, for instance.

The on-board system also checks actual position against desired position and corrects vehicle control to stay on route. The system may run multi-state models to enhance this checking capability. The system also checks for errors or failures of the system and vehicle components. If detected, the system provides for fail-safe DC:TJII:TG:#17679.SPC 5 Apll 199S -9shutdown, bringing the vehicle to a stop.

The on-board navigation system provides for different modes of control.

These include a fully autonomous mode, where navigation of the vehicle is automatically handled by the on-board system; a tele or re:mote control mode, where a remote human operator may control the direction and motion, and so on, of the vehicle; and a manual mode, where a human operator sitting in the cab can take control of the vehicle and drive it manually.

In the autonomous mode, obstacle detection is critical, and is provided for in the navigation system of the present invention. Boulders, animals, people, trees, or other obstructions may enter the path of a vehicle unexpectedly. The on-board system is capable of detecting these, and either stopping, or plotting a path around the obstruction, to return to its original route when it is safe.

Accurately tracking the desired route is another function of the on-board navigational system. System function anO architecture has been designed for real time tracking of the path at speeds up to approximately 30 mph. safely.

As an introduction to the next section, "POSITIONING," recall that the vehicle positioning system (VPS) of the present invention is a positioning system based on the incorporation of Inertial Reference Unit (IRU) Data, Odometer Data S..and Global Positioning System (GPS) Data.

20 The VPS incorporates these three subsystems with extensive innovative methodology to create a high accurate position determination system for moving or stationary vehicles, or any point on the Earth.

The GPS comprises a space segment and a land or atmospheric based processing system. The space segment includes 24 special purpose satellites (not yet fully implemented) which are being launched and operated by the U.S. Government.

These satellites continually transmit data to the Earth that can be read by a GPS receiver. The GPS receiver is part of a processing system which produces an estimate of the vehicle position based on the transmitted data.

When multiple satellites are in view, the GPS receiver can read each satellite's navigation messages and determine position based on known triangulation methods.

DCC:TJll;TGa#17679.SPC 5 April 1995 The accuracy of the process is in part dependent on how many satellites are in view.

The IRU comprises laser gyroscopes and accelerometers which produce position, velocity, roll, pitch and yaw data. The IRU combines this information into an estimate of vehicle position. The odometer produces data on distance travel that is incorporated with the IRU data.

A base station provides a geographic proximate reference point and reads the data transmitted by each satellite. Using the transmitted data, the base station makes improvements in accuracy on the estimate of the vehicle position.

A method is disclosed to predict the position of any satellite at the current time or any future time. Using that information, the GPS receiver can determine the optimum satellite constellation.

The present invention also includes a method for reducing vehicle "wandering" by using the path history of the vehicle to statistically determine the accuracy of future estimates of position.

The present application also includes numerous methods for determining how accurate the data transmitted from the satellite is. Included in these methods are techniques to compensate for data with error.

1I Positioning SA. Overview Referring now to Figs. 7 through 11, the present invention includes a Vehicle Positioning System (VPS) 1000 which is a highly accurate position determination system. The VPS incorporates three subsystems and extensive innovative methodology to create an unsurpassed position determination system for moving or stationary vehicles 310.

The VPS includes global positioning system (GPS) 700 and a motion positioning system (MPS) 900 including an inertial reference unit (IRU) 904, and an odometer 902 which have been enhanced and combined to produce a highly effective position determining system. The GPS 700 comprises a space segment and a land or atmospheric (for example, one located on an aircraft) based processing system.

The space segment includes 24 special purpose satellites (not yet fully implemented; DCC:Tjl I:TG:#17679.SPC 5 April 1995 11see Fig. 1) which are being launched and operated by the U.S. Government. These satellites constantly transmit data to the Earth that is read by a GPS receiver 706.

The GPS receiver 706 is part of a GPS processing system 700 which produces an estimate of the vehicle position based on the transmitted data. The vehicle's estimated position is transmitted to a VPS main processor 1004.

When multiple satellites are in view (see Fig. the GPS receiver 706 can read each of their navigation messages and compute the position Jf the antenna 702 using triangulation methods. The accuracy of the process is in part dependent on how many satellites are in view. Each satellite in view increases the accuracy of the process.

The IRU 904 comprises laser gyroscopes and accelerometers which produce position, velocity, roll, pitch, and yaw data. The IRU 904 combines this information into an estimate of the vehicle position. The odometer 902 produces distance traveled. The data from the IRU 904 and the odometer 902 are also transmitted to the VPS main processor 1004.

The VPS main processor 1004 combines the data from the MPS 900 (the IRU 904 and the odometer 902) with the estimated position from the GPS processing system 700 to produce a more accurate position.

Referring now to Fig. 3, a base station 314 provides a geographic proximate 20 reference point for the VPS 1000. The base station 314 includes its own GPS receiver 706 for reading the data transmitted by the satellites. Using the transmitted data, the base station 314 allows the VPS 1000 to make satellite based accuracy improvements on the estimated of the vehicle position.

The present invention includes a method (as shown in the flowchart of Fig.

18) to predict the position of any satellite at the current time or any future time.

Given an estimate of the position of the GPS antenna 702, the GPS receiver 706 can predict which satellites it will be able to read at any time. Using that information, the GPS receiver 706 can determine the optimum satellite constellation.

The present invention also includes a method for reducing vehicle "wandering" and spurious position computations. In this method, the path history or o oo ooro oo o o oooo oooo o oo o o oooo or oDo DCCfF1!I:T# 17679.PC 5 AprU 1995 -12 of the vehicle is used to statistically determine the accuracy of future estimates.

The present invention also includes a method for determining if the data received from the satellites is valid. The GPS receiver 706 determines the distance from the satellite to the GPS receiver 706 based on the data being transmitted by each satellite. The GPS receiver 706 compares this determined distance with an expected distance based on the time and an estimated position. If the distances are within a given range, then the data from satellites is assumed to be valid. Otherwise, the data is adjusted such that the determined distance between the satellite and the GPS receiver is within the given range.

B. Global Positioning System (GPS) I. GPS Space Segment As shown in Fig. 1, 24 man-made electronic satellites which make up the global positioning system (GPS) are planned for deployment by the early 1990s. As currently envisioned, the satellites will orbit the Earth at an altitude, of approximately 10,900 miles and encircle the globe twice a day. With this conventional GPS system, it will be possible to determine terrestrial position within 15 meters in any weather, any time, and most areas of the Earth.

As of the date of the filing of this application, there are known to be seven experimental GPS satellites in orbit, and several manufacturers that are building GPS 20 receivers. Until additional satellites are deployed and operational, there are only two windows each day when three or more satellites are available for position tracking.

The location of these satellites (and all others once deployed) is very predictable and can be plotted in terms of elevation and azimuth.

Reference is again made to Fig. 1 of \be drawings wherein the configuration of the fully operation-.0 GPS system is schematically illustrated. Twenty-four (twenty-one operational, three spare) medium orbiting satellites in six sets of orbits continuously transmit unique identifying signals on a common carriec frequencN Each of the 24 satellites transmits a unique navigation signal which can be used to determine the terrestrial position of an Earth antenna sensitive to the signal.

The navigation signal is comprised of a data bit stream which is modulated with a DCC:TJ11.TG:#17679.SPC 5 April 1995 13 pseudo-random type binary code which biphase modulates the carrier frequency.

In the coarse acquisition mode, each satellite has an established unique pseudo-random code, which is a gold code sequence having a length ot 1,023 chips that repeats itself once every millisecond. To facilitate separation of different satellite's signals, the gold code sequence for each satellite has a low correlation with the gold code sequence from the other satellites.

In the carrier mode, the data transmitted by the satellites is encoded on the carrier frequency in a manner similar to the C/A code. In the carrier mode, more data can be encoded and therefore more precise position computations can be made.

A GPS receiver can decode a navigation signal whose gold code sequence highly correlates to an identical gold code sequence generated by the GPS receiver.

The separation of the different satellite's signals from a common carrier is based on cross correlation of the signal with the locally generated gold code sequences, on a chip by chip basis, and then within a chip, until the maximum cross correlation value is obtained; the gold code sequence which maximizes this value can be used to extract the navigation data which is used in determining position of the receiving antenna.

Turning now to Fig. 2, a diagrammatic representation of the GPS in operation is shown. Four satellites 200, 202, 204, 206 comprise the current group (constel- 20 lation) of satellites in the view of the Earth antenna (user 210).

As is shown in the description block 208, each satellite is transmitting a navigation signal that includes timing (GPS time) and ephemeris (satellite position) data. Using the navigation equations 212, which are well-known in the art, and the timing and ephemeris data from the navigation signal, the position of the user 210 can be determined.

2. GPS System Operation Turning now to Fig. 6, a representative GPS system is shown in operation.

Four GPS satellites 200, 202, 204 and 206 are transmitting navigation signals in which satellite position (ephemeris) data and transmission time data are encoded.

Both a vehicle 310 and a base station 314 are receiving these signals from each of DCC:TJII:TG:#17679.SPC 5 April 1995 14 these satellites on their respective GPS antennas 312 and 316. In a preferred embodiment, both the C/A code and the carrier frequency are received at GPS antennas 312 and 316 for processing.

In addition to the four GPS satellites shown in the Fig. 6, a radio transmitter 624 is also depicted. This radio transmitter 624 is commonly known as a pseudolite.

These pseudolites, in one embodiment of the present invention, can be strategically placed around the perimeter of a mine pit and emulate the GPS satellites as shown in Fig. 6. This arrangement can be extremely useful in situations such as a mine pit or mine shaft, in which mining vehicles may be out of view of one or more of the GPS satellites, because of topographic features such as high mine pit walls. These ground based pseudolites provide additional ranging signals and can thus improve availability and accuracy of the positioning capability in the present invention.

The pseudolites are synchronized to the GPS satellites and have a signal structure that, while possibly different, is compatible with the GPS satellites. Note that when ranging to pseudolites, the ranging error does not include selective availability nor ionospheric errors. However, other errors must be accounted for such as tropospheric, pseudolite clock error and multipath errors.

In a deep pit surface mining operation, the view of the sky is limited by the walls of the mine and an inadequate number of satellites may be in view for position 0 determination. In such a case in the present invention, one or more pseudolites would be placed on the rim of the mine and used with a vehicle with the visible satellites to obtain accurate vehicle position.

Transmission channel 618 represents the electromagnetic communications channel linking the base station 314 and the vehicle 310. The transmission channel 618 is used to transfer data between the base station 314 and the vehicle 310. This transmission channel 618 is established by data-radios 620 and 622 which are transmitter/receivers.

The data-radios 620 and 622 are located at the base station 314 and vehicle 310 respectively, and are responsible for transferring various data between the base station 314 and the vehicle 310. The type of data transferred will be discussed DCC:TJII:TG:#17679.SPC 5 April 1995 15 further below. No physical medium (for example, copper wire, optical fiber) is necessary to conduct the transmission of data.

A radio transmitter/receiver which functions appropriately with the present invention can be found in the art. Such a preferred radio transmitter/ receiver is commercially available from Dataradio Ltd. of Montreal, Canada, Model Number DR-4800BZ.

Turning now to Fig. 7, a preferred embodiment of a GPS processing system 700 is shown. The GPS processing system 700 on the vehicle 310 includes a GPS antenna 702. In a preferred embodiment, the GPS antenna 702 is receptive to the radio spectrum of electro-magnetic radiation. However, the present invention contemplates reception of any signal by which GPS satellites might encode the navigation data.

An antenna receptive to the radio spectrum used by GPS satellites which functions satisfactorily with the present invention can be found in the art. Such a preferred antenna is commercially available from Chu Associates Inc. of Littleton, Massachusetts, Model CA3224.

The GPS antenna 702 is coupled to a pre-amplifier 704 so that the signals received at the GPS antenna 702 can be transmitted to the pre-amplifier 704. The present invention contemplates any method by which the GPS antenna 702 can be satisfactorily coupled to the pre-amplifier 704. Further, as will be noted at numerous places below, all devices in the GPS processing system 700, MPS processing system 900 and VPS processing system 1000 are coupled. The present invention contemplates any method by which these devices can be satisfactorily coupled.

Such coupling methods may include for example, electronic, optic, and sound methods as well as any others not expressly described herein. In a preferred embodiment, the coupling method used is electronic and adheres to any one of numerous industry standard interfaces.

The pre-amplifier 704 amplifies the signal received from the GPS antenna 702 so that the signals can be processed (decoded). The present invention DCC:TJII:TG:#17679.SPC 5 April 1995 16contemplates any method by which the received signals can be amplified.

A pre-amplifier which functions satisfactorily with the present invention can be found in the art. Such a preferred pre-amplifier is commercially available from Stanford Telecommunications Inc. (STel) of Santa Clara, California, Model Number 5300, Series GPS RF/IF.

The pre-amplifier 704 is coupled to a GPS receiver 706. The GPS receiver 706 decodes and processes the navigation message sent from the satellites in the view of the GPS antenna 702. During this processing, the GPS receiver 706 computes the latitude, longitude, and altitude of all satellites in the particular constellation being viewed. The GPS receiver 706 also computes pseudoranges, which are estimates of the distances between the satellites in the currently viewed constellation and the GPS antenna 702. In a preferred embodiment, the GPS receiver 706 can process, in parallel, pseudoranges for all satellites in view.

In a preferred embodiment of the present invention, the GPS receiver 706 produces this data when four or more satellites are visible. In a preferred embodiment of the present invention, the GPS processing system 700 can compute a position having an accuracy of approximately 25 meters when an optimal constellation of 4 satellites is in view. In another preferred embodiment of the present invention, the GPS processing system 700 can compute a position having an 20 accuracy of approximately 15 meters when an optimal constellation of 5 satellites is in view. An optimal constellation is one in which the relative positions of the satellites in space affords superior triangulation capability, triangulation technology being well known in the art.

In a preferred embodiment of the present invention, the GPS receiver 706 encodes in the outputted position data the number of satellites currently being viewed for each position computation made. In cases in which the number of satellites viewed for a series of position computations is four, the VPS weighted combiner 1204 (see Fig. 12 and discussion) uses only the first position computation received in the series. All subsequent position computations in the series are ignored by the VPS weighted combiner 1204. The VPS weighted combiner 1204 acts in this DCC:TJH:TG:#17679.SPC 5 Apnl 1995 17manner because position computations derived from four satellites are less accurate than nominally acceptable.

A receiver that functions satisfactorily in the present invention can be found in the art. Such a receiver is commercially available from Stanford Telecommunications Inc., Model Number 5305-NSI.

The GPS receiver 706 is coupled to a GPS inter communication processor 708. The GPS inter communication processor 708 is also coupled to a GPS processor 710 and a GPS Console 1 712. The GPS inter communication processor 708 coordinates data exchange between these three devices. Specifically, The GPS inter communication processor 708 receives pseudorange data from the GPS receiver 706 which it passes on to the GPS processor 710. The GPS inter communication processor 708 also relays status information regarding the GPS receiver 706 and the GPS processor 710 to the GPS Console 1 712.

An inter communication processor that functions satisfactorily with the present invention can be found in the art. Such a preferred inter communication processor is commercially available from Motorola Computer Inc. of Cupertino, California, Model Number 68000.

The GPS processor 710 is passed the satellite location and pseudorange data i* from the GPS inter communication processor 708. Turning now to Fig. 8, the *20 operation of the GPS processor is depicted. The GPS processor 710 uses methods to process this data including a GPS Kalman filter 802 (see Fig. 8) which filters out noise buried in the pseudorange data, including ionospheric, clock, and receiver noise. The GPS Kalman filter 802 also reads biases (discussed further below) transmitted by the base station 314 to the GPS processor 710.

Processor hardware that functions satisfactorily with the present invention as the GPS processor 710 can be found in the art. Such hardware is commercially available from Motorola Computer Inc., Model Number 68020. The software in the GPS processor 710, in a preferred embodiment, functions in the following way.

In a preferred embodiment, this GPS Kalman filter 802 is semi-adaptive and therefore automatically modifies its threshold of acceptable data perturbations, DCC:TH:TG:', ,679.SPC 5 April 1995 18 depending on the velocity of the vehicle 310. This optimizes system response and accuracy of the present invention as follows.

Generally, when the vehicle 310 increases velocity by a specified amount, the GPS Kalman filter 802 will raise its acceptable noise threshold. Similarly, wh,.n the vehicle 310 decreases its velocity by a specified amount the GPS Kalman filter 802 will lower its acceptable noise threshold. This automatic optimization technique of the present invention provides the highest degree of accuracy under both moving and stationery conditions.

In the best mode of the present invention, the threshold of the GPS Kalman filter 802 does not vary continuously in minute discreet intervals. Rather, the intervals are larger and, therefore, less accurate than a continuously varying filter.

However, the Kalman filter 802 of the present invention is easy to implement, less costly and requires less computation time than a continuously varying filter.

However, such a continuously varying filter could be used.

The GPS Kalman filter 802 must be given an initial value from the user at system start-up. From this value, and data collected by 'ti OK§ receiver 706, the GPS Kalman filter 802 extrapolates the current state (which :ieludes position and velocities for northing, easting and altitude) to the new "expected" position. This extrapolated position is combined with new GPS data (an update) to produce the 20 current state. The way the data is utilized is dependent on an a priori saved file called a control file (not shown). This file will determine how much noise the system is allowed to have, how fast the system should respond, what are the initial guesses for position and velocity, how far off the system can be before a system reset occurs, how many bad measurements are allowed, and/or bow much time is allotted between measurements.

The GPS processor 710 then computes position, velocity, and time using the filtered data and biases. However, the GPS processor 710 discards the computed velocity datum when the C/A code rather than the carrier is processed by the GPS receiver 706 because experimentation has shown that this datum is not accurate.

Velocity data derived from the carrier frequency is not discarded because it DCC:TJI ITO:#17679.SPC 5 April 1995 19 is much more accurate than the C/A code velocity data. The computed position and time data (and velocity data if derived from the carrier frequency) are encoded on GPS Signal 716 and sent on to the VPS main processor 1004 shown on Fig. In a preferred embodiment of the present invention, the GPS processor 710 reads both of these codes depending on the availability of each. Unlike data transmitted using the C/A code, the carrier frequency data transmitted by the satellite is available from the GPS receiver 706 at approximately 50 Hz (rather than approximately 2 Hz.) This increased speed allows the present invention to produce more precise position determinations with less error.

A preferred embodiment of functions of the GPS processor 710 is shown in Fig. 8. However, the press. 'tion contemplates any method by which data transmitted by GPS satellites can cessed. As shown at a block 816, a console function controls the operation of GPS console 2. This console function regulates the operation of the GPS Kalman filter 802 by providing the user interface into the filter.

The VPS communications function shown at a block 818, controls the outputs of the GPS Kalman filter 802 which are directed to the VPS system 1000. At a block 806, it is shown that the GPS Kalman filter 802 requests and decodes data from the GPS receiver 706, which data is routed through an IPROTO function 20 shown at a block 804. The IPROTO function is not a function of the GPS processor, as indicated on Fig. 8 by the dashed line. Rather, the IPROTO function resides on the GPS inter communications processor 708 and executes tasks associated with the GPS inter communications processor 708. An IPROTO function that operates satisfactorily with the present invention can be found in the art. One model is commercially available from Xycom Inc., model number XVME-081.

As shown at a block 810 the data transmitted over the transmission channel 618 is decoded and transmitted into the GPS Kalman filter 802. The communications manager function shown at a block 808, coordinates the incoming data from the IPROTO function. The communications manger function 808 also coordinates data received from an ICC function which is shown in a block 812. The DCC:TJH:TG:#17679.SPC 5 April 1995 20 ICC function 812 exchanges data with the data-radio 714 and the GPS data collection device 718 as shown.

The GPS console 1 712, is well known in the art. Many types of devices are commercially available which provide the desired function. One such device is commercially available from Digital Equipment Corporation of Maynard, Massachusetts Model Number VT220. The GPS Console 1 712 displays processor activity data regarding GPS inter communication processor 708, and GPS processor 710.

The GPS processor 710 is coupled to a GPS Console 2 722 and a GPS communications interface processor 720. The GPS console 2 722, is well known in the art. Many types of devices are commercially available which provide the desired console function. One such device is commercially available from Digital Equipment Corporation of Maynard, Massachusetts Model Number VT220. The GPS console 2 722 provides the user interface from which the GPS processor 710 can be activated and monitored.

The GPS communications interface processor 720 is coupled to a data-radio 714 and a GPS data collection device 718. The GPS communications interface processor 720 coordinates data exchange between the GPS processor 710 and both the data-radio 714 and the GPS data collection device 718. A communications 20 interface processor which functions appropriately can be found in the art. A preferred communications interface processor is commercially available from Motorola Computer Inc., Model Number MVME331.

The data-radio 714 communicates information from the GPS processor 710 (through the GPS communications interface processor 720) to a similar data-radio 620 located at the base station 314 (see Fig. In a preferred embodiment, the data-radio 714 communicates synchronously at 9600 baud. These data-radios provide periodic updates on the amount of bias (as detected by the base station 314) are transmitted to the vehicle 310 at a rate of twice a second. Base station 314 computed bias will be discussed further below.

The GPS data collection device 718 can be any of numerous common DCC:TJI:TG:#17679.SPCA S Apil 1995 21 electronic processing and storage devices such as a desktop computer. The International Business Machines Corporation (IBM) PC available from IBM of Boca Raton, Florida can be used.

C. Motion Positioning System (MPS) In a preferred embodiment, the pres( .vention also includes the combination of inertial reference unit (IRU) 904 an odometer 902 components.

These together with a processing device 906 comprise the motion positioning system (MPS) 900. IRUs and odometers are well known in the field, and are commercially available from Honeywell Inc. of Minneapolis, Minnesota, Model Number H61050- SR01, and from Caterpillar Inc. of Peoria, Illinois, Part Number 7T6337 respectively.

Turning now to Fig. 9, a preferred embodiment of the motion positioning system 900 is depicted. A Vehicle odometer 902 and an IRU 904 are coupled to the MPS inter communications processor 906.

The IRU 904 comprises laser gyroscopes and accelerometers of known design.

An IRU which can satisfactorily be used in the present invention is a replica of the system used by Boeing 767 aircrafts to determine position, except that the system used in the present invention has been modified to account for the lesser dynamics (for example, velocity) that the vehicles of the present invention will be exhibiting compared to that of a 767 aircraft.

20 In a preferred embodiment, the IRU 904 outputs position, velocity,, roll, pitch, and yaw data at rates of 50 Hz (fifty (50) times a second); the vehicle odometer 902 outputs distance traveled at 20 Hz.

The laser gyroscopes oi the IRU 904, in order to function properly, must be given an estimate of vehicle latitude, longitude and altitude. Using this data as a eve 25 baseline position estimate, the gyroscopes then use a predefined calibration in conjunction with forces associated with the rotation of the earth to determine an estimate of the vehicle's current position.

This information is then combined by the IRU 904 with data acquired by the IRU 904 accelerometers to produce a more accurate estimnate of the current vehicle position. The combined IRU 904 data and the vehicle odometer 902 data are DCC:TJIH:TG:#17679.SPC SApnril 1995 22 transmitted to the MPS inter communications processor 906.

The MPS inter communications processor 906 forwards the IRU and odometer data on to the VPS 1/O processor 1002 (see Fig. 10), as shown at signals 910 and 908, respectively.

An inter communications processor that functions satisfactorily with the present invention can be found in the art. Such a preferred inter communications processor is com:-.ercially available from Motorola Computer Inc., Model Number 68000.

The present invention contemplates any method by which the signals 716, 908 and 910 can be received by the VPS I/O processor 1002 from the GPS system 700 and MPS system 900 and forwarded on to the VPS main processor 1004. An I/O processor which functions satisfactorily with the present invention can be found in the art. Such an I/O processor is commercially available from Motorola Computer Inc., Model Number 68020.

15 D. Vehicle Positioning System (VPS) Architecture Turning now to Fig. 10, a preferred embodiment of the VPS system i architecture 1000 is depicted. Fig. 11 shows a diagram of the same VPS system architecture 1000, with the GPS processing system 700 and MPS processing system 20 900 in detail.

GPS processing system 700 and MPS processing system 900 are independently coupled to the VPS I/O processor 1002. Because they are independent, the failure of one of the systems will not cause the other to become inoperative. Thus, if the GPS processing system 700 is not operative, data will still be collected and processed by the MPS processing system 900 and the VPS 1000.

GPS processing system 700 and MPS processing system 900 transmit signals 716, 908, 910 to the VPS I/O processor 1002, as shown. These signals contain position, velocity, time, pitch, roll, yaw, and distance data (see Figs. 7 and 9 and associated discussions).

The VPS I/O processor 1002 is coupled to the VPS main processor 1004.

DCC:TJI1:TG:#17679.SPC 5 Apnl 1995 _I _ql I -23 The VPS I/O processor 1002 transmits signals 1006, 1008, and 1010 to the VPS main processor 1004, as shown. These signals contain the GPS, IRU and odometer data noted above.

Turning now to Fig. 12, a preferred embodiment of the operation of the VPS main processor 1004 is depicted. As shown, the GPS signal 1006, and the odometer signal 1008 are transmitted into a weighted combiner 1204. The IRU signal 1010 is transmitted into a VPS Kalman filter 1202. In a preferred embodiment, the GPS signal 1006 is transmitted at a rate of 2Hz; the Odometer signal 1008 is transmitted at a rate of 20Hz and; the IRU signal 1010 is transmitted at a rate of The VPS Kalman filter 1202 processes the IRU signal 1010 and filters extraneous noise from the data. The VPS Kalman filter 1202 also receives a signal from the weighted combiner 1204, as shown, which is used to reset the VPS Kalman filter with new position information.

The weighted combiner 1204 processes the signals and gives a predetermined 15 weighing factor to each datum based on the estimated accuracy of data gathering technique used. Thus, in the best mode of the present invention, the position component of the GPS signal 1006 is weighted heavier than the position component of the IRU signal 1010. This is because GPS position determination is inherently more accurate than IRU position determination.

20 However, velocity can be more accurately determined by the IRU, and therefore the IRU velocity component is weighted heavier than the GPS velocity component in the best mode of the invention.

The weighted combiner 1204 produces two outputs. One output contains all computed data and is sent to two locations: as shown at an arrow 1206, to the VPS Kalman filter 1202; and as shown at the arrow 1016, out of the VPS main processor 1004. The second output shown at the ar-ow 1018 contains only velocity data and sent out of the VPS main processor 1004 to the GPS processing system 700. The output shown at the arrow 1016 contains GPS time, position, velocity, yaw, pitch, and roll data and is transmitted at a rate of 20 Hz.

The present invention contemplates any method by which the signals 1006, DCC:TIII:TG:#17679.SPC 5 Apnr 1995 I~ I _alL- 24 1008, and 1010 can be processed at the VPS main processor 1004 in accordance ivith the above noted process steps. Processor hardware that functions satisfactorily with the present invention as the VPS main processor can be found in the art. Such hardware is commercially available from Motorola Computer Inc., Model Number 68020. The software in the VPS main processor 1004, in a preferred embodiment, functions as described above.

Referring now back to Fig. 10, the VPS main processor 1004 is coupled to a VPS communications interface processor 1020.

A communications interface processor which functions satisfactorily with the present invention can be found in the art. One preferred model is commercially available from Motorola Computer Inc., Model Number MVME331.

In a preferred embodiment, the VPS communications interface processor 1020 is coupled to, and routes the data contained in Output 1016 (at 20 Hz.) to, three different devices: a VPS console 1012, a data collection device 1014, and (3) 15 a navigation system 1022.

The VPS console 1012 is well known in the art, and is commercially available S.0. from Digital Equipment Corporation, of Minneapolis, Minnesota, Model Number VT220. This VPS console 1012 is used to display the current status of the VPS main processor 1004.

20 The VPS data collection device 1014 can be any of nunicrous common electronic processing and storage devices such as a desktop computer. The Macintosh computer available from Apple Computer of Cupertino, California, can be used successfully.

The navigation system 1022 comprises the features associated with the navigational capabilities of the present invention. The VPS system 1000 transmits its final data regarding vehicle position etc. to the navigation system 1022 at this point.

E. Base Station Referring back to Fig. 7, the present invention includes GPS components at the base station 314 that are identical to those which comprise the GPS pr cessing DCC;T II:TG: '679.SPC C .pril 1995 25 system 700 (as shown on Fig. The purpose of the base station 314 is to (1) monitor the operation of the vehicles, provide a known terrestrial reference location from which biases can be produced, and provide information to the vehicles when necessary, over a high-speed data transmission channel 618.

In a preferred embodiment of the present invention, the base station 314 will be located within close proximity to the vehicle 310, preferably within 20 miles.

This will provide for effective radio communication between the base station 314 and the vehicle 310 over the transmission channel 618. It will also provide an accurate reference point for comparing satellite transmissions received by the vehicle 310 with those received by the base station 314.

A geographically proximate reference point is needed in order to compute accurate biases. Biases are, in effect, the common mode noise that exists inherently in the GPS system. Once computed at the base station 314, these biases are then sent to the vehicle 310 using the data-radios 714 (as shown in Fig. The biases are computed using various methods which are discussed further below.

In a preferred embodiment of the present invention, a host processing system 402 is located at the base station 314 to coordinate the autonomous activities and interface the VPS system 1000 with human supervisors.

F. Satellite Based Accuracy Improvements 20 The present invention improves the position determining accuracy of the 11.chicle positioning system 1000 capability through a number of differential techniques. These techniques are designed to remove errors (noise) from the pseudoranges that are calculated by the GPS receiver 706. The removal of this noise results in more precise position determination.

In a preferred embodiment, the base station 314 GPS processing system 700 is responsible for executing these differential techniques. The term "differential" is used because the base station 314 and the vehicle 310 use independent but virtually identical GPS Systems 700. Because the base station 314 is stationary and its absolute position is known, it serves as a reference point from to measure electronic interference (noise) and other error inducing phenomena.

DCC:TJlI:T:17679.SPC 5 Aprl 1995 -26 1. Constellation Effects Fig. 13 depicts a flowchart which shows the steps required to make the best use of the satellite constellations which are in view of the GPS antenna 702. For a given vehicle 310, there may be numerous satellites in view of which only a subset will form a particular constellation. The constellation effects flowchart 1300, describes how the optimal constellation can be chosen depending on the satellites currently in view, and the desired path that the vehicle 310 is taking.

The beginning of the constellation effects flowchart 1300 is shown at a block 1302. The first step in the flowchart is shown at block 1304. The positions of each satellite in view of the GPS antenna 702 are computed using almanac data and GPS time. Almanac data refers to previously recorded data which stores GPS satellites' positions at specific times during the day. Because the operators of the GPS satellites know the course and track across the sky that the satellites take, this information is available.

15 The positions of the satellites are represented by polar coordinates which correspond to azimuth and elevation from the base station 314. Once these satellite :positions have been determined, the satellites are then mapped into a polar map with the estimated vehicle position in the center of the map. This polar map is illustrated in Fig. 14. As shown in Fig. 14, four satellites 200, 202, 204, and 206 are depicted.

20 As shown in the next block 1308, circles are drawn around each of the satellites so that the circles form an intersection at the center of the polar map as shown in Fig.

14. The error input position determination can be shown as the inner section of the circles, which resemble an ellipsoid.

Ana important parameter in the ellipsoid representation is the ratio between the semi-major and semi-minor access of the ellipsoid, called the geometric ratio of access factor (GRAF). This factor is used along with the angle of the major access to compute a weighing factor, which in effect assists the GPS system 700 to compute a more accurate position. This is done by modifying the GPS Kalman filter 802 in the vehicle GPS processing system 700 to accommodate the shape of the estimated ellipsoid and the computed northing-easting coordinates of the vehicle, as DC:TJII:TG:#17679.SPC 5 April 1995 27 shown at a block 1310 and Fig. 14.

As shown at a block 1312, these steps are taken with respect to all possible constellations in v;ew of the GPS antenna 702. After all possible satellite constellations have been computed and plotted, the optimal satellite constellation for the sired vehicle path is determined as shown at a block 1312. The optimal constellation will be one that gives the least error perpendicular to the desired vehicle path as shown at a block 1312. As shown at a block 1314, the optimal satellite constellation data is transmitted to the vehicle 310 over the data radio 714.

2. Differential Correction Techniques a. Original Bias Technique Referring now to Fig. 15, an original bias flowchart 1500 is depicted. As shown at a block 1502, the original bias flowchart computation begins. As shown at a block 1504, the pseudorange (PSR) for each satellite in view of the GPS antenna 702 is computed. The pseudoranges are computed by the GPS receiver 706 15 located at the base station 314. The block 1504 also shows that the positions of each satellite (SV where SV indicates position, and indicates a particular satellite for which this data is being computed, is determined.

The satellite positions are determined independently of the pseudorange computations made by the GPS receiver 706. The satellite positions are determined by decoding the GPS time data which is encoded in the navigation signal transmitted from each satellite. This time of transmission is the time the satellite actually transmitted the navigation sequence which was received by the GPS receiver 706.

Because the p 'h of the satellites is known and recorded in the almanac as described above, the position of the satellite can be determined by correlating the GPS transmission time with the almanac data. Thus, for time T1 the polar coordinates of the satellite can be determined through a correlation table in the almanac.

After the steps in the block 1504 have been completed, the known base station 314 position is subtracted from each satellite position as shown in a block 1506. Because the position of the satellites and base station 314 are in polar DCC:TJIITG:117679.SPC 5 Apni 1995 28 coordinate form, a polar coordinate equation using the euclidian norm, which is well known in the art, must be used to make this subtraction. The actual equation subtracts the polar coordinates of the base station 314 from the polar coordinates of the satellite of which difference the euclidian norm is computed. The square root of the euclidian norm is then computed.

The steps shown in the block 150' produce the actual range of the satellite from the GPS antenna 702 located on the base station 314. As shown in a block 1508, the pseudorange for each satellite (computed by tbh GPS receiver 706 using the GPS time) and the satellite's clock bias are subtracted from the actual range determined in the block 1506. This step produces the actual bias (error) of the pseudorange computation made by the GPS receiver 706. This error is caused by many different effects, such as atmospheric conditions, receiver error, etc.

Because the vehicle 310 is in close proximity to the base station 314, the error in pseudorange computation is assumed to be identical. Therefore, the pseudorange error which has been determined as shown in the blocks 1504, 1506 and 1508 can be used to modify the position estimates produced by the vehicle 310 GPS .processing system 700. Of course, this error calculation could not be made at the vehicle 310 GPS processing system 700, because the actual position of the vehicle is not known, whereas the actual position of the base station 314 is known.

20 As shown at a block 1510, the base station 314 GPS Kalman filter 802 is updated with the bias information. Finally, as shown as a block 1512, the base station 314 computed biases are transmitted to the vehicle 310 using the data radios 620 and 622.

The biases are then used by the vehicle 310 GPS processing system 700 to provide more accurate position determination by updating the vehicle 310 GPS Kalman filter 802.

b. Parabolic Bias Technique As the GPS satellites rise and fall in the sky, the path formed by each satellite follows a parabola. Therefore, a parabolic function can be computed which represents the path of each satellite in the sky. Turning now to Fig. 16, part of the DCC:TJH;TG:#17679.SPC 5 Api 1995 29 flowchart 1600 shown describes a method in which such a parabolic function (model) is computed for each satellite in the view of the base station 314 GPS antenna 702.

Referring now to Fig. 16, a block 1602 shows that the flowchart begins. As shown at a block 1604, at a time pseudoranges are determined for each satellite in view of the base station 314 GPS antenna 702, using the GPS receiver 706, as described above. As shown at a block 1606, the pseudoranges (for each satellite) are incorporated into parabolic best fit models for each satellite. Thus, one point on the parabolic model for each satellite is added.

As shown at a block 1608, a test is made as to whether enough points on the parabolic model have been determined to estimate a parabolic function which the satellite would conform to along the rest of its path. The number of points that have been collected will determine a particular statistical R squared value. As shown at the block 1608, if this R squared value is greater than 0.98, then the parabolic model 15 is deemed to be accurals enough to estimate the future path of the satellite. If the R squared value is less than or equal to 0.98, then more points on the parabolic model must be computed. These points are computed by incorporating the pseudorange data which is continually being computed by the GPS receiver 706.

As shown at a block 1610, the N value increments to show that the time at 20 which the pseudorange is computed, as shown in the block 1604, has increased.

Because the GPS receiver 7G5 outputs pseudorange data for each satellite twice a second, each N increment should represent approximately one half second.

If enough data points have been collected such that the R squared value is greater than 0.98, then, as shown in a block 1612, the parabolic model is deemed accurate enough to represent each satellite's orbital path. As shown in the block 1612, the parabolic model represents points on the past and future satellite path.

Now that the parabolic model is complete, future points on the model can be calculated as shown at a box 1614. As shown at the block 1614, for the time T(n) the locus point on the parabolic model is computed. Here the locus point is the expected location of the satellite at time Once this locus point is computed, DCC:TfJI:TG:# 17679.SPC 5 April 1995 I 30 the range for the locus point (distance between the GPS antenna 702 and the satellite) is computed, as shown at a block 1616 As shown at a block 1618, the pseudorange is then computed for time T(n+l) (the current time). The pseudorange is computed by the GPS receiver 706 as described elsewhere in this application. This pseudorange is also incorporated into the parabolic best fit model, as indicated at an arrow 1626. As shown at a block 1620, the pseudorange computed at time T(n+l) and tile base station 314 clock bias are subtracted from the parabolic locus point range to get the bias. The bias data are then smoothed in accordance with methods well known in the art, as shown at a block 1622. This bias is then transmitted to the vehicle 310 using the data radio 714, as shown at a block 1624.

c. Base Residuals as Bias Technique Fig. 17 uses base residuals to compute the bias. A base residual is the difference in the actual polar coordinate position of the base station 314 and the 15 position which is computed by the GPS processing system 700 on the base station 314. To illustrate how this functions, assume the base station 314 is at the corner of Elm and Maple streets. Also assume the base station 314 GPS processing system 700 estimates the position of the base station 314 to be four miles due south of the actual known position of the base station (the corner of Elm and Maple). Then it is obvious that the bias is a distance equal to four miles in a due south direction.

Because the GPS processing system 700 on the vehicle 310 is identical to the GPS processing system 700 on the base station 314, the four mile error in computation can be deemed to be occurring at the vehicle 310 as well as the base station 314. The vehicle 310 can then use this information in its GPS processor 710.

In effect, the GPS processor on the vehicle 310 will modify its position computations to account for a four mile due south errr in the data. Thus, the position computed by the vehicle 310 GPS processor 710 will reflect a position which is four miles north of the position computed by the vehicle 310 GPS processor 710 which used the pseudorange computed by the vehicle 310 GPS receiver 706.

Turning now to Fig. 17, a block 1704 shows that the exact polar coordinates DCC:TJH:TG:#17679.SPC S Apn1 199S 31 xO, yO, z0 of the base station must be obtained. The pseudoranges for the satellites which are in view of the GPS antenna 702 are then computed. If the GPS receiver 706 on the vehicle 310 is configured to read data from a particular constellation of satellites (not necessarily all which are in view), then the GPS receive:" '06 at the base station 314 must also be configured to use the same constellation.

The location of the base station 314 is then computed by the GPS processor 710 located at the base station 314. If there is a difference in this computed base station 314 location and the actual known base station 314 location, (such as the four miles in the above example), then these differences, called residuals, as shown at a block 1710 represent the new biases for the vehicle 310. As shown at a block 1712, the residuals are then outputted to the vehicle 310 over the data radio 714 to be processed at the GPS processor 710.

G. Weighted Path History The weighted path history technique of the present invention improves the positioning capability of the present invention. The weighted path history technique ""depicted in Fig. 19, uses previous path positions to provide an estimation of future positions. Use of this technique results in a reduction of position "wandering" and enhanced immunities to spurious position computations. Wandering describes the tendency of the vehicle positioning system 1000 to determine positions that deviate 20 from the actual path of the vehicle 310.

Turning now to Figs. 19 and 20, the weighted path history technique is :i described. At a block 1902, the weighted path history flowchart begins. The position of the vehicle 310 is computed at a block 1904 by the vehicle positioning system 1000 as can be seen in Fig. 20, for vehicle positions, 2002, 2004, 2006, 2008, and 2012 are diagrammed. The vehicle 310 moves along the path 2022 which is created by the sequence of points just noted.

Turning back now to Fig. 19, at a block 1906, each new position -omputation is incorporated into the weighted path history best fit algorithm. At a block 1908, the R square value is shown to be compared against 0.98. This R squared value represents the number of position points that have been taken thus far, and therefore DCc:TH:TG:#17679,SPC 5 Ap' 1995 32 how statistically accurate a future estimation can be. The window referenced in the block 08 refers to the previous number of computed position points.

If the R squared value of the new window is not greater than or equal to 0.98, then a test is made at a block 1910 to see if more than 20 position points have been computed. If there have been more than 20 points collected, then the window is restarted as shown at a block 1914. The window is restarted in this case because, if the R squared value is less than 0.98 after more than 20 points have been collected, the collected points are deemed to be inaccurate and are therefore not relied upon. To restart the window means that all new data points will start being incorporated into the best fit algorithm.

If lesb than or equal to 20 position points have been calculated, then the window is not restarted. Rather, the raw position is outputted as shown at a block 1912. This raw position is simply the position that was computed as is shown in the block 1904. If there were not more than 20 points collected, then it is assumed that not enough points have yet been collected to produce a R squared value greater than or equal to 0.98. After the window is restarted as shown in a block 1914 the raw position is also output as shown in the block 1912.

.go• If the R squared value of the new window is greater than or equal to 0.98, then as shown in a block 1916, the output position is modified to be the best fit 20 prediction. This is depicted in Fig. 20 at a position point 2010. Position point 2010 represents the position point as computed by the vehicle positioning system 1000.

According to the weighted path history technique, the position point 2010 is different from the expected position point 2006. Therefore, the actual position outputted is modified from the position point 2010 to become thte position point 2006. This assumes that there were enough position points previously computed so that the R squared value of the new window was greater than or equal to 0.98 as required in the block 1908 of Fig. 19. Once the position point 2010 is modified, it is outputted to the navigation system 1022.

DCC:TJII'T:#17679.SPC 5 Apfl 1995

Claims (4)

1. A system for determining an estimate of the position of a vehicle based on pseudoranges derived from a vehicle positioning system and prior position estimates, the system comprising: first means for receiving signals from a plurality of satellites, responsively determining respective pseudoranges from said satellites to the vehicle, and deriving a first position estimate of the vehicle's position as a function of said pseudoranges; second means responsive to said first means for receiving the prior position estimates and responsively determining a vehicle path model using a best fit algorithm; third means responsive to said second means, said third means for S. S. extrapolating a second position estimate from said vehicle path model; and fourth means responsive to said third means, said fourth means for 15 setting said first position estimate equal to said second position estimate if said second position estimate is more accurate.

2. A method for determining an estimate of the position of a vehicle based on a vehicle position system and prior position estimates via a best fit algorithm, the 0b a method comprising the steps of: 20 receiving signals from aplurality of satellites, responsively determining respective pseudoranges from said satellites to said vehicle, and computing a first *55'05 position estimate using the vehicle positioning system as a function of said pseudoranges; determining a vehicle path model using the best fit algorithm as a function of the prior position estimates; extrapolating a second position estimate from said vehicle path model; and setting said first position estimate equal to said second position estimate if said second position estimate is more accurate.

3. A method for improving the accuracy of position estimates of a vehicle based DCC:TJHiTG.# 17679RS I 31 July 1996 C- I -34- on a vehicle positioning system by considering prior position estimates via a best fit algorithm, the method comprising the steps of: computing positions using the vehicle positioning system; incorporating said positions into a best fit algorithm; extrapolating an estimated subsequent position from said best fit algorithm; and substituting a subsequent position computed by said vehicle positioning system with said estimated subsequent position extrapolated from said best fit algorithm. The method of claim 3, further comprising the step of outputting said estimated subsequent position to a vehicle navigator. The system as claimed in claim 1 substantially as hereinbefore described with reference to the accompanying figures.

6. The method as claimed in claim 2 or claim 3 substantially as hereinbefore described with reference to the accompanying figures. DATED: 31 July 1996 CARTER SMITH BEADLE Patent Attorneys for the Applicant: CATERPILLAR INC. D o P O 0 C DItC:TJH:TG:#17679 RS I 31 July 1996 ABSTRACT A weighted path history technique uses previous path positions of a vehicle (310) to provide an estimation of future positions. The position of the vehicle (310) as it moves along a path is computed by a first means (1000). The position data is incorporated into a weighted path history best fit algorithm. An estimated subsequent position is extrapolated from the best fit algorithm. The subsequent position calculated by the first means (1000) is substituted with the estimated subsequent position derived from the best fit algorithm. *e *0 0**t0 0 a e DCC:TJIH:T:#N17679.SPC 5 April 1995