Advances in tracking and location technologies are enabling new application development. These applications can be found in environmental, utility, transport and other industrial applications. Many new approaches are being developed for more personal and local use and these applications include in-store location, hospitals, airports and inside buildings. In this column Steven Feldman takes a look at some of these applications and provides his unique views on some of the recent developments.
Most of us take GPS for granted today, the benefits of the Global Positioning System can be experienced in navigation devices, personal and vehicle tracking systems and a wide range of applications. The almost universal incorporation of GPS into the latest generation of smartphones has spawned a host of location based applications for the consumer and business.
A limitation of GPS has been the need to maintain line of sight to the satellites potentially causing GPS black holes where positioning is lost or becomes inaccurate, e.g. dense urban areas with high buildings. To some extent this problem has been solved by supplementing the GPS positioning with location inferred from cell towers and known WiFi hotspot locations.
Skyhook, an alternative positioning provider, have developed a software based solution that combines these 3 sources of location information to determine the location of a mobile device with greater accuracy and faster resolution times that claims to achieve accuracy of 10-20 metres even in challenging environments.
But even these hybrid position systems don’t help us when we step inside a large office building, hospital or industrial complex or shopping mall, within these environments inferred location is limited. However a range of technologies including RFID, ultrasound, ultra wideband radio and WiFi location are opening up the possibility of more precise location determination within buildings and with that a whole new range of applications.
Applications
For a few moments let’s park the question of which technology will provide ubiquitous high accuracy location indoors and ponder the potential for what some are calling Local Positioning Systems or micro-location. Broadly this next generation of location based applications can be divided into two categories; personal navigation and the tracking or positioning of assets.
Trying to find your way around a large hospital complex, airport or office building can be a challenging and frustrating task, trying to find your way out in an emergency when possibly visibility is restricted could be even worse.
In addition to navigation, content could be streamed to users dependent on their location – information updates could include case notes for a doctor called to a hospital bed, lecture notes for a student entering a lecture hall or perhaps the holy grail of LBS, location based advertising in the form of details of special offers within a shopping mall or even as you are walking down the aisle of a supermarket. At exhibitions and trade fairs location sensing devices could determine what stands were visited and provide digital details about products on request in preference to the carrier bags of leaflets that rarely get read.
Local positioning also offers the potential to track people’s movement through a facility enabling an understanding of flows, the location of key specialist personnel and potentially the detection of unauthorised entry to restricted facilities. In hospitals and prisons precise positioning could provide increased levels of staff safety particularly if devices are equipped with some form of panic alarm.
Tagging objects with a location determining device opens up applications from asset management through inventory control to optimised manufacturing workflows. Many environments utilise mobile or portable assets, frequently time is lost locating a specific piece of equipment and often surplus inventory is required to compensate (some hospitals have estimated up to a third). Mobile assets are also inevitably vulnerable to misappropriation and the introduction of a geofence around tagged items can greatly reduce such losses.
Motion detection, either through a sensor or through positioning signals can provide an indication that an item is in use and the conjunction of a tagged member of staff with an asset can help identify who is using it.
In automated manufacturing lines, location tags are being used to track vehicles through the production process and to ensure that the correct custom options, tools and settings are used in assembly and recorded for quality control purposes.
In a shopping centre the paths of shoppers can provide valuable insight into behaviour and interplay between different retail brands can help to optimise the tenancy mix and marketing strategies for centre landlords who are increasingly sharing risk and success with their tenants. Within supermarkets tracking the paths of trolleys through the store and linking to purchases recorded at the checkout can optimise merchandising and even drive a more personalised shopping experience and in-store advertising.
These are just some of the micro-location applications that are already in use, the range of potential applications will explode as the sensing technologies become more accurate and hopefully integrate with our universal communications device, the mobile phone. As in all systems design, the applications will determine the most appropriate technology in terms of accuracy, frequency, volume of tags, power, area to be covered and infrastructure.
Technology modes
Looking at some of the different technologies that are available, they fall into two groups or modes; those that leverage the current capabilities of a mobile phone and those that require a dedicated tag. Additionally some approaches are active in that they broadcast a signal that enables location determination at regular intervals down to a few seconds and some are passive relying on the device passing close to a sensor for its location to be determined. Listed below are a few of the most promising technologies.
{sidebar id=303 align=right} Solutions that utilise existing infrastructure and infer location from it have a cost advantage over those that require a dedicated infrastructure to be installed.
Skyhook have developed a database of the location of over 100m WiFi access points which can be used as part of a positioning system, it is not yet sufficiently granular to provide accurate positioning within building complexes. WiFi has the potential to be extended from external use to internal where building complexes have a large number of wireless access points. The local databases of WiFi locations will need to be accessible to a mobile device in a similar manner to the Skyhook database which was recently announced to have topped 100m points worldwide, perhaps Skyhook have plans to licensed their positioning algorithms for use in these restricted environments.
TV signals provide another method of locating a device – they are high power, low frequency and consequently have the ability to penetrate most buildings. Rosum have developed a system which utilises a dedicated chipset to be integrated into the handset or tag combined with a monitor unit that is installed in the building to provide local calibration of TV signal strength and time delays. It is claimed that room level accuracy can be achieved within multi-storey buildings.
Path Intelligence, a UK based company, have developed an alternative approach using the identifier signals broadcast by a mobile phone to the network to derive a medium resolution anonymous location across large areas such as shopping centres and large complexes with the deployment of a small number of sensors to triangulate signals. The system provides “path intelligence” rather than individuals’ locations although inevitably it has prompted some concerns over privacy.
Whilst these solutions largely use existing infrastructure and in the case of WiFi and Path Intelligence could use relatively unmodified mobile phones, accuracy may be inadequate for some applications. A dedicated infrastructure and/or tag combination is required to achieve metre or better resolution. There are several solutions based upon radio waves – RFID, Ultra Wide-Band and Bluetooth appear to have the most potential at the moment.
RFID typically operates over short distances, usually 10-50cm for a passive (unpowered) tag and up to 10-20m for an active (powered) tag so location determination in a large area requires a significant number of scanners to detect tags. However for monitoring people or objects leaving or entering a space where access routes are limited the technology is highly effective. In well controlled and defined environments such as warehousing, manufacturing and processing flowlines RFID offers a cost effective but somewhat restricted solution. The range of detection of active and passive tags is increasing and they could provide low cost medium accuracy solutions for high volume applications in the future.
Ultrasound offers room and sub-room accuracy but requires a receiver to be located within each room because ultrasound does not penetrate solid walls. The tags and receivers are relatively low cost offsetting some of the disadvantages of requiring room level sensors. Some successful experiments have been undertaken to triangulate ultrasound signals to provide sub metre accuracy.
UWB currently offers the most precise location determination available in a commercially deployed solution. The lower power signals are suited to sensitive environments such as hospitals where interference with other instrumentation is a major concern and short pulses allow higher numbers of tags to communicate than some other systems. Ubisense claim 15cm 3D accuracy for their system which is deployed on several automotive production lines.
Map intelligence
Any application of indoor location needs more than a set of coordinates. Applications will often benefit from maps or more likely geo-referenced CAD drawings to provide some context and a framework for any spatial rules or processes that will form part of the application. As indoor location applications gain traction one can foresee the Building Information Models (BIM), such as those that Autodesk and Bentley have been promoting for several years, will become the base mapping for local positioning systems.
Personal navigation applications may require simplified pedestrian routing capabilities in large complexes combined with a less detailed map than would be required for in building tracking. Evacuation routes could also be incorporated, perhaps in a format that could be downloaded to the device to avoid a dependency on connectivity that may fail in an emergency.
Handover
For personal navigation to be a seamless door to door experience we will need a single connected device to support global and local positioning. WiFi location at its best offers a viable solution that is already embedded in many mobile phones, the challenge that remains to be solved is to provide a seamless handover from a Google Maps type application to a detailed site or building map. Either the navigation would need to switch to an application hosted by the complex owner/operator for the last part of the journey or perhaps more likely someone will offer the capability for building owners to upload their layouts into one of the GYM platforms (I seem to recall that Microsoft had a fairly simple way to host a local tile server that could merge geo-referenced CAD drawings into Virtual Earth), just leaving the navigation bit to sort out.
Privacy
When Google recently launched Latitude, their location sharing application, there was a flurry of concern in the media about issues of privacy. Some have argued that in today’s world we have almost given up on privacy as we skitter around the mobile web leaving a digital footprint that is becoming ever more detailed. Others (in my opinion those who have a limited understanding about the technologies) are concerned that these applications would allow “them” to spy on “us” and certainly the idea that Government, Google, my employer or even a partner or friend should be able to track me without my explicit knowledge and agreement would be just reason for concern.
Nothing changes with Local Positioning except that the information is more granular and one can only hazard a guess at the embarrassing or confidential information that could be inferred from such precise information
Applications that are tag based will usually be deployed by employers or the owners of buildings that we visit, who will make their use a condition of employment or entry. No doubt there will be some interesting issues arising around how the information from these devices is used and how long the information is retained.
A final thought
When GPS was first conceived it was for state of the art military applications, it would have been difficult to imagine the mass appeal of the technology or the range of applications that we have today. No doubt somewhere within the military today advanced Local Positioning technology is being developed for battlefield and training applications, in a few years time this technology will be commonplace and will probably be integrated with our communications device which may still be called a phone.
What will be the tipping point? Perhaps it will be a robot. If we are going to see the deployment of mobile robot devices they will need ways of navigating through the space that they occupy. They could learn through experience but initially they will want on board navigation and positioning.
——————————————————————–
About the Author
Trained as an economist, Steven Feldman spent the first 20 years of his working life in building materials before stumbling upon GIS. He has been Managing Director of GDC and MapInfo UK and is now an independent consultant and mentor as well as Chair of the AGI GeoCommunity Conference.
Read more at http://giscussions.blogspot.com or follow him at http://twitter.com/stevenfeldman
