Showing papers on "GPS/INS published in 1991"

The Theory of the GPS Composite Clock

Abstract: In essence, GPS is a network of atomic clocks related by measured signal transit times, specifically receive times measured by ground station clocks minus transmit times measured by satellite clocks. Such measurements allow both orbit and clock estimation, but since clocks appear only in differences, any error common to all the clocks would remain unobservable. Early GPS designs therefore treated one of the clocks as a master, “correct” by definition and dropped from the problem. Although such artificial reliance on a particular clock was obviously undesirable, suggestions of clock ensembling were inhibited by a lack of theory. The combined orbit/clock problem clearly required a Kalman filter or the equivalent, and that seemed to demand a completely observable system. This paper presents the solution (in use since June 1990) known as the GPS Composite Clock. Essentially it is a Kalman filter applied to the original orbit/clock problem, including all clocks, treating none as master. There are no explicit ensembling equations, and the system is not completely observable, but a series of new theorems shows that (a) the Kalman filter implicitly creates an ensemble time from all the clocks, and (b) the unobservability can be handled in any of several acceptable ways. A concept of transparent variations is developed to show the role of unobservable dimensions, and permit a dynamic separation of covariance into unobservable and observable parts. The observable part is recognized as the covariance of the various clocks about an implicit ensemble mean of them all, and the unobservable part as the covariance of the implicit ensemble mean relative to an ideal clock. To within the small error levels of the observable parts of the system, every corrected clock (physical clock corrected by its estimated error) behaves as if controlled by the implicit ensemble mean. Thus a corrected clock represents the implicit ensemble mean (to within such errors) and has the same long-term stability. For example, long term Allan variances are typically reduced, relative to stand-alone values, by a factor of the effective number of clocks.

Kalman filter mechanization for INS airstart

Abstract: A strapdown mechanization and associated Kalman filter were developed to provide both ground align and airstart capabilities for inertial navigation systems using Doppler velocity and position fixes, while not requiring an initial heading estimate. This is accomplished by use of a Doppler velocity sensor and a position source such as the global positioning system (GPS) manual fly-over update or target sighting systems. Filter transition from coarse to fine align mode is accomplished without disrupting the estimation of the inertial instrument or aiding sensor errors, by defining the azimuth error state as wander angle error, and using a simple, but effective, manipulation of the filter covariance matrix. >

Autonomous reconfigurable GPS/INS navigation and pointing system for rendezvous and docking

Abstract: This paper describes the results of an integrated navigation and pointing system software development effort sponsored by the NASA MSFC through a SBIR Phase 2 Program. The integrated Global Positioning System (GPS)/Inertial Navigation System (INS) implements an autonomous navigation filter that is reconfigurable in real-time to accommodate mission contingencies. An onboard expert system monitors the spacecraft status and reconfigures the navigation filter accordingly, to optimize the system performance. The navigation filter is a multi-mode Kalman filter to estimate the spacecraft position, velocity, and attitude. Three different GPS-based attitude determination techniques, namely, velocity vector matching, attitude vector matching, and interferometric processing, are implemented to encompass different mission contingencies. The integrated GPS/INS navigation filter will use any of these techniques depending on the mission phase and the state of the sensors. The first technique, velocity vector matching, uses the GPS velocity measurement to estimate the INS velocity errors and exploits the correlation between INS velocity and attitude errors to estimate the attitude. The second technique, attitude vector matching, uses INS gyro measurements and GPS carrier phase (integrated Doppler) measurements during a spacecraft rotation maneuver to determine the attitude. Both of these techniques require only one GPS antenna onboard to determine the spacecraft attitude. The third technique, interferometric processing, requires use of multiple GPS antennae. In order to determine 3-axis body attitude, three GPS antennae (2 no-coplanor baselines) are required.

GPS/INS integration for civil aviation

Abstract: The mechanization equations for generating the INS (inertial navigation system) solution in the proper coordinates are presented. Two different Kalman filter error models for integrating GPS (Global Positioning System) with INS are derived. The model which uses spheroidal component differences rather than Cartesian vector differences is preferable, since it uses error states which can be directly corrected in the INS mechanization equations. >

GPS Inertial Attitude Estimation via Carrier Accumulated-Phase Measurements

Abstract: This paper describes the design and simulation of an integrated GPS/INS that accomplishes space vehicle navigation and attitude estimation. The simulation contains models of an inertial measurement unit (IMU), a GPS receiver processing signals from an array of antennas placed on the spacecraft structure, and an integrated navigation filter. Interferometric equations are employed to process GPS carrier accumulated-phase measurements from precise antenna locations in the user vehicle. Vehicle orientation in inertial space is then obtained by relating baseline-difference phase measurements to orientations of antenna baselines relative to satellite line-of-sight vectors. The method of GPS-aided attitude estimation described herein is distinguished from other methods in that it does not employ search algorithms to resolve the carrier phase cycle ambiguity. Instead, the ambiguity is modeled as a random process and is included as an error state of a statistical filter. As a result, the attitude estimation time history evolves directly for any dynamic environment.

Graceful Degradation of GPS/INS Performance With Fewer Than Four Satellites

Abstract: Precision first pass weapon delivery with exceptionally accurate navigation are vital to the operational survivability of attack aircraft. Systems such as the Global Positioning System (GPS) are currently being fielded and integrated with inertial navigation systems (INS) to support such a capability. It offers the potential for less costly weapon targeting systems if target coordinates relative to the GPS reference can be provided. The current standard GPS receiver requires simultaneous signals from 4 satellites (with good geometry) to calculate and output useful position and velocity information. However, signals from 4 satellites may not always be available, especially in and around heavily defended target areas protected by high power jammers. These high power jammers, even when recognized and appropriately nulled out by adaptive array GPS antenna, can result in large sectors of the sky surrounding these jammers no longer being monitored for any GPS satellite signals. This paper recommends changes to the existing GPS equipment and new aircraft Kalman filter algorithms which provide graceful degradation of aircraft navigation performance when only one, two, three, or intermittent satellite signals are available, and quantifies the level of relative performance expected for these conditions. This paper also analyzes the relative performance of various Kalman filter mechanizations as a function of filter size.

High-precision real-time gps positioning concepts: modeling and results

Abstract: GPS has been used successfully for a wide range of high-precision surveying applications Most precise positioning results have been achieved through postprocessing of data This denies the use of GPS for those surveying and navigation applications which require on-line position computation A Kalman filtering scheme is presented which is designed to perform high-integrity, real-time GPS relative positioning in static and kinematic modes to centimeter-level accuracy A continuous-kinematic data set for a train has been processed as an example to explain the operation of the system and illustrate achievable accuracy Results of the test proved that the repeatability of vertical positions for a train traveling at 120 km/h was better than 5 cm

A Discussion of GPS/INS Integration for Airborne Photogrammetric Applications

Abstract: The concept of using integrated GPS/INS for precise aircraft positioning is discussed and the integration algorithm is briefly reviewed. Airborne tests jointly conducted by the University of the Federal Armed Forces, Germany, The University of Calgary, and the Rheinbraun Company, Germany are used to assess the accuracy of the integrated system. Two Trimble 4000SX receivers, an LTN 90–100 strapdown inertial system and a Zeiss RMK 15a/23 photogrammetric camera were used in the experiment allowing for a comparison of perspective centers of the camera determined from traditional photogrammetric techniques and from GPS/INS integration. Results show that decimetre accuracy is achievable. The accuracy of the integrated system for attitude determination is also assessed but a conclusive result is not possible at this time. Relative angular velocities between the camera and the INS may have caused discrepancies in the respective attitudes.

Use of Distance-measuring Equipment (DME) for Correcting Errors in Position, Velocity, and Wind Measurements from Aircraft Inertial Navigation Systems

Abstract: Aircraft distance-measuring-equipment (DME) data are used to update position, velocity, and wind measurements from inertial navigation systems (INS) measurements Data from conventional single-channel DME sets, suitably calibrated, are shown to be adequate to resolve the Schuler oscillation and correct INS positions to better than 1-km accuracy The satellite-based NAVSTAR global position system (GPS) is rapidly superseding other systems for external position reference However, DME is reliable and very accurate and has been recorded on many research datasets The principal limitation of the DME is that it is restricted to land-based navigation The regression technique used does not necessitate multiple DME receivers or station switching and involves few restrictions on the collection of the data However, the results improve when more than one station is used Comparisons with other navigation systems (interferometer and loran) demonstrate the method's skill in resolving INS errors Intercompar

Vehicle position calculator

Abstract: PURPOSE:To maintain continuity of the travel trace of a vehicle in a vehicle position calculator of a navigation system to be mounted on a vehicle and also ensure high vehicle position accuracy by resetting to a vehicle position measured by a GPS when there is a shift in position due to a sensor error. CONSTITUTION:A vehicle position calculator comprises a speed sensor 11 and an angular speed sensor 12, a GPS receiver 13, an estimated position calculating means 14, map data 15, a candidate road searching means 16, a vehicle position correcting means 17 and a vehicle position resetting means 18. The respective sensors 11, 12 are used to estimate the vehicle position, the vehicle position is corrected to a candidate road whose feature coincides with a vehicle travel trace based on the data, and only when an accumulated value of corrected distances by position correction of a constant number of past corrections or correction in a constant travel area on the candidate road reaches a threshold or higher, the vehicle position is reset to a position measured by the GPS receiver 13.

On-satellite position measuring system filter reset system

Abstract: PURPOSE:To prevent the dispersion of filter estimation when it is conducted with the information obtained from a GPS receiver loaded on an artificial satellite. CONSTITUTION:A GPS receiver 2 provided on an artificial satellite 8 receives the radio waves transmitted from NAVSTARs 4, 5, 6, 7 through an antenna 3 and calculates the pseudo distances between itself and the NAVSTARs and their change ratios. The data measured by the GPS receiver 2 is transmitted to a position measurement arithmetic processor 1 to conduct the position measurement of the self position by one-point position measurement and a filter. At this time, by adapting this reset system, the estimating calculation of the filter is regularly operated, and one forming a precision index is outputted. The resulting position measurement result is outputted to a position measurement result utilizing equipment 9. Thus, a system such that the orbit estimation result of the filter is never dispersed can be constructed. A value showing the certainty of the estimated value can be outputted to the filter estimation which has no method for knowing the certainty of the estimated result.

Performance Analysis of Integrated GPS/INS Guidance for Air-to-Ground Weapons in a Jamming Environment

Abstract: This paper addresses several findings of a recent study undertaken by Systems Control Technology, Inc. and supported by Mayflower Communications Company. A phase 1 study [l], evaluated candidate low-cost integrated Global Positioning System and Inertial Navigation System (GPS/INS) configurations for guided standoff weapons. The focus was directed toward three low-cost GPS/INS concepts. The first uses a receiver on-board the vehicle and is an autonomous GPS/INS guided weapon which is initialized and launched from a delivery aircraft. The second and third configurations use a GPS translator on-board the weapon to re-transmit the GPS signals to a GPS receiver on the aircraft, or to a ground station operating in a differential mode. The Phase 2 effort reported here examines the issues related to integration techniques and EW vulnerability. Only the autonomous configuration was investigated in Phase 2. An analysis was conducted of the jamming vulnerability of various receiver/INS configurations by evaluating the performance degradation in the presence of jamming using two simulations. In the first simulation, actual real-time GPS receiver tracking loop software was used to generate a functional relationship between jammer power level and RMS tracking error, for two configurations. One, an integrated design and the second a "stand-alone" receiver, one navigation without the benefit of aiding obtained from an inertial system. This numerically derived functional model was incorporated in teh second simulation which utilized both covariance propagation equations and measurem,ent "aiding", or updates, to the INS based navigator. The measurement noise used to update the Kalman filter equations directly utilized the functional model obtained from the GPS receiver trackign loop. The covariance simulation was "flown" along a trajectory and used to generate CEPs for various weapon/jammer configurations. The paper is organized as follows: the background section touches on points of general interest and sketches general requirements, integration issues are highlighted next, followed by a simulation section which discusses how the analyses for these studies were conducted. The results are presented next, and the final section contains the conclusions.

GPS/INS Trajectory Determination for Highway Surveying

Abstract: The conventional way of surveying is the establishment of the control network and then surveying the densifying points. This has a major drawback in a number of applications — its discrete feature. The determined positions relate uniquely to the measured points while positions outside these points have to be interpolated using certain terrain representation. In some applications, the exact determination of geometrical characteristics is crucial.This includes road and railroad profiling, pipeline profiling, determining the verticality of mining shafts. Applying the conventional methods for such tasks is very tedious and in some cases almost impossible. It was the reason while geodesists were dreaming long time about some kind of tool which would enable them nearcontinous positioning with sufficient level of accuracy.

Development of a TRN/INS/GPS integrated navigation system

Abstract: The main objective of the work reported was to develop and evaluate an algorithm merging data from a terrain reference navigation system (TRN), an inertial navigation system (INS), and a Global Positioning System (GPS) receiver. This was accomplished by merging all available sensor data in one large Kalman filter. The navigation sensors were regarded as smart sensors, providing not only positional data but also an associated covariance matrix. The integration of these navigation sensors was accomplished around a MIL-STD-1553B data bus. Initial results show that the integrated navigation system showed good performance. In the case that the GPS was jammed and the TRN had a low accuracy, due to flying over flat terrain, the sensor models kept the accuracy of the integrated system to an acceptable level. >

A Navigation Reference System (NRS) Using Global Positioning System (GPS) Aiding

Abstract: : To quantify the performance abilities of existing or proposed navigation systems, the U.S. Air Force has for the last several years compared the performance of the system under test to the performance of a baseline navigation system known as the Completely Integrated Reference Instrumentation System (CIRIS). CIRIS obtains a highly accurate navigation solution by combining the output from three major subsystems: inertial navigation system (INS) information, barometric altitude information, and range and range-rate data from ground transponders which have been precisely surveyed. Although the navigation solution produced by CIRIS is highly accurate, it will soon be inadequate as the standard against which future navigation systems can be tested. This research proposes an alternative to CIRIS - a hybrid Navigation Reference System (NRS) which is designed to take advantage of a newer INS (the LN-93), certain features of the current CIRIS, and certain features of the Global Positioning System (GPS). Analysis is conducted using a Kalman filter development package known as the Multimodel Simulation for Optimal Filter Evaluation (MSOFE). Both a large order truth model for the NRS (in which a full 24 satellite constellation is modeled) and a full-order Kalman filter are developed. Results suggest that the proposed NRS (with GPS aiding) provides a significantly improved navigation solution as compared to CIRIS.

Ship navigation for geophysical applications using Kalman-filtered GPS fixes

Abstract: Based on a state vector representation of ship position and velocity, we have developed a Kalman filter that provides accurate navigation from position fixes supplied by a Global Positioning System (GPS) receiver. Quasi‐constant position offsets occurring for fixes associated with switches in observed satellite constellation are modeled by including constellation biases as state vector components. A proper choice of statistics for state propagation and measurement noise leads to improved positioning. However, it may also increase the statistical dependence of Kalman estimates over characteristic time periods. Since a quantitative measure of this dependence is generally important for further processing, we have derived general expressions for the autocovariance of state vector Kalman estimates. These expressions contain products of propagation matrices and Kalman gains and, hence, relate directly to stability properties of the filter. For many applications, only certain components of the state vector are o...

A new way of integrating GPS with INS

Abstract: Direct use of INS to determine GPS integrity will not meet the accuracy requirements for the non-precision approach phases of flight because gradual drift in the GPS signal cannot be distinguished from INS drift. Rather than directly integrating GPS with INS, the inertial system can instead be used to give the external Integrity Monitor (IM)system (such as SATZAP), more time to react. Inertial system errors grow very slowly for the first few minutes after initial alignment or calibration by an external reference such as GPS. After calibration, one can “coast” for several minutes using the INS by itself without GPS but with very little loss in accuracy from that obtainable with additional GPS updates. Consequently, GPS corrections may be stored and only used after an integrity-assurance delay time T, which is the maximum time required for the external IM system to assure the integrity of the GPS data. The GPS data are used first to determine an integrated GPS/INS solution at time t-T in the past. This solution, which includes optimally calibrated inertial error states, is then immediately propagated forward to present time, t. In effect, the navigation solution is always “coasting” from time t-T in the past to the present time t, because the propagated solution is based on a pure inertial solution without additional GPS updates. This type of integration is referred to as delayed, integrity-assured GPS/INS integration. With this approach, the integrity and redundancy requirements for sole means of navigation are achieved, while meeting both the accuracy requirements and goals for all phases of flight, including non-precision approach. It requires only the presently planned 21 or 24 satellite constellation, and a standard commercial 2 nmile/hr. INS, together with the existing GPS Operational Control system. Future systems, like GLONASS or the GPS integrity channel, are not required.

Autonomous navigation of small co-orbiting satellites using C/A GPS code

Abstract: A combined absolute/relative navigation experiment, currently under analysis in Italy, is described. Two space vehicles are constrained to take measurements from the same GPS satellites. The absolute and relative state vectors are determined by a Kalman filter that processes the differences between the two vehicle measurements relevant to the same GPS satellite and the absolute measurements of each vehicle. The experiment is implemented using two small coorbiting satellites, Pegasus/Scout II compatibles, flying in circular orbits at an altitude of about 400 km and manuevering to vary the distance between the two satellites. Absolute and relative position and velocity are autonomously computed onboard the vehicles with accuracies sufficiently high to allow GPS-based rendezvous operations and autonomous navigation.