- SPECIAL REPORTS
- THE MAGAZINE
Black box is a technical term for a device or system … viewed primarily in terms of its input and output characteristics. Almost anything might occasionally be referred to as a black box: a transistor, an algorithm, humans, the Internet. The opposite of a black box is a system where the inner components or logic are available for inspection ... sometimes known as a white box, a glass box or a clear box. Source: www.Wikipedia.org.
Surveying is an activity that involves the science of measurement; thus, understanding the theories of mathematics and physics that undergird the measurements is fundamental. Surveyors sometimes nod and talk about how difficult it is to understand the inside workings of a GPS receiver.
They correctly see GPS as a potential black box. But they often fail to see that almost everything they use today that involves electronics, microprocessors and software is a potential black box.
A black box can be a simple total station or a top-of-the-line robotic, reflectorless instrument. Surveying software can also be a black box, and that includes software carried into the field in a data collector or field controller, firmware in the instrumentation (GPS, theodolite, EDM, even digital level) as well as post-processing software.
I am fond of saying, “When you use a steel tape, you know when it’s broken. But how do you know when your EDM or RTK rover isn’t giving you accurate measurements?” This is one of the dangers of treating any step in the measurement process as a black box. Even when using a steel tape, surveyors have been known to unduly trust that the measurements they make with the tape are correct. Most know about the potential personal and environmental errors even if they don’t make the corrections. What they often forget is that the instrumentation itself can have errors. The most common errors are due to a lack of adjustment or calibration. However, software-driven tools used by surveyors in the field and the office can create a whole different series of errors and yes, even mistakes.
I define a mistake (or blunder) as a discrepancy in a measurement that is avoidable. It can be eliminated by making measurements carefully. Eliminating mistakes involves understanding the surveying system (the instrumentation, environment and the people directly and indirectly operating it) and applying carefully considered methods and procedures to collect the data.
With software-based technology, the first potential area for error is due to not having an understanding of how the software is constructed. Very often there are parameters that can be set so that corrections are made for some systematic natural errors as the data is collected. For example, in EDM measurements, this might mean compensating for how the index of refraction of the atmosphere changes with fluctuating temperature and pressure, or systematic instrumental errors such as compensating horizontal directions in a total station due to misleveling in the direction transverse to the line of sight of the telescope. Surveyors’ responses to these available corrections range from ignorance to indifference to misapplication to (fortunately) proper, informed use.
Allow me to illustrate the first three of these responses using the previous examples of atmospheric refraction and misleveling in the transverse direction to the telescope. An ignorant surveyor is unaware of the potential for error from various causes and is further unaware that the instrumentation can be set up to make corrections (given the correct inputs and proper measurement procedures). Ignorance can even arise from thinking that “one part per million is really tiny, so I don’t need to worry about it,” without ever actually evaluating what the error might be under the conditions of the measurement being planned.
An indifferent surveyor doesn’t really care whether the software makes those corrections or not. He or she may be aware that the capability to make corrections exists and may even take pains to make sure that the software settings for making those corrections are “on,” but doesn’t take the trouble to ensure that the proper temperature and pressure have been entered. Instead, he trusts that some correction, no matter how badly applied, surely is better than no correction at all. Examples of this indifference abound. One popular case is to use whatever EDM PPM setting was used previously.
In addition to the examples given above, hidden mistakes can frequently occur, and sometimes this is because the algorithms used by the software developer are incorrect or improperly applied. How many surveyors know whether the algorithms for making so-called iono-free corrections in their measurements are correct? How many surveyors know whether the antenna phase center corrections being applied are correct? How many surveyors know (or understand) the assumptions they are asked to make when using least squares software? This type of software evaluates and adjusts the random error in each measurement based on those assumptions.
Very often the only thing the surveyor can do is to measure “known” lines, figures or networks. Even in this instance, knowing how to set up such “known” checks takes a thorough understanding of the technology and the science of measurement. The demands on the surveyor’s knowledge and skills with “push-button” instrumentation (black boxes) is actually greater than with more tangible instrumentation such as the steel tape. He must imagine the impact of the errors and figure out how to compensate for them and/or determine if the automatic systems contained within do it correctly and appropriately. The surveyor does this based on a thorough understanding of how the technology works and how to correctly use the instrumentation. The surveyor who does this is the proper, informed one--and the one who sets an example for other surveyors.