1. Read in the Lost Person Behavior book that “a user friend ground-based computer model that automatically generates a POA map does not currently exits. (page 306) However, there appears to be other programs in development which calculates probability distributions based on different data. Example would be Incident Commander Pro which uses a “comprehensive library of lost person data”. It appears we are not alone on the problem. It also appears they are getting their data from SAR statistics, which is very valuable. Predicting human behavior is really difficult. Our research in just trying to figure out how people move as a function of terrain may produce some interesting results.

2. While scanning through some of the geocaching tracklogs, it seems that the majority of geocachers like to drive to their caches and either geocache in a city or in parks with set trails.

3. Geocaching.com has some guidelines to geocaches. There are obvious no trespassing, protected land that restricts caches. So off trail caches are allowed. (This would be a real damper if we were looking for off trail data from an activity that has it against their guidelines.)

4. I still haven’t been able to find a database/GIS of trails in either the US or the world.

5. Had some technical difficulties in setting up the facebook app. But it’s ready to go and can easily incorporate new features.

6. Advertising on facebook: First you give a bid of how much you are willing to pay per click (They recommend $.52-$.69) and you set a daily budget of how much you are willing to pay per day. 7. Skimmed Tracking: A Blueprint for Learning. This book seems pretty interesting. It talks a lot about tracking in general with different purposes. This didn’t help too much in trying to figure out how people move, but a good reminder that there are also some easily disrupted signs on the ground that can give more clues to where a lost person is. Signs that would be too unrealistic for a UAV to pick up. (The author also gives the impression that trained trackers are the best at finding a lost person. 8. So there’s activity similar to geocaching called geohashing. The idea is that everyday a coordinate is hashed (based off the date and the closing value of the DOW) to produce a random coordinate in each graticule (we are in the region from Spanish Fork to about Kayville). There are about 15 active participants on the geohash wiki for the salt lake graticule. Maybe if we ask them if they’d like to participate in our research by submitting their track logs. (And wouldn’t a facebook app be real easy to remember where to submit?)

9. Finished reading the paper Judgment under Uncertainty: Heuristics and Biases by Amos Tversky and Daniel Kahneman. The main idea I took away from the paper is that we come across decision making where we have limited information and must choose something which we can't tell if it's correct or not. It goes through 3 mechanisms which we use to guess. Right now in WiSAR we strategically guess where a missing person is and we hope to be able to create computer generated models which will give us more information to better guess where someone is.

10. Checked out developing for the Iphone/IPod Touch. Apparently you need a mac to develop with. They have a simulator as part of the SDK. Development primarily uses objective C. Some cool features is that you have access to the socket programming and the units can get a location (using GPS, cell phone towers, or wifi signals)

1. Began working on visualizing the data from tracklogs. It appears that in the representation there may be some distortions. (Just like map makers. But for a small region it’s unnoticeable.)

2. I couldn’t find/figure how to convert between meters and degrees on the earth. I created an algorithm that approximates how many degrees it takes to move a specific distance.

3. Talk Mike about the iphone app. He says that once he gets it setup there will be areas others can build it up. I’ve decided I’ll wait for him to get started on it.

4. Mike showed me how we can download the GIS data from http://gis.utah.gov/sgid-vector-download/utah-sgid-vector-gis-data-layers-by-category. I noticed one of the categories from there was wilderness. This kind of information may be a little more useful. (Question: What is defined as wilderness? Specifically does it contain trails or not.)

1. Worked on the visualization more mostly.

2. Couldn’t put in all of my hours.

1. I was thinking about long range wifi, and here is a device that provide wireless networking up to 50 miles. http://www.afar.net/wireless/ethernet-bridge/. However this is relatively slow (2.75 mbps compared to the normal 54 mbps.) I’m thinking it won’t be fast enough for the video.

2. Should just tag everyone with satellite beacons in case they get lost…..Guess cell phones are the closest we’ll probably get to that point. The limitations are obvious.

3. It seems as I go through off trail logs (ones with bushwhacking) It’s interesting to note it seems that even on trips where they go out and come back, they don’t follow exactly the same route. Especially with changes in elevation/vegetation.

4. Sarbayes.org seems like a good resource once we really start modeling.

5. Only put in about 10 hours this week

1.Low slope is about 1% 2. Medium slope is about 5% 3. A steep slope is about 30%

1. I have done a lot of labor intensive work (coding, looking for off trail data, etc).

2. Wilderness.com has a resource for locations of national wilderness areas. However these appear to have trails

3. Found a good reference of how people tend to walk in circles without the aid of navigational devices (ie sun.) (Walking Straight into circles by jan souman)

4. The wonder client and phairwell are written in c++ using the QT toolkit. This is help a lot in integrating the programs.

1. http://www.mrlc.gov/ provides land cover data. This data is presented in erdas .img format. Gdal (http://www.gdal.org/) provides a conversion tool to convert from the complicated and propreitary format to geotiff. It also provides another tool to provide the meta data of the geotiff (the projection used, the origin in that projection, parameters for the projection, etc.)

2. These geotiffs contain the land cover data we need. Each pixel represents 30×30 meters. To extract the pixels would just require opening the image files and selecting the pixel. However, since each image file is large(700-1400 megabytes, I had problems opening them with standard image libraries(QT's and java's.) Sun provides an advance imaging library (https://jai.dev.java.net/) that is able to handle the large files.

3. In order to extract the land cover of a given geographic coordinate, the Cartesian coordinate which the projection represents that be calculated. The geotiffs use the Alber's Equal Area projection. I couldn't seem to get the right results with the equation provided by http://mathworld.wolfram.com/AlbersEqual-AreaConicProjection.html (Using Quantum GIS to verify since it provides the projected coordinates for a given point.) I found a java library (http://www.jhlabs.com/java/maps/proj/) that is able to calculate map projections, but it gives points approximately 200 meters off. I found a more reliable c library called proj4(http://trac.osgeo.org/proj/) that calculates the projection.

4. I am now working on a java app that is able to take a geographic coordinate, and give the land cover value at the point using the land cover data and proj4.

1. Went to the demo field trial. I have a better understanding of all the parts that go into making our UAV fly.

2. Would someone navigating the outdoors with a GPS have more situational awareness if they had audible cues? (In other words diminishing the need of the visual interface.) Just an idea since when people (or maybe WiSAR searchers) are navigating to a given point, they may spend a lot of visual time looking at their GPS instead of what's around them. Would this have an effect the one's ability to search?

1. Took some time to understand how java threads and sockets work. This is a new field to me since I don't really have any experience with them. This will be used in an opening a socket with the code I wrote to find the land cover at a given point.

Here is the format that I will use for the webserver. (8 bit chars)

To request an answer:

<WISAR>
<Request> <latitude in decimal degrees as a string> , <longitude in decimal degrees as a string> </Request>
</WISAR>

The response

<WISAR>
<Response><Answer as an int string></Request>
</WISAR>

Finished writing the java land cover server.

Started writing the network classes for warnings in Phairwell. However, Mike has plains for rewriting his network classes. I guess I will work on making the warning class to send out a QT signal when there is a warning.

Refactored my track visualization code so that it's easier to add meta data to a tracklog.

Now have it where I am having my track builder read in the values from the land cover server. Next step will be to make a mapping of those values to spare, medium, or dense vegetation.

Here is a meaning of the responses from the land cover server

    11. Open Water - All areas of open water, generally with less than 25% cover of vegetation or soil.

    12. Perennial Ice/Snow - All areas characterized by a perennial cover of ice and/or snow, 
generally greater than 25% of total cover.

    21. Developed, Open Space - Includes areas with a mixture of some constructed materials, 
but mostly vegetation in the form of lawn grasses. Impervious surfaces account for less than 
20 percent of total cover. These areas most commonly include large-lot single-family housing 
units, parks, golf courses, and vegetation planted in developed settings for recreation, 
erosion control, or aesthetic purposes

    22. Developed, Low Intensity - Includes areas with a mixture of constructed materials and 
vegetation. Impervious surfaces account for 20-49 percent of total cover. These areas most 
commonly include single-family housing units.

    23. Developed, Medium Intensity - Includes areas with a mixture of constructed materials 
and vegetation. Impervious surfaces account for 50-79 percent of the total cover. These areas 
most commonly include single-family housing units.

    24. Developed, High Intensity - Includes highly developed areas where people reside or work 
in high numbers. Examples include apartment complexes, row houses and commercial/industrial. 
Impervious surfaces account for 80 to100 percent of the total cover.

    31. Barren Land (Rock/Sand/Clay) - Barren areas of bedrock, desert pavement, scarps, talus, 
slides, volcanic material, glacial debris, sand dunes, strip mines, gravel pits and other accumulations 
of earthen material. Generally, vegetation accounts for less than 15% of total cover.

    32. Unconsolidated Shore* - Unconsolidated material such as silt, sand, or gravel that is subject 
to inundation and redistribution due to the action of water. Characterized by substrates lacking 
vegetation except for pioneering plants that become established during brief periods when growing 
conditions are favorable. Erosion and deposition by waves and currents produce a number of landforms representing this class.

    41. Deciduous Forest - Areas dominated by trees generally greater than 5 meters tall, and 
greater than 20% of total vegetation cover. More than 75 percent of the tree species shed foliage 
simultaneously in response to seasonal change.

    42. Evergreen Forest - Areas dominated by trees generally greater than 5 meters tall, and 
greater than 20% of total vegetation cover. More than 75 percent of the tree species maintain their 
leaves all year. Canopy is never without green foliage.

    43. Mixed Forest - Areas dominated by trees generally greater than 5 meters tall, and greater 
than 20% of total vegetation cover. Neither deciduous nor evergreen species are greater than 
75 percent of total tree cover.

    51. Dwarf Scrub - Alaska only areas dominated by shrubs less than 20 centimeters tall with 
shrub canopy typically greater than 20% of total vegetation. This type is often co-associated 
with grasses, sedges, herbs, and non-vascular vegetation.

    52. Shrub/Scrub - Areas dominated by shrubs; less than 5 meters tall with shrub canopy 
typically greater than 20% of total vegetation. This class includes true shrubs, young trees in 
an early successional stage or trees stunted from environmental conditions.

    71. Grassland/Herbaceous - Areas dominated by grammanoid or herbaceous vegetation, generally 
greater than 80% of total vegetation. These areas are not subject to intensive management such 
as tilling, but can be utilized for grazing.

    72. Sedge/Herbaceous - Alaska only areas dominated by sedges and forbs, generally greater 
than 80% of total vegetation. This type can occur with significant other grasses or other 
grass like plants, and includes sedge tundra, and sedge tussock tundra.

    73. Lichens - Alaska only areas dominated by fruticose or foliose lichens generally greater 
than 80% of total vegetation.

    74. Moss - Alaska only areas dominated by mosses, generally greater than 80% of total vegetation.

    81. Pasture/Hay - Areas of grasses, legumes, or grass-legume mixtures planted for 
livestock grazing or the production of seed or hay crops, typically on a perennial cycle. 
Pasture/hay vegetation accounts for greater than 20 percent of total vegetation.

    82. Cultivated Crops - Areas used for the production of annual crops, such as corn, 
soybeans, vegetables, tobacco, and cotton, and also perennial woody crops such as orchards 
and vineyards. Crop vegetation accounts for greater than 20 percent of total vegetation. 
This class also includes all land being actively tilled.

    90. Woody Wetlands - Areas where forest or shrubland vegetation accounts for greater 
than 20 percent of vegetative cover and the soil or substrate is periodically saturated 
with or covered with water.

    91. Palustrine Forested Wetland* -Includes all tidal and non-tidal wetlands dominated 
by woody vegetation greater than or equal to 5 meters in height and all such wetlands that 
occur in tidal areas in which salinity due to ocean-derived salts is below 0.5 percent. 
Total vegetation coverage is greater than 20 percent.

    92. Palustrine Scrub/Shrub Wetland* - Includes all tidal and non-tidal wetlands 
dominated by woody vegetation less than 5 meters in height, and all such wetlands 
that occur in tidal areas in which salinity due to ocean-derived salts is below 0.5 percent. 
Total vegetation coverage is greater than 20 percent. The species present could be true 
shrubs, young trees and shrubs or trees that are small or stunted due to environmental conditions.

    93. Estuarine Forested Wetland* - Includes all tidal wetlands dominated by woody 
vegetation greater than or equal to 5 meters in height, and all such wetlands that 
occur in tidal areas in which salinity due to ocean-derived salts is equal to or greater 
than 0.5 percent. Total vegetation coverage is greater than 20 percent.

    94. Estuarine Scrub/Shrub Wetland* - Includes all tidal wetlands dominated by woody 
vegetation less than 5 meters in height, and all such wetlands that occur in tidal areas 
in which salinity due to ocean-derived salts is equal to or greater than 0.5 percent. 
Total vegetation coverage is greater than 20 percent.

    95. Emergent Herbaceous Wetlands - Areas where perennial herbaceous vegetation accounts 
for greater than 80 percent of vegetative cover and the soil or substrate is periodically 
saturated with or covered with water.

    96. Palustrine Emergent Wetland (Persistent)* - Includes all tidal and non-tidal 
wetlands dominated by persistent emergent vascular plants, emergent mosses or lichens, 
and all such wetlands that occur in tidal areas in which salinity due to ocean-derived 
salts is below 0.5 percent. Plants generally remain standing until the next growing season.

    97. Estuarine Emergent Wetland* - Includes all tidal wetlands dominated by erect, 
rooted, herbaceous hydrophytes (excluding mosses and lichens) and all such wetlands 
that occur in tidal areas in which salinity due to ocean-derived salts is equal to or 
greater than 0.5 percent and that are present for most of the growing season in most 
years. Perennial plants usually dominate these wetlands.

    98. Palustrine Aquatic Bed* - The Palustrine Aquatic Bed class includes tidal and 
nontidal wetlands and deepwater habitats in which salinity due to ocean-derived salts 
is below 0.5 percent and which are dominated by plants that grow and form a continuous 
cover principally on or at the surface of the water. These include algal mats, detached 
floating mats, and rooted vascular plant assemblages.

    99. Estuarine Aquatic Bed* - Includes tidal wetlands and deepwater habitats in 
which salinity due to ocean-derived salts is equal to or greater than 0.5 percent and 
which are dominated by plants that grow and form a continuous cover principally on or 
at the surface of the water. These include algal mats, kelp beds, and rooted vascular 
plant assemblages.

    * Coastal NLCD class only

Created a mapping from the return value from the server to sparse vegetation, medium vegetation, dense vegetation, water, swamp, other, developed.

Sparse = {31,51,81,71}
Medium = {52,82, 90}
Dense = {41,42,43}
Water = {11}
Swamp = {32, 91, 92, 93, 94, 95, 96, 97}
Other = {98, 99, 12, 72}
Developed = {22, 23, 24}

Just cleaned up my desktop and made sure all of my code was in a repository.

I've been thinking about over/under confidence in autonomous systems, particularly that of the UAV.

Is it better for the UAV/video operator give false alarms versus being missed pron What happens with the searchers as they are directed around because “the UAV saw something”. How would an incident commander direct their resources based off the feedback from the UAV verses their own experience? How is the UAV best used? I think if there is too much confidence in the narrow point of view from the UAV, then the chance to build/use experience of others is lost. If there is too little, then the UAV would never be used. How would an incident commander actually use a computer generated probability distribution?

I've been working again on track visualization again. I'd like to improve the interface. Also add editing features so that we can easily add meta-data to a track log manually.

After refactoring my old track visualization, I almost at the point to be able to select points to add meta-data to.

Sometimes a little extra reading pays off. I was wondering why a track log of one of my geocaching adventures through rock canyon was having it's land cover as “Developed, Open Space” and “Developed, low Intensity”, and mostly for the points on/near the trail. For points where I was clearly off trail it properly categorizes them. Part of the definition is “These areas most commonly include large-lot single-family housing units, parks, golf courses, and vegetation planted in developed settings for recreation, erosion control, or aesthetic purposes.” I am thinking that we could use this as a heuristic for determining on/off trail. This would be better then trying to eyeball through google earth.

Ignore the previous statement :(

Met with Dr. Goodrich. Gave suggestions for the track editor to overlay an elevation model and picture with the tracklog. This will provide a better visualization of the tracklog. This would make annotating easier.

As for trust research, I need to figure out how it is defined and how it can be measured.

Current I have a 3d track visualization. It is able to dynamically download large amounts of elevation data and imagery within about a minute from USGS. Also, after having written the software in C++ (and using a Java module to interact with USGS's web interface) I decided to port the program over to Java since there excellent language tools, managed memory, and development speed. The port was successful.