Spatial Computing∗:

Accomplishments, Opportunities, and Research Needs

Shashi Shekhar, Steven Feiner, Walid G. Aref

Abstract

We present a perspective on the societal accomplishments, recent cultural shifts, challenges,and opportunities in spatial computing based on the discussions at the 2012 Computing Com-munity Consortium (CCC) visioning workshop, “From GPS and Virtual Globes to Spatial Com-puting – 2020,” held at the National Academies’ Keck Center to assess interdisciplinary devel-opments and research challenges in spatial computing. In this article, we first provide detailedexamples of transformative accomplishments resulting from spatial computing research. We thendiscuss the cultural shift resulting from the deep integration of spatial computing technologiesinto the everyday lives of citizens. This integration has presented a number of short-term op-portunities for new developments in spatial computing along with long-term research needs tofurther transform our society.

1 Introduction

We present a perspective on the societal accomplishments, recent cultural shifts, challenges, andopportunities in spatial computing based on the discussions at the 2012 Computing CommunityConsortium (CCC) visioning workshop [1], “From GPS and Virtual Globes to Spatial Computing– 2020,” held at the National Academies’ Keck Center to assess interdisciplinary developmentsand research challenges in spatial computing. Earlier, the US Department of Labor identifiedgeospatial technologies, along with nanotechnology and biotechnology, the three most importanthigh-growth industries for the 21st century [9]. Subsequently, the Association for ComputingMachinery (ACM) in 2008 created a new Special Interest Group (SIG) for Spatial Computing(SIGSPATIAL).

Spatial computing encompasses the ideas, solutions, tools, technologies, and systems thattransform our lives and society by creating a new understanding of spaces, their locations, places,as well as properties; how we know, communicate, and visualize our relation to places in a spaceof interest; and how we navigate through those places. From virtual globes to consumer globalnavigation satellite system devices, spatial computing is transforming society. We have reachedthe point where a hiker in Yellowstone, a schoolgirl in DC, a biker in Minneapolis, and a taxidriver in Manhattan know precisely where they are, know the locations and details of nearbypoints of interest, and know how to efficiently reach their destinations. Large organizationsalready use spatial computing for site selection, asset tracking, facility management, navigationand logistics. Scientists use the Global Positioning System (GPS) to track endangered speciesand better understand animal behavior, while farmers use GPS for precision agriculture toincrease crop yields and reduce costs. Virtual globes [2] such as Google Earth and NASAWorld Wind are being used to teach school children about their local neighborhoods and the

∗Spatial computing is used in a broad sense to include the study of computing in spatial, temporal, spatio-temporal,and non-geographic spaces.

1

Table 1: Representative Spatial Computing Organizations

ACM SIGSPATIALAmerican Society of Photogrammetry and Remote Sensing (ASPRS)Association of American Geographers (AAG)IEEE Geoscience and Remote Sensing Society (GRSS)Institute of NavigationNational Academy of Sciences

Board of Earth Science and Resources (e.g., Mapping Sciences Committee)Society of Photo-optics Instrumentation Engineers (SPIE)University Consortium for Geographic Information Science (UCGIS)

world beyond in an enjoyable and interactive way. In the wake of recent natural disasters (e.g.,Hurricane Sandy), Google Earth’s service has allowed millions of people to access imagery tohelp in disaster response and recovery services [17]. Within days of the 2010 Haiti earthquake,navigable digital roadmaps were available thanks to citizen volunteers submitting to the popularvolunteered geographic information [10] website OpenStreetMaps [20].

These tools are just the tip of the iceberg. In the coming decade, spatial computing promisesan array of new transformative capabilities. For example, where GPS route finding today isbased on shortest travel time or travel distance, companies are now experimenting with eco-routing, finding routes that reduce fuel consumption and greenhouse gas emissions. Smartrouting that avoids left turns has already saved UPS over three million gallons of fuel annu-ally [30]. Such savings can be multiplied many times over when eco-routing services becomeavailable for consumers and other fleet owners (e.g., public transportation). The ubiquity ofmobile phones presents an incredible opportunity for gathering information about all aspects ofour world and the people living in it [11]. Already research has shown the potential for mobilephones with built-in motion detectors carried by everyday users to detect earthquakes mereseconds after they begin [8]. Navigation companies frequently utilize mobile phone records toestimate traffic levels on busy highways [32]. How can computers efficiently utilize this prevalentspatio-temporal sensing power of mobile phones without drastically impacting battery life orpersonal privacy concerns? There is a growing need for a cyber-infrastructure [3] to facilitate ourunderstanding of the Earth as a complex system. Technological advances have greatly facilitatedthe collection of data (from the field or laboratory) and the simulation of Earth systems. Thishas resulted in exponential growth of geosciences data and the dramatic increase in our abilityto accommodate complexity in models of Earth systems. It may be crucial for understandingthe entire Earth and its physics (e.g., ocean, atmosphere and land), biology (e.g., plants animals,ecology), sociology (e.g., sustainable economic development, human geography), etc.

2 Transformative Accomplishments

Spatial computing initially arose to support computational representation and analysis of mapsand other geographic data across many disciplines represented by professional organizationslisted in Table 1. Since then, a number of transformative technologies have deeply inte-grated themselves in our society. Representative accomplishments include the global position-ing system, location-based services, geographic information systems, spatial statistics, spatialdatabases, spatial data models, and the digital earth.

Global Positioning System (GPS): Where am I? In the eighteenth century, “the longi-tude problem” was among the hardest scientific problems. Lacking the ability to measure theirlongitude, sailors throughout the great ages of exploration had been literally lost at sea as soonas land was out of sight. Through a combination of a compass, maps, star positions, and the

2

invention of the chronometer (a clock that worked on moving ships), it was eventually possibleto position oneself with some level of precision even in the middle of the ocean with few land-marks. With the launching of the Global Positioning System, it is now possible to quickly andprecisely locate oneself anywhere on the surface of the Earth. The Global Positioning System(GPS) is a space-based satellite navigation system that provides location and time informationanywhere on Earth where there is an unobstructed line of sight to four or more GPS satellites(out of a few dozen) [23]. GPS’s accurate timekeeping facilitates everyday activities such asclock synchronization in computer networks (e.g., Internet), geographic distributed sensor gridsto monitor moving objects (e.g., missiles, planes, vehicles), electric power distribution grids, etc.GPS’s localization capabilities have made possible a number of location-based services (LBS)for end users, such as turn-by-turn navigation, local search, and geo-coding. GPS and LBS havebecome widely deployed and useful tools for commerce, science, tracking, and surveillance.

Location-based Services (LBS): Which police car or ambulance is near a 911 caller usinga smartphone? When outside a building, it can be difficult for a caller to articulate their position,and for the dispatcher to identify a nearby police car or ambulance. These challenges led to slowresponse times for emergency services for mobile callers. With the advent of systems like E-911,a requirement where mobile phones would self-locate with GPS or ground-based localizationinfrastructure to automatically report location information to a 911 center, dispatching nearbypolice cars and ambulances to correct locations is faster and more accurate. This is an exampleof Location-based Services. The OpenLS standard, a technical specification for LBS, describesa set of core services: location utilities, directory service (geocoding), presentation service,gateway service, and route determination service [28].

Geographic Information System (GIS): What countries are reachable by North Korea’smissiles? Figure 1 is a well known example of erroneous distance information computed on aplanar map using circular distance, an easy mistake without the help of GIS supporting sphericalmeasurements to a flat paper map. GIS can understand a large number of map projections usedby common geographic data producers and aid in fusing map data from diverse sources. As theEarth is not a perfect sphere, GIS systems also understand more accurate representations ofthe Earth, such as ellipsoid representations and non-parametric representations that use land-based geodetic reference points for localization. Geographic Information Systems capture, store,analyze, manage, and visualize spatial data [14, 29]. While today’s GIS systems contain spatialstatistical analysis, and spatial database management systems, they retain a number of uniquecapabilities such as map projections, cartography, geodetic data, and map layers. For example,a map of the earth is a representation of a curved surface on a plane. While map projectionsretain topological properties (e.g., neighbors), they do not retain most metric properties (e.g.,distance, area).

Spatial Statistics: Is there an outbreak of disease? Where? In 1854, Dr. John Snow man-ually plotted Cholera locations on a street map of London to visually identify outbreak hotspotaround the Broad Street water pump. It took several days to perform this analysis even for onedisease over a small geographic area. Today, disease surveillance organizations monitor scoresof infectious diseases over very large geographic areas using spatial statistical tests for detectingoutbreaks (e.g., scan statistics) and hotspots and distinguishing those from natural variations.These techniques are also being used to analyze spatio-temporal event datasets in public safety(e.g., crime reports), transportation (e.g., accidents), etc. Spatial statistical theories (e.g., pointprocesses, spatial auto-correlation, geo-statistics) address unique challenges (e.g., violation ofindependent identical distribution assumption) in applying traditional statistical models (e.g.,linear regression, pearson correlation coefficient) to geographic data.

Spatial Database Management System (SDBMS): Which national parks are withinan hour driving distance of Route 66? Before the development of spatial databases, these typesof spatial queries required extensive programming and suffered from long computation times,due to the mismatch between 2-dimensional spatial data and the 1-dimensional indexes used

3

(a) Flat Earth (b) Spherical Earth

Figure 1: Geographic Information System: A 2003 article in The Economist significantly under-estimated the distance that North Korean missiles could travel because its map did not account forthe spherical shape of the world. The correct version is shown on the right [7].

by traditional database systems (e.g., B+Tree). Spatial Databases [27, 28], such as OracleSpatial and PostGIS, introduced spatial data structures (e.g., R-trees, voronoi diagrams) andalgorithms (e.g., plane-sweep, shortest-path, nearest neighbor, range query) to efficiently answerspatial queries. This reduced programming effort and resulted in more compact code and fastresponse times.

Spatial Data Models: What concepts (e.g., data types) and relationships (e.g., opera-tions) facilitate multi-stage queries of spatial data? A naive collection of spatial data typessuch as points, lines, and polygons is inadequate for multi-stage queries since the result of somequeries (e.g., union of disjoint polygons) cannot naturally be represented as a point, line, orpolygon. The Open Geospatial Consortium [22] proposed an OGIS standard with 6 spatialdata types (point, linestring, polygon, point collection, linestring collection, polygon collection)to facilitate multi-stage queries (e.g., union of two disjoint polygons is represented as a poly-gon collection). It also provided topological operations (e.g., inside, overlap, disjoint, touch)based on a mathematically rigorous 9-intersection model, and metric operations (e.g., distance,area). This facilitated industry commitment and investment as a consensus standard reducesinvestment risks. It led to SQL3-based products from IBM, Oracle, Microsoft, Postgres, etc.

Digital Earth: What is the spatial distribution of a population before and after an earth-quake? In the past, two challenges prevented fast and efficient response to this question. First,detailed spatial information (e.g., remotely sensed imagery, census data) did not have the broadaccess it has now through the advent of tools such as Virtual Globes (e.g., Google Earth, NASAWorld Wind, Bing Maps). Second, surveying population location and movement was a timeconsuming and expensive procedure. With today’s Volunteered Geographic Information sys-tems, such as Ushahidi [21] and Open Street Maps (OSM), everyday citizens can create maps,submit information, and aid in overall relief efforts. For example, after the Haiti earthquakedisaster, people across the world used OSM to digitize post-earthquake arial-imagery to quicklyupdate maps of Haiti’s roads, hospitals, triage centers, and refugee camps that were still func-tioning [20]. These collections of tools enable and extend Al Gore’s vision of a “Digital Earth”where citizens could not only “access vast amounts of scientific and cultural information to helpthem understand the Earth and its human activities,” but also contribute new information [2].

4

Table 2: Cultural Shift in Spatial Computing

Late 20th Century 21st Century and Beyond

Only sophisticated groups (e.g., Department ofDefense, oil exploration) used GIS technologies

Everyone is a mapmaker and every phe-nomenon is observable

Maps were produced by a few highly trainedpeople in government agencies and surveyingcompanies

Everyone uses location-based services

Only specialized software (e.g., ArcGIS, OracleSQL) could edit or analyze geographic informa-tion

Every platform is location aware

User expectations were modest (e.g., assist inproducing and distributing paper maps andtheir electronic counterparts)

Rising expectations due to vast potentialand risks

3 Cultural Shift

In the late 20th century, almost all maps in the US were produced by a small group of highlytrained people in government agencies and surveying companies. Only a few sophisticatedgroups such as the Department of Defense or oil exploration groups used GIS technologies.These groups depended on highly specialized software such as ESRI ArcGIS and Oracle Spatialfor editing or analyzing geographic information, and their goals did not extend much beyond thedistribution of paper maps and their electronic counterparts. As summarized by Table 2, recentadvances in spatial computing have changed this situation drastically. Today, “everyone” can bea mapmaker and most phenomena are observable, most people use location-based services, andmore and more platforms are location-aware. The very success of spatial technologies has raisedusers’ expectations of spatial computing in the future. At the same time, users increasinglyworry about the potential misuse of location data.

Everyone uses location-based services: The proliferation of web-based technologies,cell-phones, consumer GPS-devices, and location-based social media have facilitated the widespreaduse of location-based services [26]. Internet services such as Google Earth and OpenStreetMaphave brought GIS to the masses. With cell-phones and consumer GPS-devices, services suchas Enhanced-911 (E-911) and navigation applications are consumed by billions of individuals.Facebook check-in and other location-based social media are also used by over a billion peoplearound the world.

Everyone is a mapmaker and every phenomenon is observable: The fact that userswith cell phones and access to the Internet now number in the billions is a new reality of the21st century. Increasingly, the sources of geo-data are smart-phone users who may passivelyor actively contribute geographic information. The immediate effect is wider coverage and anincreased number of surveyors for all sorts of spatial data. More phenomena are becomingobservable in the sense that sensors are getting richer for 3D mapping and broader spectrumsat finer resolutions are being captured. This affords the ability to observe more phenomenaat higher levels of precision, but presents new challenges based on the increased data volume,velocity, and variety that are exceeding the capacity of current spatial computing technologies.

Every platform is location aware: Traditionally, spatial computing support was limitedto application software layers (e.g., ArcGIS), web services (e.g., Google Maps, MapQuest),and database management (e.g., SQL3/OGIS). In recent years, spatial computing support isemerging at all levels of computer architecture such as HTML 5, social media check-ins, InternetProtocol Version 6 (IPv6), and open location services (OpenLS).

5

Figure 2: Augmented reality applications are becoming commonplace for smartphones.

Rising expectations due to vast potential and risks: Location-based services, navi-gation aids, and interactive maps have arguably exceeded users’ expectations. Their intuitivebasis and ease of use have earned these products a solid reputation. Consumers see the poten-tial of spatial computing to reduce greenhouse gas emissions, strengthen cyber-security, improveconsumer confidence and otherwise address many other societal problems. However, the verysuccess of spatial computing technologies also raises red flags among users. Geo-privacy con-cerns must be addressed to avoid spooking citizens, exposing economic entities to liability, andlowering public trust.

4 Short-Term Opportunities

The above challenges provide new avenues of research in spatial computing. Solving these willgive rise to a number of new and interesting opportunities:

Spatial Abilities Predict STEM Success: Spatial abilities include navigation, learningspatial layouts, and performing mental rotation, transformation, scaling and deformation ofobjects across space-time. Spatial skills strongly predict who will go into and succeed in thefields of science, technology, engineering, and math (STEM) [31]. As it stands, our societyis facing a challenge in educating and developing enough citizens who can perform jobs thatdemand skills in STEM domains. Improving spatial training at K-12 levels is likely to increasethe number of students who excel in and pursue careers in STEM fields.

Emerging Spatial Big Data: Examples include trajectories of cell-phones and GPS de-vices (shown in Figure 3), mobile check-in’s, wide-area motion imagery, and location-basedsearch information. Spatial big data has the potential of providing new understanding andspurring innovation. A 2011 McKinsey Global Institute report estimated savings of $600 billionannually by 2020 via reductions in idling and fuel used via smarter navigation [13]. Location in-formation from cellphones will allow urban informatics, allowing for real-time census informationto be gathered for public health, safety and prosperity.

Augmented Reality Systems: Augmented reality enriches our perception of the realworld by overlaying spatially aligned media in real time. For example, it can alter a user’s viewof their environment by adding computer graphics to convey past, present or future informationabout a place or object as shown in Figures 2 and 4. It is already used in heads-up displays

6

Figure 3: GPS trace data can be used to highlight delays and slowdowns in driving, suggesting moreefficient routes.

in aircraft and has become a popular feature in smartphone applications. As lightweight, butpowerful, computer-driven eyewear becomes more commonplace, augmented reality will play acrucial role in specific fields such as medicine, architecture, tourism and commerce, engineering,civil/urban planning, assembly and maintenance, as well as general day–to–day intelligenceamplification.

Time-Travel and Depth in Virtual Globes: Virtual globes such as Google Earth, BingMaps, and NASA World Wind are being used to understand our changing planet in an enjoyableand interactive way. Time-travel and depth (e.g., subsurface, atmosphere) in virtual globes willprovide the ability to visualize historical and future scenarios on a global scale for use casessuch as visualizing changing arctic ice-sheets over recent decades, as well as climate projectionsunder alternative policy scenarios.

Spatial Predictive Analytics: Recent progress in spatial statistics [25] and spatial datamining have the potential to improve accuracy and timeliness of predictions about the futurepath of hurricanes, spread of infectious diseases, and traffic congestion, which have confoundedclassical prediction methods. Spatial models can be invaluable when making spatio-temporalpredictions about a broad area of issues including the location of probable tumor growth in ahuman body or the spread of cracks in aircraft wings or highway bridges.

Persistent Environmental Hazard Monitoring: Environmental influences on the airwe breathe, the water we drink, or the food we eat, can have significant impact on health andsafety [4]. Spatial computing (e.g., VGI, smart-phones and location-aware sensor networks) cangreatly enhance spatio-temporal precision and accuracy of exposure data in sensitive environ-ments such as schools, hospitals, fragile eco-systems, and vulnerable public gathering places.

Geo-collaborative Systems, Fleets and Crowds: Spatial computing will take the Inter-net beyond cyberspace, enabling connections among fixed structures and moving objects such ascars, pedestrians, and bicycles, to help prevent collisions or coordinate movement. For example,the city of Los Angeles recently interconnected all of its 4,500 traffic signals to improve traf-fic flow during rush hour through coordinated signals. Spatial computing enables smart-mobs(groups of people) to come together quickly for common causes without the need for any oneperson to direct. For example, drivers, smart cars, and infrastructure may cooperate in thefuture to reduce congestion, speed up evacuation, and enhance safety.

Localizing Cyber Entities: Location is fast becoming an essential part of Internet services,

7

Figure 4: Experimental augmented reality assistance for an aircraft engine assembly task.

with HTML 5 and IPv6 providing native support for locating browsers and GPS-enabled phoneslocating people on the move. Location authentication on the Internet may enhance cyber-security by helping verify the identity and location of message sources. For example, geo-targetedwarnings for people in predicted tornado paths can help save lives by reducing false warnings.Societal adoption will require trust in such systems against threats related to security, privacy,confidentiality, etc.

Moving Localization Indoors and Underground: Despite worldwide availability, GPSsignals are largely unavailable indoors, where human beings spend 90% of the time. Location-based services, such as route navigation, are currently present in 10% of our lives but withemerging technologies such as indoor localizations, routing, and navigation (already availablein major airports and hospitals), the new reality in the 21st century is that our spatial contextwill be available essentially all the time, leveraging localization indoors and underground (e.g.,mines, tunnels) via cell-phone towers, Wi-Fi transmitters, and other indoor infrastructure.

Beyond Geographic Space: Spatial computing ideas can transform information manage-ment in non-geographic spaces, such as those listed in Table 3. For example, defect locations onsilicon wafers may be modeled as statistical point processes to identify hotspots. Neuro-maps(shown in Figure 5) may organize patient data (e.g., MRI, CT) and intra-human body positiontracking may facilitate navigation along a least-invasive route to reach and remove brain tumors.Astro-maps chart the stars and interplanetary localization systems may improve space travel,whereas knowledge-maps plot our ideas and thoughts.

Geo-Privacy and Geo-Security: While location information (GPS in phones and cars)can provide great value to emergency response personal, consumers, and industry, streams ofsuch data also introduce spooky privacy concerns of stalking and geo-slavery [6]. Computerscience efforts at obfuscating location information to date have largely yielded negative results.Thus, many individuals hesitate to indulge in mobile commerce due to concern about privacy oftheir locations, trajectories and other spatio-temporal personal information [12]. Policy makershave an opportunity to improve consumer confidence by paving the way for next generationLBS respecting individual user privacy.

5 Long-Term Research Needs

Spatial computing is no doubt providing society with tremendous value. Out of these successes,however, significant challenges are emerging that will impact the future of spatial computing atall levels–science, systems, services, and society. Meeting these challenges will require expertisebeyond the realm of spatial computing itself. First, overcoming the challenges of everyone andanyone being a mapmaker and most phenomena being observable will require moving fromfusion of data from a few trusted sources to synergizing data across a multitude of volunteers.

8

Figure 5: An example of non-geographic space. Spatial computing ideas can transform informa-tion management in non-geographic spaces, such as the brain. Neuro-maps may organize patientdata (e.g., MRI, CT). Figure courtesy of Technische Universitat Munchen/ KAUST/University ofUtah/University of Konstanz/DFKI Saarbrucken.

Second, facilitating the spread of location-based services will afford widely available servicesfor everyone as opposed to services for only the GIS-trained few. Third, surmounting thechallenge of equipping every platform to be location-aware will move spatial computing from afew platforms (e.g., cellphones) to all platforms (e.g., sensors, PCs, clouds). Other opportunitiesdue to rising expectations are crosscutting such as geo-privacy and ubiquitous computing.

Spatial Computing Sciences – From Fusion to Synergetics: Historically, spatial com-puting science dealt with geographic data from highly trained GIS professionals in authoritativeorganizations with data quality assurance processes. Today, an ever-increasing volume of geo-graphic data is coming from average citizens via check-ins, tweets, geo-tags, geo-reports fromUshahidi, and donated GPS tracks. Volunteered geographic information (VGI) raises challengesrelated to data quality, trustworthiness, bias, etc. Such data requires transformation of tra-ditional data fusion ideas into a broader paradigm of data synergetics, thereby raising manynew issues. For example, we need to be able to manipulate qualitative spatio-temporal data inorder to reason about and integrate the qualitative spatial and temporal information that maybe gleaned from VGI (e.g., geo-tags, geo-reports, etc.). Spatio-temporal prediction may assistin inferring the described location of a tweet from its content. Additionally, since contendingnarratives in VGI data may lead to alternative maps of a common area from different perspec-tives, handling multiple competing spatial descriptions from the past and future is essential.Furthermore, spatial and spatio-temporal computing standards are needed to more effectivelyutilize VGI such as geo-tags with known geographical locations via history-aware gazetteers.

Spatial Computing Services – Spatial Cognition First: Previously, spatial computingservices were defined for a small number of GIS-trained professionals who shared a specializedtechnical language, not understood easily by the general public. With the public quickly be-coming mapmakers and using location-based services, there is a great need to understand thepsychology of spatial cognition. Such understanding will improve the use and design of mapsand other geographic information products by a large fraction of society. Further research onspatial cognitive assistance is needed to explore ideas such as landmark-based routing for indi-viduals who cannot read maps or for navigating inside a new space such as a building or campuswhere not all areas (e.g., walkways) are named. Understanding group behavior in terms of par-ticipative planning (e.g., collaboration on landscape, bridge, or building design) or smart mobsfor coordinating location movement will also enhance spatial computing services for groups ofpeople, as opposed to individuals. Context (e.g., who is tweeting, where they are, and physicalfeatures in the situation) should also be brought into each of these scenarios to investigate newopportunities for tweet interpretation for warning alerts during emergencies such as natural

9

disasters (e.g., Hurricane Sandy). New ways of improving the public’s spatial abilities (e.g.,navigation, learning spatial layouts, and reading maps) for different groups (e.g., drivers andnon-drivers) must be further investigated to leverage some of these opportunities.

Spatial Computing Systems – From Sensors to Clouds: In the 20th century, the pub-lic face of spatial computing was represented by spatial computing systems such as Arc softwareand Oracle Spatial Databases. Today, all levels of the computing stack in spatial computingsystems are being influenced by the fact that more and more platforms are location-aware due tothe widespread use of smart-phones and web-based virtual globes. Spatial computing infrastruc-ture will be needed to support spatial computing at lower layers of the computing stack so thatspatial data types and operations may be appropriately allocated across hardware, assemblylanguages, operating system kernels, runtime systems, network stacks, database managementsystems, geographic information systems, and application programs. Augmented reality willbe needed to accommodate devices such as eyeglass displays and smart-phones for automated,accurate, and scalable retrieval, recognition, and presentation of information. Sensing oppor-tunities exist for providing pervasive infrastructure for real-time centimeter-scale localizationfor emergency response, health management, and real-time situation awareness for water andenergy distribution. Computational issues [5] for Spatial Big Data will create new researchopportunities for cloud computing by addressing the size, variety, and update rate of spatialdatasets that currently exceed the capacity of commonly used spatial computing technologiesto learn, manage, and process data with reasonable effort.

Cross-cutting Issues in Spatial Computing: Emerging spatial computing sciences, sys-tems, and services give unprecedented opportunities for research and application developmentsthat can revolutionize our ways of life and in the meantime lead to new spatial-social questionsabout privacy. An example of the potential may be seen in the ubiquity of GPS-enabled devices(e.g., cell-phones) and location-based services. As localization infrastructure and map data setsreach indoors, there is expectation that the support that existed for an outdoor context willalso be available indoors. An example of the risks is the issue of geo-privacy. While locationinformation (GPS in phones and cars) can provide great value to users and industry, streamsof such data also introduce privacy concerns of stalking and geo-slavery. Computer science ef-forts at obfuscating location information to date have largely yielded negative results. Thus,many individuals hesitate to indulge in mobile commerce due to concern about privacy of theirlocations, trajectories and other spatio-temporal personal information.

Table 3: Representative spaces of interest in Spatial Computing

Outer Space Moon, Mars, Venus, Sun, Exoplanets, Stars, Milky Way, Galaxies

Geographic Terrain, Transportation, Ocean, Mining

Indoors Inside Buildings, Malls, Airports, Stadiums

Human Body Arteries/Veins, Brain, Genome Mapping, Chromosomes, Neuromapping

Micro / Nano Silicon Wafers, Materials Science

6 Final Considerations

Spatial computing promises an astonishing array of opportunities for researchers and entrepreneursalike during the coming decade. However, societal impact needs to be taken into account. It isvital that US policymakers clarify users’ geo-privacy rights. Without that, it will be difficult forspatial computing to achieve its full transformative potential. We must also acknowledge theunique and daunting computational challenges that working with spatio-temporal data poses.

10

Successfully harnessing the potential of these datasets will require significant US investmentand funding of spatial computing research. Currently, many spatial computing projects are toosmall to achieve the critical mass needed for major steps forward. Federal agencies need tostrongly consider funding larger and more adventurous efforts involving a dozen or more facultygroups across multiple universities. Bolder ideas need to be pursued perhaps by leveraging exist-ing mechanisms such as: NSF/CISE Expeditions in Computing, NSF Science and TechnologyCenters, NSF Engineering Research Centers, US-DoD Multi-disciplinary University ResearchInitiative, NIH Program Project Grants, US-DoT University Transportation Centers, US-DoEAdvanced Scientific Computing Research Centers, and US-DHS Centers of Excellence.

Furthermore, spatial computing scientists need more institutional support on their homecampuses. Beyond one-time large grants, it will be necessary to institutionalize spatial comput-ing research programs to leverage enduring opportunities as acknowledged by a large number ofresearch universities establishing GIS centers (akin to computer centers of the 1960s) on campusto serve a broad range of research endeavors including climate change, and public health. NSFCISE can establish computer science leadership in this emerging area of critical national impor-tance by creating a dedicated and enduring research program for spatial computing parallel toCNS, IIS, ACI, and CCF, given its cross-cutting reach.

A number of agencies have research initiatives in spatial computing [16, 18, 19, 17, 15] (e.g.,the National Cancer Institute’s Spatial Uncertainty: Data, Modeling, and Communication, andthe National Geospatial-Intelligence Agency’s Academic Research Program ). Neverless, spatialcomputing and the agencies themselves could benefit from multi-agency coordination to reducecompeting projects and facilitate interdisciplinary and inter-agency research. Spatial computinghas already proven itself as a major economic opportunity to our society and further spatialcomputing research can make possible even more revolutionary advances.

7 Acknowledgments

We thank the CCC Spatial Computing 2020 workshop participants, especially the organizingcommittee members: Peggy Agouris, Michael F. Goodchild (member, NAS), Erik Hoel, JohnJensen (ASPRS, AACI), Craig A. Knoblock, Richard Langley (Inst. of Navigation), EdwardMikhail (ASPRS), Ouri Wolfson, and May Yuan (Former President, UCGIS). We also thankthe Computing Community Consortium (CCC), including Erwin Gianchandani, Kenneth Hines,Hank Korth, and Eric Horvitz for guidance and valuable feedback. We thank Michael Evansand Dev Oliver at the University of Minnesota for all their help in the Spatial Computing 2020visioning activities. We also thank contributors to the Vision and Challenge Paper Track at the12th International Symposium Spatial and Temporal Databases [24]. We thank Kim Koffoltfor improving the readability of this report. We also thank the advising committee and theorganizing committee for their direction and leadership.

11

8 References[1] P. Agouris, W. Aref, M. F. Goodchild, S. Barbra, J. Jensen, C. A. Knoblock, R. Langley,

E. Mikhail, S. Shekhar, O. Wolfson, et al. From GPS and Virtual Globes to SpatialComputing-2020, Sept. 10th-11th, 2012.

[2] Al Gore. The Digital Earth: Understanding our planet in the 21st Century, given at theCalifornia Science Center, Los Angeles, California, on January 31, 1998, 1998.

[3] D. Atkins. Revolutionizing science and engineering through cyberinfrastructure: Report ofthe national science foundation blue-ribbon advisory panel on cyberinfrastructure. 2003.

[4] A. H. Auchincloss, S. Y. Gebreab, C. Mair, and A. V. D. Roux. A review of spatial methodsin epidemiology, 2000–2010. Annual review of public health, 33:107, 2012.

[5] M. De Berg, M. Van Kreveld, M. Overmars, and O. Schwarzkopf. Computational geometry.Springer, 1997.

[6] J. Dobson and P. Fisher. Geoslavery. Technology and Society Magazine, IEEE, 22(1):47–52,2003.

[7] Esri. Understanding Geodesic Buffering. http://www.esri.com/news/arcuser/0111/

geodesic.html.[8] M. Faulkner, M. Olson, R. Chandy, J. Krause, K. Chandy, and A. Krause. The next

big one: Detecting earthquakes and other rare events from community-based sensors. InInformation Processing in Sensor Networks (IPSN), 2011 10th International Conferenceon, pages 13–24. IEEE, 2011.

[9] V. Gewin. Mapping opportunities. Nature, 427(6972):376–377, 2004.[10] M. F. Goodchild. Citizens as sensors: the world of volunteered geography. GeoJournal,

69(4):211–221, 2007.[11] A. Krause, E. Horvitz, A. Kansal, and F. Zhao. Toward community sensing. In Proceedings

of the 7th international conference on Information processing in sensor networks, pages481–492. IEEE Computer Society, 2008.

[12] J. Krumm. A survey of computational location privacy. Personal and Ubiquitous Comput-ing, 13(6):391–399, 2009.

[13] J. Manyika, M. G. Institute, M. Chui, B. Brown, J. Bughin, R. Dobbs, C. Roxburgh, andA. H. Byers. Big data: The next frontier for innovation, competition, and productivity.McKinsey Global Institute, 2011.

[14] R. B. McMaster and E. L. Usery. A research agenda for geographic information science.CRC Press, 2004.

[15] R. Muntz, T. Barclay, J. Dozier, C. Faloutsos, A. Maceachren, J. Martin, C. Pancake,and M. Satyanarayanan. IT roadmap to a geospatial future, report of the committee onintersections between geospatial information and information technology. Washington,DC: National Academy of Sciences, 2003.

[16] National Research Council. Priorities for GEOINT research at the National geospatial-intelligence agency. National Academies Press, 2006. ISBN: 9780309101493.

[17] National Research Council. Successful Response Starts with a Map: Improving GeospatialSupport for Disaster Management. National Academies Press, 2007. ISBN: 0309103401 /0-309-10340-1.

[18] National Research Council. New Research Directions for the National Geospatial-Intelligence Agency. National Academies Press, 2010. ISBN: 0309158656 / 0-309-15865-6.

[19] National Research Council. Understanding the changing planet: Strategic directions forthe geographical sciences. National Academies Press, 2010. ISBN: 9780309150750.

[20] New York Times. Digital Help for Haiti. http://goo.gl/yYn66, 2010.[21] O. Okolloh. Ushahidi, or’testimony’: Web 2.0 tools for crowdsourcing crisis information.

Participatory learning and action, 59(1):65–70, 2009.[22] O. G. OpenGIS. Consortium. Inc, http://www. opengis. org.[23] B. W. Parkinson and J. J. Spilker. Global Positioning System: Theory and Applications

(volume One), volume 1. Aiaa, 1996.[24] D. Pfoser, Y. Tao, K. Mouratidis, M. A. Nascimento, M. Mokbel, S. Shekhar, and Y. Huang.

Advances in Spatial and Temporal Databases: 12th International Symposium, SSTD 2011,Minneapolis, MN, USA, August 24-26, 2011. Proceedings, volume 6849. Springer, 2011.

[25] B. D. Ripley. Spatial statistics, volume 575. Wiley-Interscience, 2005.[26] J. Schiller and A. Voisard. Location-based services. Elsevier, 2004.

12

[27] S. Shekhar and S. Chawla. Spatial Databases: A Tour. Prentice Hall, (ISBN 013-017480-7),2003.

[28] S. Shekhar, S. Chawla, S. Ravada, A. Fetterer, X. Liu, and C.-t. Lu. Spatial databases-accomplishments and research needs. Knowledge and Data Engineering, IEEE Transac-tions on, 11(1):45–55, 1999.

[29] S. Shekhar and H. Xiong. Encyclopedia of GIS. Springer, 2008.[30] N. Y. Times. Left-hand-turn elimination. http://goo.gl/3bkPb, 2007.[31] D. H. Uttal and C. A. Cohen. Spatial thinking and stem education: When, why and how.

Psychology of learning and motivation, 57:147–181, 2012.[32] Waze Mobile. http://www.waze.com/.

13