Comprehensive GNSS Technology Training For Resource Mapping. Module 3 Quality Assurance and Quality Control Procedures

Transcription

1 Comprehensive GNSS Technology Training For Resource Mapping Module 3 Quality Assurance and Quality Control Procedures Revised: 2011

2 Contents 1) Quality Assurance Procedures... 4 a) Pre-Survey... 4 b) Field Data Collection... 4 c) Data Processing... 4 d) Mapping ) Mission Planning... 5 a) Terrain and Canopy Blockage... 5 b) Dilution of Precision (DOP)... 7 c) Satellite Outages and Periodic Maintenance Updates... 8 d) RT-DGPS Signal Availability... 8 e) Using Online (Web) Mission Planning Applications ) Ashtech Online Mission Planning Application... 9 a) Selecting Mission Location... 9 b) Selecting Time...11 c) Selecting Satellites...11 d) Viewing Mission Planning...12 i) Satellites and Count ii) Satellites and DOP iii) Polar Diagram iv) Smart 3D ) Data Management a) Field Data Models ) Quality Control Procedures a) Common Error Types...18 b) Detailed Audits ) Conclusion List of Figures Figure 1: Causes for Good and Poor PDOP... 6 Figure 2: Example PDOP chart (NavCom)... 7 Figure 3: Select Site page Figure 4: Select Time page Page 2

3 Figure 5: Select Satellites page Figure 6: Satellites and count page Figure 7: Satellites and DOP page Figure 8: Polar Diagram Figure 9: Smart 3D and Google Earth List of Tables Table 1: DOP ratings and their description... 7 Table 2: GNSS, GPS Status links... 8 Page 3

4 1) Quality Assurance Procedures Quality Assurance (QA) procedures refer to planned and systematic processes that provide confidence of a product s or service s effectiveness. This applies to all forms of activities; design, development, production, installation, servicing and documentation stages. These procedures must be implemented by any organization performing GNSS surveys to ensure that information leaving their office has the highest appropriate quality. All data must meet project requirements for accuracy, reliability, and completeness. Following are considered essential QA procedures that should be carefully planned and executed throughout a project: a) Pre-Survey This procedure ensures that all requirements of the project are understood between client and contractor and that all personnel understand the procedures to be followed (e.g. semi-permanent markers, map ties, etc.). b) Field Data Collection All GNSS and supplemental survey equipment must be in good working order. It is a good practice to double-check survey equipment before departing for a field mission. Field operators must be well trained in data capture methods and note keeping, as well as operating and maintaining any survey equipment. Project standards (e.g. DOP, SNR, elevation maximums, etc.) must be adhered initially. These values must be discussed with field crew in every detail. All necessary attributes must be captured in the field (not guessed at later). Supplementary digital photographs and sketches are always valuable to understand the field that data was collected. c) Data Processing All files (field and reference) must be downloaded and archived in proper formats with a system that can be readily accessed. Corrected data must be archived before any editing process so that the editing - smoothing - interpretation may be verified later. Page 4

5 d) Mapping Mapping operators must be well trained in recognizing and interpreting the errors in GNSS data, especially if data collection is in difficult conditions (e.g. under forest canopy or in urban areas). Specified cartographic standards must be followed throughout - much time can be spent doing excellent mapping to the wrong standards. All final products should be reviewed by field operators to verify interpretation of field notes and data noise and to check for completeness. This practice also provides more understanding to field operators about their data collection methods. Ensure that all required files, maps, reports, etc. are included in the proper format and are archived internally. 2) Mission Planning For data acquisition projects using GNSS technology, mission planning refers to predicting optimum satellite geometry for the survey location. It is essential that there are enough satellites available that would provide best geometry possible to meet project accuracy and productivity. Mission planning will help: Determine the best (most productive) observation times for a given survey. Optimize field data collection periods and crew planning given necessary constraints. The following must be considered for proper mission planning: Survey location terrain and canopy blockage, Dilution of Precision (DOP) values for better precision, Satellite outages and periodic maintenance downtime, RTK and real-time DGNSS Signal Availability. a) Terrain and Canopy Blockage With the addition of GLONASS satellites there is usually more than enough satellites available for obtaining good geometry. However, surveyors may encounter locations with heavy canopy and terrain blockage that would still require careful planning. Many GNSS hardware and software manufacturers supply utilities that would predict location specific conditions for the planning stage. However, using these utilities may not be relevant to real-world conditions and coherent planning is impossible due to the reality of terrain and canopy blockages. Page 5

6 Before working in an area with heavy vegetation canopies, it is useful to perform advance mission planning and assume a high (e.g degree) elevation mask. Although signals from low angle satellites may occasionally reach the antenna, planning for the worst case gives a useful estimate of the optimum observing periods. Figure 1: Causes for Good and Poor PDOP The Field Operator should consult to mission planning graphs. This location specific chart may be very useful when working in areas with difficult tracking conditions. Figure 2 below illustrates DOP results generated with NavCom Internet mission planning application (Web link provided below) for Victoria, BC area, February 8 th, Many other manufacturer specific applications and online tools generate similar graphs for referral. It is easy to recognize the high PDOP spike around 8pm local time by looking at the chart. Page 6

7 Figure 2: Example PDOP chart (NavCom) b) Dilution of Precision (DOP) Dilution of Precision represents the additional multiplicative effect of satellite geometry on GNSS precision. A low DOP value represents a better positional precision due to the wider angular separation between the satellites (Figure 1 above) used to calculate a receiver unit's position. DOP can be expressed as a number of separate measurements. HDOP, VDOP, PDOP, and TDOP are respectively Horizontal, Vertical, Positional (3D), and Time Dilution of Precision. GPS receivers allow the display of these positions (Skyplot Figure 8) as well as the DOP values. DOP Value Rating Description 1 Ideal Highest possible confidence level to be used for applications demanding the most critical possible precision at all times. 1-2 Excellent Positional measurements are considered accurate enough to meet all but the most critical applications. 2-5 Good Represents a level that marks the minimum appropriate for making business decisions Moderate Positional measurements could be used for calculations, but the fix quality could still be improved Fair Represents a low confidence level. Positional measurements should be discarded or used only to indicate a very rough estimate of the current location. >20 Poor Measurements are not reliable and should be discarded. Table 1: DOP ratings and their description Page 7

8 c) Satellite Outages and Periodic Maintenance Updates Orbiting satellites are occasionally taken out of service for scheduled (or unscheduled) maintenance. If a particular satellite is critical to the constellation geometry, it is important to ensure it will be operating during the planned field survey. A satellite outage may have little or no impact on field surveys, or it may have a significant impact. The only way to know is to run the prediction software and test the impact when the satellite is removed. The status of GNSS satellites is available from a private GNSS equipment manufacturer; Javad. The official civilian point of contact for GPS information, the US Coast Guard Navigation centre (NAVCEN). Below table provides the Internet links for GNSS and GPS status: GNSS Status GPS, DGPS Status Table 2: GNSS, GPS Status links d) RT-DGPS Signal Availability If real-time differential GNSS (RT-DGPS) is planned to be used, consideration must also be made for obstructions and coverage extent of this signal. Ground-based broadcasting systems such as the Coast Guard system or custom radio systems have a limited range. Check here for locations of Canadian Coast Guard beacons: Wide Area Augmentation System (WAAS) is available without any cost to everyone living in North America and beyond. WAAS uses a network of ground-based reference stations, in North America and Hawaii to generate GPS corrections, which are broadcast from 2 geostationary satellites. According to the FAA, WAAS specification requires it to provide a position accuracy of 7.6 metres (25 ft) or better (for both lateral and vertical measurements), at least 95% of the time. Actual performance measurements of the system at specific locations have shown it typically provides better than 1.0 metre (3 ft 3 in) laterally and 1.5 metres (4 ft 11 in) vertically throughout most of the contiguous USA and large parts of Canada and Alaska. Geo-stationary satellites appear fairly low on the horizon for users in British Columbia, and there must be a clear line of sight to the satellite in order to receive corrections, therefore WAAS is not a reliable correction source when working under canopy. Page 8

9 e) Using Online (Web) Mission Planning Applications There are a few proprietary mission planning applications available by different GNSS manufacturers and other entities. Some of these available on the Internet are listed below: Ashtech Web mission planning: (Google Earth integration) Navcom: System Effectiveness model (WSEM): 3) Ashtech Online Mission Planning Application In this section, due to recent popularity with online applications, Ashtech s Web mission planning application will be explored. This is not an endorsement or recommendation of any sort to use this application. Most online or stand alone mission planning applications are similar to each other, and produce very similar results. Knowledge gained here with Ashtech s application can easily be adapted to operate other similar applications. This particular application is selected since Google Maps is integrated and most people nowadays have used this clever mapping tool and have at least basic knowledge of using it. Mission planning steps are simple and described briefly below: 1. Specify time and location of the survey location, 2. Confirm the settings are appropriate (e.g. elevation mask, GPS or GPS + GLONASS), 3. Produce detailed planning graphs / lists (e.g. delete individual satellites to see NANU impact, change elevation mask, simulate obstructions, etc.). 4. Prepare mission plans with tolerance for DOP values and satellite availability. a) Selecting Mission Location Figure 3 below illustrates the simple user interface of Ashtech online mission planning tool and the initial Select Site page to begin mission planning. Users can enter their location at the address toolbar, either by typing their Latitude and Longitude coordinates, or simply by typing the name of the survey location. Also, there is a home icon displayed on the map, which predicts the user location roughly by resolving the user s Internet IP address and its server location. This setting needs to be enabled from the Web browser. Furthermore, it is simple to pinpoint the survey location with the marker provided on the left hand side of the map window. Just drag to the location, and map window will update itself automatically. Elevation Mask needs to be set to predict the visibility of the satellites. More specifically, the value set at Elevation Mask option causes any satellite seen below this angle from the selected site to be rejected from the mission planning. The mask should be set to the elevation mask that will be used for the GNSS campaign, typically As mentioned before, in areas of heavy Page 9

10 canopy a mask of will give a more realistic estimate of the number of satellites that will actually be tracked. Draw Local Mask on Panoramic View window on the bottom of the screen describes additional obstructions close to and around the selected site. The local mask includes the set elevation mask. Users can simply click on the diagram to add a new vertex to the existing mask. This action would take into account the azimuth and elevation of the point. The mask is then updated instantly. It is possible to add as many vertices as necessary. Every time a vertex is added, the mask is updated accordingly. To delete a vertex, just drag the corresponding round handle and drop it on the closest vertex. The two vertices will be merged into a single one and the mask will be updated accordingly. Figure 3: Select Site page Page 10

11 b) Selecting Time Select Time page is straight forward. Select the month from the first drop-down list. You can alternatively use the two blue arrow keys located nearby to select directly the previous or next month. Then, select the year from the second drop-down list. This would define the survey date and display it upper part of the Select Date area. Select Start and End Time, and verify the mission planning duration in the lower bottom of the page. Users also have the option to choose Coordinated Universal Time (UTC) or Local time. UTC time is based on Greenwich Mean Time (+8 hours for PST). Figure 4: Select Time page c) Selecting Satellites This page includes three tabs, one per GNSS system (GPS, GLONASS and SBAS). SBAS represents the global Satellite Based Augmentation Systems such as OmniSTAR, EGNOS and WAAS satellites. Users can choose the satellites they would like to use in their mission planning, and exclude the others. Page 11

12 For each of the three systems, 50% transparent color represents a user excluded satellite, where 80% transparency informs the user of an unhealthy satellite that cannot be included in the mission planning. Also, green represents healthy GPS satellites, where red and blue defines active GLONASS and SBAS satellites respectively. The bottom of the page lists all of the enabled satellites for all three systems. The displayed list of satellites complies with the color conventions mentioned in the above paragraph. Figure 5 below represents the Select Satellites page. Figure 5: Select Satellites page d) Viewing Mission Planning View Mission Planning page illustrates the results of the settings completed in previous pages. Users will notice a menu that gives access to the results of the mission planning. The same color conventions are used also on this page as with the Select Satellites page as far as the satellites are concerned; Green: GPS, Red: GLONASS, Blue: SBAS. Page 12

13 i) Satellites and Count This diagram illustrates (Figure 6) illustrates the following: - Availability of each satellite selected throughout the mission duration. Each horizontal segment represents the period of time of the selected satellite availability. - Number of all visible satellites vs. Time is represented by the jagged dotted line. - Bottom of the chart represents the mission duration; from start time to end time. - Vertical axes (left hand side of the chart) show satellite numbers; starting from GPS to GLONASS and to SBAS, represented by their assigned colors. Users will notice that mouse cursor operates as a kind of reticule when moved over the application. This is a very helpful tool to read the specific time results in more detail. Figure 6: Satellites and count page ii) Satellites and DOP Satellites and DOP diagram (Figure 7) is identical to the previous except that the number of available satellites is replaced by DOP values shown on the vertical axes on the right hand side. In this diagram, reticule enables users to view the specific time and DOP values on top of the Page 13

14 diagram. Each DOP type uses a different color such as PDOP as light blue, VDOP as black and etc. Figure 7: Satellites and DOP page iii) Polar Diagram Polar Diagram (Figure 8) represents the orbit of each satellite selected in previous pages that will be visible during the mission. Satellite orbits are represented with a plot based on the azimuth and elevation of the satellite applicable to the survey location. Users can place their mouse cursors on specific satellite information displayed under Visible Satellites screen and instantly observe its orbit on the Sky Plot. This applies to all satellite types as they are represented with different colors. Also, users can use the slider in the View Sat Position at Time window to view the satellite orbits in different times during their mission. Page 14

15 Figure 8: Polar Diagram iv) Smart 3D Smart 3D page (Figure 9) requires installation of Google Earth plug-in. This plug-in allows you to navigate and explore geographic data on a 3D globe using a web browser. It is free to download at below link: - This page enables users to visualize active satellites as they are represented by stars and lines connecting to the survey location. Rejected and invisible satellites can also be visualized by color spots, but they are not represented by lines connecting them to the earth. It is simple to identify the number (PRN) of a specific satellite by clicking on the color spot. It is also possible to animate the view by clicking on the play button or simply moving the slider as it is a built-in function of Google Earth. There are also settings present for users to adjust the animation, if there is any need. Page 15

16 Figure 9: Smart 3D and Google Earth 4) Data Management GNSS data management is an important part of a successful GNSS project. Without a consistent system of naming and archiving data files, essential project data may be lost. Organizations should develop a system of data management considering file names, folder structure and names, field and processing notes, project reports, and archive procedures. A data management and archiving scheme should be consistently and regularly followed within an organization. Most applications manage GNSS data in project folders. A typical GNSS job will create many files; some very large in size and some with cryptic names. Projects are a simple way to help keep these files organized. Page 16

17 a) Field Data Models A field data model is used during the field survey to control the collection of features (objects) and attributes (information about the objects) relevant to a particular survey project. A data model structures data collection to ensure that information collected in the field is complete, relevant, consistent and in a format compatible with the output requirements. It can also increase efficiency in the field by providing data entry forms and quick menu selection options. A data model allows captured feature positions and attributes to be immediately compatible with GIS and database structures. Some examples of useful data models are: Road inventories to capture: surface types, condition, road signs, culverts, bridges, etc. Water license inventories requiring a variety of features to be captured: point of diversion, point of interest, creek junctions, gates, chutes, causeways, dams, etc. as well as attribute information about each feature. Forest health inventories requiring feature and attribute information such as: pest type, stand type, number of attacked trees, slope, aspect, and operability. It is essential to carefully design a data model with consideration for the project specifications, field practicality, efficiency, and database compatibility. The model should be developed in conjunction with field and mapping/gis personnel to ensure it meets all needs. In some instances, a client or contracting organization may have a specific data model, which should be followed. It is a good idea to develop or adapt a data model in conjunction with the client if they do not already have one for the specific GNSS being used. Below are some of the considerations to create a data model: Keep the structure as simple as possible; unnecessarily complicated data models are inefficient during field surveys wasting time and leading to errors, Review GIS database requirements to ensure all relevant features and attributes are included and structured appropriately; Avoid using special characters in attribute fields (e.g. colons, commas, slashes, quotation marks, etc.). These characters may conflict with database export field delimiters. Underscores can be used to indicate spaces; Test the data model thoroughly before sending out with production field crews; Be prepared for multiple build/test/edit cycles before an efficient, useable data model is created. 5) Quality Control Procedures These procedures apply at the end of data acquisition stages. Quality Control (QC) is basically a detailed assessment of the quality of the GNSS data. This stage requires understanding of many concepts and need procedures that requires carefully executed. Page 17

18 If proper QA procedures applied in the beginning of the project, QC procedures will be easier to follow through. The frequency and level of detailed QC procedures will depend on resources available. Depending on the amount of data, the experience of the operator, and any automated tools available, the procedure can take from 15 minutes to 6 hours. Briefly, QC procedures include the following steps: o Visual assessment of the data o Search for noise in the data and other random errors o Assess field data collection errors o Look for systematic errors o Analyse the data with strict QC threshold levels. Some user applications can use statistical methods to analyse to accept and/or reject individual fixes that are obviously in an error range. If any point result is greater than the set threshold value for standard deviation they may be rejected as an outlier. a) Common Error Types Radial error is defined as the distance an individual position fix is from the true location of the point it has no direction. At least 95% of the points must be within the project accuracy specification of the true location of the point. Cross-track error is defined as the perpendicular distance of an individual position fix from the true location of a line. At least 95% of individual position fixes must be within the accuracy specification of the true location of the line. Along-track error is defined as an error in the same direction of the line, usually seen as a gap in the data. There will be a maximum distance between individual position fixes defined in the project specifications. It is recommended that most position fixes be much closer than this maximum distance. Interpretation errors can be a significant error in GNSS-derived line features with noisy data. This is often a problem with small, complex features under heavy forest canopy (e.g. reserve areas in a cut block). Another common problem is consistent, systematic misinterpretation of errors by a mapping operator. b) Detailed Audits A Quality Control audit of submitted data is a fairly rigorous process. In some cases, the submitted data would be independently re-processed and in others, the entire survey will be independently re-done. In either case, superior methods and procedures should be used. These procedures should be done by expert GNSS personnel or consultants with much experience in the technology, geodetic methods and statistical analysis. The auditor and the Page 18

19 methods should be accepted by all parties and any findings must be statistically and legally defensible. Usually, a very small portion of submitted GNSS data for a project or an organization would be subject to detailed audits. Field re-surveys are seldom done unless there is some question as to what was actually captured in the field. Times for detailed audits will of course vary depending on the scope of the original work and the methods followed. A reasonable rule of thumb would be to expect a detailed audit to take twice the time of the original, production survey. Allowances must be made for errors in both the initial survey and in the audit survey. Usually, the initial survey would have been done using the best methods and equipment for the conditions and project requirements. Although more rigorous methods may be used, the audit process can seldom be considered error-free. Any analysis must be based on reasonable statistical expectations. 6) Conclusion Quality Assurance goes a long way with GNSS data acquisition mission planning. As noted throughout this training module, this is more of a practical approach with guidelines. There are basic fundamentals as practised daily, and there are project and database specific requirements that need to be addressed very clearly well ahead of time. Many QA procedures are practiced with expertise of personnel, with no established documentation and standards. As expected, various practices are overlooked or even forgotten with the urgency of mission completion, leading to project deadline delays, additional expenses and mission failures. This module presented essential guidelines of QA practices for GNSS data acquisition mission planning as GeoBC recommends. Since most aspects of QA procedures need each other for conclusion, it is necessary not to disregard any steps. Due to the nature of GNSS field surveys, procedures that may appear to be insignificant or unnecessary could result in dramatic unsatisfactory results. It is always a good practice to remind the survey community with the importance of QA procedures to comply with Do it right, the first time. Page 19