Activity Feed › Discussion Forums › Education & Training › Instrument Accuracy Research Paper
Instrument Accuracy Research Paper
Posted by eagle1215 on July 2, 2022 at 6:03 pmI am currently taking a 400-level research class that is directed toward a research paper. I have attached my introduction. I am looking for advice on the feasibility of the topic. I am also looking for relevant research in regards to the accuracy of a robotic total station vs GPS vs an automatic level. I am still in a stage of the class where the topic can change. What are your thoughts?
dave-karoly replied 1 year, 8 months ago 8 Members · 15 Replies- 15 Replies
I think you have a topic that is very interesting. You will need to do a bit of research on a few items as it is written in regards to first.
accuracy vs precision
relative accuracy
absolute accuracy
use your upper math courses to help in identifying the things like this RMSE etc.
a point to consider is GPS over longer distances will be more accurate than a RTS or total station. Where one might be able to measure relative accuracy at closer distances etc. scanning has its own set of rules and precision and accuracy statements as well. But you are very well on the river track. Great topic. I look forward to reading your final product. You might want to visit NGS website since you are referencing a state plane system. Look in publications for the state information. Some great pre recorded webinars on datums and coordinate systems. Also on gps accuracy etc. look at total station and robot and gps specs from various manufacturers. On gps ck the difference between rtk and static. Look up the din or iso specs for the total stations and robot standards you get the accuracy specs on. This way you can use your math skills to validate or un validate my statement on gps being more accurate over longer distances. Also do some reading on absolute point positioning or PPP vs baselines etc. (hint absolute accuracy no base station needed) it is very apparent you are thinking for sure. Great start keep up the great work. Another resource that can help a bit is read the ALTA 2021 standards for relative position tolerances. This might fit well into your work as a baseline. Again math. Now there is one more accuracy you need which is not necessarily a measurement and that is directly related to boundary. We can be the most accurate and precise measurements in the world. But make in accurate assessments of boundary resolution. Fyi.
Some thoughts:
Your topic is too large for you to adequately cover. Find a portion of it that you can investigate in depth.
Yes, measurement is an important part of land surveying, but in many situations legal analysis is more important. Relatively few lawsuits are filed over the accuracy of measurement, versus over what was measured.
The quick answer is that each tool has situations where is is best. Using GNSS for a lot survey may not be as accurate as with a total station. GNSS in open country will be much faster and more accurate for breaking down a PLSS section.
There is no question that a level is superior for tight vertical control over projects of moderate size.
.@bill93 great point. It is a a large area to cover for sure and one could easily get down many rabbit holes before completion. I think as he begins to research his spider sense will kick in and he we figure out exactly the focus he needs to bring in narrowing it down some. Heck one could write a complete paper on any one instrument type with its accuracy and precision and pros and cons . Great to see though a young person thinking and going after knowledge. He will succeed. He just needs to keep wading through the mud to get to high ground. Always look to read your perspective. Always packed with wisdom.
FWIW:
Do you intend to make observations yourself or use other data sources?
Do you have access to all the instrument types you wish to compare?
Do you intend to make this a ??case study? of specific instruments of the types you define?
You refer to GRID as being an acronym. I do not know what the initial letters are abbreviations for. Grid coordinates, like the US SPC system, are values on a specific reference surface. In the case of the US SPC it is the surface of the reference ellipsoid.
I assume you are familiar with the reductions needed to place observed distances to the ground, horizontal or grid. You might appreciate this paper by Dr Charles Ghilani: https://www.xyht.com/professional-surveyor-archives/3088/
Obviously an upper-division course will require significant research. Google Scholar will provide lots of articles. The Research Gate and Academia.com sites are helpful.
Your paper will benefit from some word-smithing e.g. use ??plumb not ??plum.?
There is some interesting content on Metrology at the NIST site. See: https://www.nist.gov/metrology
Good luck,
DMM
I would encourage this effort. Bill is right. it needs to be narrowed down some, but what one learns about errors when focusing on measuring errors is valuable indeed.
Just to tighten it up a bit, variance and range of values are measures of precision instead of accuracy. The accuracy part will be difficult to control. Each experimental value has to be compared to a standard value, not to another experimental value. Standard values for these measurements are hard to come by and each comes with its own variances. Defining accuracy becomes tricky. Figuring out a reasonable, non-biased way is difficult, but worthwhile to pursue on its on merit.
Pare it down a bit, understand that accuracy can be a moving target, develop a methodology that copes with moving targets, and have at it.
I am not sure if I want to take the data myself for the project. I will investigate further on my own. I just want to get the essay portion completed. I can access the instruments when I choose to take field data. I would generalize for a GPS vs robot ts then add the specs for the particular instruments that I use for work. GRID is listed on our drawings and legals. We use it to show that we are using a Grid Coordinate system, not ground coordinates. We then list GRID – IGCS – XXX County. Indiana has State Plane East and West and Indiana Geospatial Coordinate System (IGCS). IGCS defines a plane for each county.
The professor finally stated the length of the essay to be 20-25 pages with 10 of which is the literature review.
FWIW, I offer the following comments on the ??proposal? in your original post. As it is entirely too hot to engage in outdoor activities, I have gone on too long.
I taught in a surveying program where we emphasized the importance of clear communication as well as technical knowledge. I hope you will read these comments as intending to help not merely criticize.
Your statement never explicitly states what you propose to do. You refer to the capabilities and limitations of different surveying instruments but never make a statement like ??I propose to analyze the relative performance of three types of surveying equipment: GPS, automatic levels and robotic total stations with respect to various surveying tasks.?
An analysis of the appropriateness of the three fundamentally different types of instruments to meeting standards and specifications requires that you provide a listing of the tasks. For example, you wish to compare comparisons of distance measurements, positioning accuracy with respect to known control points, height differences obtained using the three instrument types.
As an automatic level has neither distance measuring capability (other than stadia readings) nor the ability to measure angles it cannot be usefully compared with the other two instrument types.
All three instrument types can be used to determine height differences.
Using your terminology: while AL seems the most straightforward there are many factors that can reduce its accuracy including rod calibrations, inappropriately long sightings, varying sight lengths, setup or rod settling during observations, refraction, and blunders. Remember as well that errors in leveling accumulate.
RTS leveling (trigonometric leveling) can be quite accurate but is also subject to errors due to atmospheric effects (temperature and humidity, refraction, the same instrument and rod settling issues.
GPS (or GNSS) vectors can yield differences in height. In addition to possible setup errors, settling, and the impact of the wet troposphere, the ability of your software reduction package to correctly fix integers is crucial.
I am unclear whether you propose to analyze the differences in distances between points or differences in coordinates? If comparing coordinates will this involve a small project where data is adjusted and then compared?
On the separate matter of your definition (terminology):
You define GPS as: ???? a system of locating the receiver in relation to the Earth??s center.? First of all, the antenna is being located not the receiver which are sometimes separate units. Positioning your antenna using a base station involves the determination of the vector from the base station to the antenna or point over which it is established. There are, of course, various methods of using GPS signals including RTK, static and point-positioning. In the case of precise point positioning your antenna is positioned with respect to the GPS/GNSS satellites which, of course, are positioned wrt the geocenter.
I disagree with your assertion that GRID is an acronym. I understand that you wish to differentiate grid/ground distances/coordinates. Labeling distances reduced to the grid as IGCS grid (you should specify which grid you are using) and those field measurements reduced to the ground as ??ground? should make things clear.
General comments:
You state with respect to various instruments: ????has the tightest ?? control.? Don??t you mean that the instrument is the most appropriate when establishing ?? control?
You should consider using a term other that ??situation? when referring to task or condition.
Are accuracy standards ??higher? or more rigorous/stringent?
On the general issue of standards. One works to achieve a standard. Specifications are established to help meet that standard.
In closing, I wish you success on this and other projects.
DMM
@geeoddmike offers sound advice.
To clearly define your research goals, you need an h0 and an hA, null and alternate hypotheses, clearly stated with a predefined rejection criterion.
For example, if you’re using the 430-meter mark on an NGS Calibration Base Line National Geodetic Survey – Calibration Base Line (CBL) Web Map (noaa.gov) as a standard, then you might say: h0 = There is no significant difference between the measured distance and the published distance for (instrument) at the 68% level, and hA = There is a significant difference between the measured distance and the published difference at the 68% level.
You choose the significance level. 68% lowers the risk of a Type I error but raises the risk of a Type II error. You have to balance the relative importance of each in setting your acceptance interval. The important thing is to set it before you start. Using the data to set the interval is cheating and it leads to bad conclusions.
You can also set h0 and hA so that you’re comparing the differences between measurements taken with different instruments, but you need a good, published measurement value either way to verify your overall work. You’ll have the data to do it both ways with little extra work, so doing it both ways would be a good approach.
You’ve got this! Now produce good research for your professor and for us!
- Posted by: @eagle1215
The professor finally stated the length of the essay to be 20-25 pages with 10 of which is the literature review.
Boo! Yuk! Where is the dislike button?
God speed, young fella. You will do well. Getting sound advice, already.
You guys have been great help. My professor knows nothing about surveying. I am told not to get to technical in my essay. I think it is because he wants to make sure I am not just making stuff up. I have sources and citations for the information that I am using. I also think he is too lazy to sort through the resources that I have cited.
Be careful with the assumptions about his checking your sources. There is software that does that and checks for plagiarism and it’s fast and easy to use.
On the other hand, two weeks ago I reviewed a published research brief produced by a grant-funded university research organization using PhD candidates for researchers. The data were publically available. The misinterpretations and outright false claims in the brief were shocking. I spent maybe 90 minutes on the review and stopped about half way through. There was no point in continuing, so I submitted my findings and went to the beach.
I’m told by a respected educator that all such research is terrible and is excused because it often represents the author’s first=ever attempt at research. I think that is a reprehensible way to develop researchers To me, it would be far better to limit the scope of the assignment and vigorously police the standards. I suspect that your assessment of your professor is closer to the source of the quality problem that I saw.
Be different. Produce a quality product regardless of how you think it may be handled. It’ll have your name on it and there will be electronic copies, so it will never go away.
@mathteacher Where were you when I was in school. You and GeoidMike. Geezers. The statement above at professor not checking is what I witnessed first hand when I went back to school during the recession using my gi bill. We need more teachers like you for sure. Teachers that guide push develop those who want to learn.
I hope this young man follows your advice and goes in 100% . He can learn and strive to better himself for sure. No matter what is thrown his way.
There may be some nuggets in this research for you.
https://www.oregon.gov/odot/Programs/ResearchDocuments/SPR304-821_UpdatedSurveyStds.pdf
I wouldn??t use ??tight,? I would use more correct terminology.
Log in to reply.