Jerry L. Wahl
Branch of Cadastral Surveys
Bureau of Land Management
California State Office
2800 Cottage Way, E-2841
Sacramento, California 95825
Raymond J. Hintz
University of Maine
Corwyn J. Rodine
Bureau of Land Management
Eastern States Office
This paper is a discussion of the background for development of an Electronic Field Book whose requirements for data collection and other features are directly applicable to boundary retracement surveying. This application of data collection offers unique challenges that are not specifically addressed in commercial data collection systems. Among these attributes are high efficiency in traverse as opposed to radial survey, and onboard computation capabilities specific to the U.S. Public Land Survey System. An additional requirement is a thorough and easy to use mechanism for recording descriptive and evidentiary information as well as topographic and planimetric data relevant to the legal aspects of boundary surveying.
Since 1988 the Bureau of Land Management has been actively investigating the feasibility of survey data collection for it's Cadastral Survey function. At that time data collection was not generally considered worth implementing on surveys involving any significant amount of traversing. Various studies of data collection in the 1985 time frame as well as common sense indicated that station setup time overrode measurement time savings in all but pure radial shootout applications.
Neglecting the lack of field advantage, it was still recognized that some efficiencies were achievable in post processing with integrated software. In actual practice these advantages were hard to realize because of the lack of an effective common post processing system at the time within Cadastral Survey BLM, and that data collection capable surveying instruments were not generally available or in use.
Other factors which had led to failure of earlier attempts at using data collection were an apparent large implementation curve. That is, problems in setting up a system frequently prevented a user from becoming operational. Things as simple as obtaining a proper cable, or learning a different field procedure were common. Investigations indicated that there were a significant percentage of purchasers of data collection systems that ultimately did not use them.
Nevertheless over the years a few individuals within BLM Cadastral Survey had successfully implemented data collection systems via the HP-41C even though the primary advantage gained was automation of self written post processing. No commercial systems were successfully implemented.
Changing Times:
By 1988 things were changing. First, PC's began going along with the field surveyor to remote projects. The first use of the PC was to assist in preparation of the official record field note returns required in BLM Cadastral work. With the PC there was at last the potential for a place for the data to go. This was enhanced by the emerging automation of other portions of the field to finish cycle. Plat drafting was becoming established and in some areas and AutoCad was even being used for survey computations in some project situations. Also at this time the slow conversion and upgrading of field survey equipment had allowed data collection capable EDM and electronic total stations to start being available `on-line'.
Out of this scenario there arose an increased desire by a number of field people to try data collection. As a result Cadastral Survey began investigating data collection in order to identify what capabilities would be desirable for BLM Cadastral Survey applications, as well as what things seemed to be hindrances to success.
The approach taken was to try a number of different existing systems based upon proposals received for individual projects. Hopefully this would allow for feedback and the individual surveyors making the proposals would follow through and implement the systems allowing for a thorough evaluation.
As a result of these test projects and other field input as an ongoing process, much was learned about desirable attributes of a Cadastral data collection system.
Test Projects - Prototyping
Through a number of these prototype tests Cadastral learned a number of lessons working with the following 5 classes of data collection systems.
a. 41C stand alone systems: These systems were typified by no interface to an instrument, but used real time keyed in measurement data. Advantages were that implementation was basic, inexpensive, available to almost everyone for the cost of a PC interface, flexible, and defined by the user.
Lessons: Had good flexibility, field computations, low size and weight, lacked capacity and data integrity.
b. 41C systems using the Topcon HA1 interface: Cadastral's Eastern States Office developed a system using this interface. This is one of the most successful functional implementations.
Lessons: Worked only with a few Topcon instruments, need for more storage and full geodetically correct field computations that were part of the system were not available on line.
c. Systems based on the Tripod Data Systems (CO-OP41) data collection sub-system. Had the advantage of considerably more storage, and built in communication with PC's as well as working with many instruments, and improved data integrity. Enhanced with in-house ROM's.
Lessons: Showed desirability of working with multiple instruments, limited descriptive fields due to 41C limitations a problem, flexibility and customisablity still desired and achieved through custom EPROM's.
d. CMT MCII and MCV handheld's: These units fall midway between the 41C systems and the true MS-DOS PC's. Light weight with moderate amounts of storage. The 41C language available for the unit allows flexibility and with the optional TSTATION software even made possible a custom data collection system in 41 code as well as in C.
Lessons: MC II no environmental integrity, MCV okay but double the cost, rapidly converging hand held PC's converging cost wise with much more flexibility. Is not a full MS-DOS machine on which it is easy to use any PC like programs.
e. Dedicated data collectors: Nikon, Lietz SDR24, Zeiss DAC500 were also tested with mixed results. Typically had problems with proprietary procedures and cost.
As a result of the testing and feedback it became apparent that the most likely course of action was to pursue custom development of a data collection system that came to be called the Cadastral Electronic Field Book (CEFB).
During the last few years Cadastral Survey has also been developing a custom PC based Cadastral Computational system known as Cadastral Measurement Management (CMM). This software is in beta test at this time, and was developed through a co-operative effort between the Bureau of Land Management (BLM) and the University of Maine (UM). CMM is expected to be the first iteration in an effective PC-based dependent resurvey computation system. The pursuit of field data collection system is closely coupled to that software development.
Two of the most difficult problems associated with CMM are the monotony and error prone task of manual entry of field measurements, and the inability to have CMM functionality in a field environment. The CEFB effort is now just beginning, however due to the testing and prototype efforts, it's major components have been identified and software prototypes for each component currently exist.
The development and testing of CEFB will follow the same successful strategies which have been discussed regarding similar factors related to CMM (Rodine, et al., 1991). A significant amount of interaction between BLM and UM will result in a field test of CEFB, and results of the field test will result in modification of CEFB prior to a beta release.
Why another data collection system? Many would argue that a plethora of data collection systems now exist, and to create another would surely be redundant. This could be true for many applications, but no system exists which truly fulfills the dependent resurvey needs of the U.S. Public Land Survey System (PLSS). If one is to totally replace the field tablet with an electronic equivalent in the Cadastral Survey application the following must be considered:
Many other highlights of CEFB will be detailed in this paper which do indeed exist in other data collector systems.
Hardware Platform.
The choice of a data collection hardware was driven by a number of factors. Primary among these are future potential of the system, flexibility, size, weight, cost and, of course, functionality. This could have resulted in a difficult choice, however the criteria developed in the testing, feedback and prototyping phase limited the prospects to only a few options. For example, the requirement to interface to a multitude of instruments is very limiting. The 41C based systems have proved to be the most flexible, but also the most limited in storage or CPU capability and had no future due to HP's discontinuance of the 41C in mid 1988.
The choice was settled on the emerging MS or PC-DOS compatible hand held PC. There are a number of vendors of this hardware base. With intense technological and market competition the price and weight of these systems is coming down while the speed and power are rapidly increasing. There is obviously a greatly increased capability for future flexibility with this platform at about the cost of current proprietary data collection systems.
The recent advances in environmentally rugged hand-held MS-DOS computers makes these devices the logical choice for data collection hardware. It can be assumed that due to the plethora of applications that exist for these units that prices will continue to drop while their computing powers expand.
The significant advantage of the MS-DOS hand-held computers as a data collector is the ability to program the unit in standard high level computer languages such as C. In theory, any software system developed for use with MS-DOS will operate successfully on these hand-held computers in the field environment. In the future this will become more and more of an advantage. The screen size and storage capacity of current units are the only factors that limit any PC type application from functionally operating on one of these units.
Lack of programming flexibility (a la 41C) is offset by free form data collection and customisable features of the software design.
CEFB is being developed generically for all standard MS-DOS hand-held computers based on what appears to be a fairly consistent storage capacity, screen size, and keyboard.
Total Station Interface: One of the critical requirements identified for Cadastral Survey data collection is the ability to interface with a multitude of existing survey instruments that may be in use. The CEFB achieves this capability through an incredible amount of cooperation with the Florida Department of Transportation (FDOT, 1990), who has been responsible for a highway-based electronic field book system. Due to this cooperation the technology has been developed that allows CEFB to interface to all commercially available total stations. CEFB will still also allow hand entry of measurements at any time.
The combination of generic PC based data collection software combined with communication to all total stations eliminates any realistic hardware restrictions regarding use of the MS-DOS hardware option for CEFB.
Design Philosophy in Data Collection: CEFB will have the following basic generic components with regards to data collection:
Design Philosophy in Data Analysis: As repetitions are being turned at a setup the user will have the ability to obtain mean and standard deviations of these redundant measurements. The user can reject data, though this data will not be eliminated from the data file but instead flagged as disregarded (analogous to drawing a line through information in a field book).
If the user has defined starting coordinates and direction, the user will be able to recall geodetic (or state plane coordinates) of any station. When existing traverse stations are closed upon the user will easily display conventional latitude, departure, and angular closures.
If record boundary lines are defined, the user will be able to recall distance from a starting point (distance down line), the distance to a defined line (offset to a line), and the distance and direction to any desired point. Upon input of an angle or bearing the user will also be provided with a distance to the defined line.
Astronomy: Astronomic measurements will be processed real-time, stored in the raw data file, and compared to an existing direction for a line. The user will be able to continue traversing with the result of the astronomic measurements, the existing direction, or some weighted average of the two values.
In short, this design can be described as "Know As You Go", which is essential to a dependent resurvey.
Feature Codes: One of the critical design factors for BLM Cadastral Survey application in a retracement survey data collector is the ability to efficiently collect descriptive information. This will be accommodated at several levels. First, the user will be able to enter descriptive information free form at any time. In addition for point descriptive data a user-definable look-up table of feature codes will be available in CEFB.
The feature coding will be more than a simple, singular definition if one desires. For example, a feature code for a bearing tree may prompt the user for species, diameter, markings, bearing, and distance. Any or all of the secondary definitions may receive a null entry. The list of feature codes will be readily expandable or customizable by the user.
Initially some thought was given to the potential for full macro assisted word processing capability to exist with the CEFB environment to generate near final field note ready descriptions. Further thought about that possibility led to the belief that field operations needed to be streamlined as much as possible in order to be used, and it was unlikely for field people to feel comfortable taking the time to fully annotate descriptions. Offsetting this is the importance of allowing a proper and unbiased description of the actual monument and evidence information. That is, not to force the user to fit the description into a predefined form if it is not appropriate. The design of a streamlined yet flexible macro-feature code system will be one of the most difficult parts of the system to implement. The feature code information will be processed into a data base when the survey measurements are being processed. This information will then be used to help automate production of the official field notes.
Field Notes: At any time during a survey a user will have the ability to enter descriptive information through the keyboard, just as one would write into a field book. This information will also be extracted and placed into data base form at processing time.
Record to the Field: The user will also have the ability to store record field notes for the survey in CEFB. This will have been entered through a word processor in the office and transferred to the hand-held PC. This digital information will aide the surveyor in his search for record corners, In computing search positions for missing corners, etc. This information may be encoded using the feature code macro language.
CMM field functionality: Since a PC is now in the field, the next question raised would be why the entire CMM software system cannot be carried into the field? At the present time storage is the primary problem since computer hard disks, at their present level of development, cannot survive the rigors of field surveying. Also, the physical size of these handheld units limits the size of the display screen. While this is not a problem technically, only a portion of the standard 25 x 80 character screen can be viewed at any given time.
One must then consider which components of CMM are not essential to field surveying operations. It was the mutual consensus of BLM and UM that the survey network analysis routines, centered around the least squares analysis, are not really required. The next items of peripheral importance are the more automated functions for reproportioning entire data sets based on available located corners.
This leaves the following computational items critical in the field (Rodine, et al., 1991):
Since all of these functions now exist in CMM and are operating geodetically correct and follow procedures described by the Manual of Survey Instructions (1973), it is simply a matter of altering the user interface for the field unit to make them functional.
Processing - A daily link between CMM and CEFB: A processor of raw field measurements into abstracted distances, angles, azimuths, and elevation differences has been developed by co-author Hintz for the Florida D.O.T. EFB (FDOT, 1990). When tied to a control file, the software can automatically identify the redundant component of the survey network, i.e., save all sideshot observations for later computation.
The redundant data, based on any network geometry, is used to generate initial coordinates for all redundant stations, and a least squares adjustment of the redundant data set is then performed. The sideshot information is then computed from the coordinates generated by the least squares solution.
This entire process is a batch operation, and in CEFB this will mean adding a day's work to the existing data and reprocessing. The user will then employ all of the Cadastral computation tools in CMM to carry out duties in a dependent resurvey.
Prior to the next day's field work, the new coordinates will be re-loaded to the field unit. This will enable one to take the best estimates of station coordinates to the field in an updated fashion.
While it is expected that CEFB development will be evolutionary and ongoing for a number of future years, there are several technologies on the horizon which are not part of the current design, but may be considered at a future date.
First is the technology of hand writing PC interface, vs. keyboard entry. This could be particularly of value in a data collection system and could even allow sketches to return to the electronic field book environment. Until systems are available to test, however, it is difficult to assess whether an interface technology will be a help or hindrance in a particular application.
Infra red link with the total station. One of the most failure prone, time consuming to maintain aspects of data collection is the cable interface to the survey instrument. It would be desirable in the future to replace this with an infrared digital link (with full data handshake checksum). This may depend greatly on total station manufacturers of development of extremely compact self contained serial port links that can be installed on any instrument.
The Cadastral Electronic Field Book (CEFB) has been discussed. CEFB is designed specifically to suit the needs of a surveyor performing a dependent retracement survey in the U.S. Public Land Survey System. The system will allow a user to always know his geodetic position, his relation to other points, his relation to record boundary lines, and perform geodetically correct PLSS computations efficiently in the field. The success of any data collection system is highly dependent upon user acceptance and functionality. Towards this end the process of development of CEFB will and must include an evolutionary process that includes user testing, feedback, responsive adaptation and iterations of that process.
Funding for this project was made possible by Cadastral Survey in the U.S.D.I., Bureau of Land Management. The authors wish to thank Bernard W. Hostrop, Chief Division of Cadastral Survey Washington Office, all the BLM Cadastral surveyors who have provided feedback on data collection testing, and an unlistable large number of people at the Florida Department of Transportation who have offered help in this project.
Blanchard, B.M. (1990), Utility Program Development: A Digitally Integrated Measurement Management System for the U.S. Public Land Survey System, M.S. thesis, University of Maine, Orono, ME, 86 p.
Florida Department of Transportation (1990), Electronic Field Book System - Field Surveying System, Tallahassee, Fl. 127 p.
Florida Department of Transportation (1990), Electronic Field Book System - Survey Data Processing System, Tallahassee, Fl. 76 p.
Hintz, R.J., Blackham, W.J., Dana, B.M., and J.M. Kang (1988), Least Squares Analysis in Temporal Coordinate and Measurement Management, Surveying and Mapping, Vol. 48, No. 3, pp. 173-183.
Hintz, R.J. and C.J. Rodine (1990), Automation and Precision in a Cadastral Surveying Environment, Proceedings of the ACSM-ASPRS Annual Convention, pp. 124-133.
Hintz, R.J. and H.J. Onsrud (1990), Upgrading Real Property Information in a GIS, URISA, Vol. 2, No. 1, pp. 2-10.
U.S. Dept. of Interior - Bureau of Land Management (1973), Manual of Instructions for the Survey of the Public Lands of the United States 1973, U.S. Government Printing Office, 333 p.
Wahl, J.L. (1990), Cadastral Measurement Management User's Manual (Beta Release), unpublished, 120 p.
Wahl, J.L., Rodine, C.J., and R.J. Hintz (1991) Progress Report on the Development of an Integrated PLSS Cadastral Measurement Management and Retracement Survey Software System