Bruce R. Harvey

Using barometers to determine the height for indoor positioning

It is well known that atmospheric pressure decreases when altitude increases. Models have been created to relate altitude or height to pressure. A barometer can measure the air pressure and then the altitude can be calculated. Before the era of GNSS, barometers were widely used to determine heights outdoors. The invention of GNSS was a revolution in positioning and navigation. However, it does not work in an indoor environment. Alternative technologies have been developed such as Wi-Fi fingerprinting mainly for 2D positioning and navigation. In some of the applications, 3D or 2.5D (the level of the building) is required. Using barometers is a possible solution and some new mobile phones have a built in pressure sensor. But there are many issues that should be considered. Is height determined from barometric pressure accurate enough? Is there a latency problem? Does the air conditioning in an almost sealed building significantly affect height readings? This paper discusses the necessary considerations to use barometers for indoor applications based on experiments. Possible solutions are suggested.

Transformation of 3D Co-ordinates

Abstract As satellite positioning techniques are more frequently used, transformations of three-dimensional co-ordinates will become common practice. Methods of determining and applying transformation parameters for transforming 3D co-ordinates are discussed. The models, appropriate statistical tests and interpretation of the results will also be considered. The methods presented are applicable to many transformations required by surveyors in Australia. Some applications are the conversion of Doppler or GPS co-ordinates into the terrestrial networks and the comparison of Doppler, GPS or terrestrial data with SLR or VLBI data. In this paper particular emphasis will be placed on transformations between terrestrial and GPS data.

GPS COORDINATE TRANSFORMATION PARAMETERS FOR JAMAICA

Abstract The Surveying and Mapping community now has the benefit of real-time 3-dimensional coordinates at the centimetre level, through the Global Positioning System (GPS). The reference frame, World Geodetic System of 1984 (WGS84), within which a user ascertains these coordinates is essentially geocentric. In Jamaica all coordinate data and mapping are based on a non-geocentric coordinate system known as the Jamaican Datum of 1969 (JAD69), which like many others around the world was realized by making basic assumptions about the geoid-ellipsoid separation at the origin. WGS84 coordinates are therefore not compatible with the JAD69. This paper presents the results of work carried out to define the relationship between the two coordinate systems, through three types of transformation parameters, namely Block Shifts, Molodensky and Seven Parameter Similarity. Three of the forty-two primary control stations together with a Continuously Operating Reference Station (CORS), previously tied to the island network, forms a fiducial network at which WGS84 coordinates were determined. The two coordinate data sets for the four fiducial stations were used to solve the parameters. Tests were carried out on nine points with coordinates known in both systems. The results indicate that the Block Shift and Molodensky values transforms WGS84 coordinates to JAD69, accurate to ±1-2m and the seven parameter similarity values produce JAD69 coordinates accurate to ±0.5m or better. Arguments are put forward to suggest why coordinate transformation is an interim solution and that the development of a Geocentric Datum is preferred for Jamaica.

Supernova 1987A: radiosphere resolved with VLBI five days after the neutrino burst

Following the detection1 of radio emission from SN1987A in the Large Magellanic Cloud, we conducted very-long-baseline inter-ferometry (VLBI) observations with an interferometer composed of a NASA Deep Space Network antenna (DSS42) at Tidbinbilla, Australia and the antenna of the Hartebeesthoek Radio Astronomy Observatory, South Africa2,3. We did not detect any emission from the supernova above a level of ˜20% of the supernovas total flux density, although signals were detected from our two calibrator sources with amplitudes roughly equal to those determined in earlier VLBI observations. We infer that we resolved the supernovas radiosphere and estimate, for an epoch 5.2 days after the neutrino burst4,5, a lower bound on the radiospheres radius of 1.2 mas. Given the photometric data from the supernova6,7, a distance to the Large Magellanic Cloud of 50 ± 5 kpc (ref. 8), and an apparent expansion velocity that varied systematically with time from 18–16 x 103 km s–1 (refs 9 and 10), as estimated from the blue-shifted Hα absorption lines on the days preceding our observations, we conclude that 5.2 days after the neutrino burst the supernovas radiosphere was at least 2.5 times larger than the inferred blackbody photosphere, and at least as large as the Hα line-forming region.

Theodolite observations and least squares

Abstract This paper examines the reduction of theodolite direction observations in connection with subsequent least squares adjustments. These reductions of direction measurements are often called station adjustments. Some of the reduction procedures are reviewed. Whether to use grand means, single arcs or half arcs in a least squares adjustment of a survey network is considered with regard to systematic instrument errors, electronic data recording equipment, a priori standard deviations and computer resources. The effect on the statistical testing of survey networks is also discussed.

Accurate radio and optical positions for southern radio sources

Accurate radio positions with a precision of about 0.01 arcsec are reported for eight compact extragalactic radio sources south of -45-deg declination. The radio positions were determined using VLBI at 8.4 GHz on the 9589 km Tidbinbilla (Australia) to Hartebeesthoek (South Africa) baseline. The sources were selected from the Parkes Catalogue to be strong, flat-spectrum radio sources with bright optical QSO counterparts. Optical positions of the QSOs were also measured from the ESO B Sky Survey plates with respect to stars from the Perth 70 Catalogue, to an accuracy of about 0.19 arcsec rms. These radio and optical positions are as precise as any presently available in the far southern sky. A comparison of the radio and optical positions confirms the estimated optical position errors and shows that there is overall agreement at the 0.1-arcsec level between the radio and Perth 70 optical reference frames in the far south.

Geodetic Surveying with Quasar Radio Interferometry

Abstract From 20 April to 3 May, 1982, five Australian radio-telescopes were linked for the first time and operated in synchronism to form a single radio-telescope. The link-up was the culmination of two years of intensive effort and co-operation to co-ordinate telescope modifications and overcome the logistic problems of undertaking such an experiment. The five telescopes are sited at the NASA Deep Space tracking facility at Tidbinbilla near Canberra, the CSIRO Radio Astronomy Observatory at Parkes, the University of Tasmanias Radio Observatory near Hobart, the University of Sydneys Fleurs Observatory near Sydney, and the LANDSAT tracking station at Alice Springs. The simultaneous but independent operation of two or more widely separated telescopes is called Long-Baseline Interferometry. This experiment was set up to provide high-resolution radio maps of distant Southern Hemisphere quasars and galaxies. However, the experiment is of great interest to surveyors, since it also provides a means of making ...

Adjustment of survey networks by topological grid search

Abstract The adjustment of survey networks by a new topological grid search (TGS) approach offers some advantages compared to conventional least squares (LS) solutions. TGS can solve systems of highly non-linear, redundant or rank defect equations, and with integer or real number parameters. The observations of survey (geodetic) data can be combined with models of geophysical processes. In this paper three numerical examples are given to assist with the understanding of the TGS process and to compare it with conventional LS and L1 adjustments. A new investigation of the residuals of the observations is included and the effects of gross errors. The application of TGS to network rank defect problems and to integer parameters is presented in this paper for the first time. The TGS approach uses an m-dimensional grid of points around starting estimates of the values of the parameters. The method does not require a design matrix that is usually formed from partial derivatives of model equations, as is required in LS and robust L1 norm (L1) solutions. Topologically each observation is tested against the model values calculated for each of the many possible combinations of the grid points. Those grid points which satisfy the set of conditions that the calculated value differs from the observed value by less than some criteria are accepted. From this set of multidimensional grid points, which define a space including the solution, we compute the mean values, which correspond to an estimate of the solution parameters

SURVEYING THE DEFLECTION OF AN ARCH BRIDGE TO SUB-MILLIMETRE PRECISION

Abstract This paper describes a survey of an engineering structure that required high precision results. It is intended as a case study, primarily for surveying students, though the application of the techniques may provide insight to a wider audience. The purpose of the survey was to monitor movements of a new arch bridge to better than ±1mm. The coordinates of targets placed on the arch structure were not required, only deflections between subsequent epochs of measurements were required. Our survey procedures provided arch deflections in the vertical direction to better than ± 0.4mm and in the in-plane direction to ±0.8mm.

Content of the bachelor of engineering in geomatic engineering degree course at the University of New South Wales (UNSW)

Abstract This paper describes the content of the course at UNSW which is internationally recognised for its quality, depth and breadth. It prepares students for careers in the broad range of fields that are identified with Geomatics. [The School at UNSW has changed its name from Surveying to Geomatic Engineering and renamed its degree a Bachelor of Engineering.] Course material includes survey measurement using fully electronic tacheometers, in practicals, laboratories and survey camps; cadastral surveying, land development and management subjects all taught by practising professionals; computer programming and using commercial computer packages; geodesy, photogrammetry, spatial information systems (ie GIS), and remote sensing; as well as a sound basis in mathematics, physics, town planning and relevant aspects of civil engineering taught by other university schools. This paper attempts to explain some of the course content in narrative form, to provide prospective employers with details of the capabiliti...