Select Page

Context awareness navigation systems discussed

Most of us are aware of location awareness services such as GPS (Global Positioning Systems) but the idea of Context Awareness Systems (CAS) is perhaps new to many of you. These systems are not new, in that they have been discussed for over 10 years, but these systems are only now really being developed and showing potential as navigational devices and as devices which truly aid the wayfinding experience. The true benefit of context awareness systems is and will be in the way in which they can take heterogeneous data and then be used through mobile technologies such as in smartphones and tablets, to offer us a better and more personal travel experience.

Can you imagine arriving in an airport and needing to catch a connecting flight and your mobile phone being able to guide you directly to whatever departure gate you need, whilst being able to direct you via the toilets (based on historical data that the app knows that you always like to freshen up in between flights) and also guides you via a coffee shop or bar. The app also gives you an alarm call if you are short of time to reach the gate. On arriving at your destination the app can guide you on the most appropriate way to get into the destination city, based on ground transportation availability and on the local weather!

Mobile context technologies

Mobile technology

What is a Context Awareness System

The key to these systems is the idea of ‘highly dynamic applications’. These systems are designed, for example, to consider the user’s surroundings, location, the climatic conditions and interpret relevant data from third parties, in order to create a better experience for you and to provide you with better options. First-hand and updated information using GPS sensors, sensors for movement, acceleration and dead reckoning to predict routes are all obvious features which can already be included. Based on your location and other conditions, the system can help you have access to the most appropriate services and information. These dynamic systems can work in two ways and they are via a:

  • Data-driven paradigm (i.e. data that comes in such as flight information or gate information, so information which is new one-way data)
  • and/or a Knowledge-driven paradigm (data based on previous habits i.e. knowledge that the system builds up your past usage).

In order to combine data-driven and knowledge-driven paradigms, a hybrid approach is needed in order to pull together the data-sets and to integrate the data for the awareness system. One of the big problems as you can probably guess, is that there is the potential for so many variables and choices in what data is included in a context aware system. This inevitably creates a level of fuzzy logic (i.e. irregular data) and this makes designing a truly context rich PNS system quite challenging. A fuzzy inference system is thus one inclusion which designers of new PNS might need to include in the design of the systems.

Personal Navigation Systems (PNS)

One of the key uses of a CNS is as a ‘Personal Navigation System’ and these already exist in that we have, for example, mobile phones which we can use with apps, which can be used for navigation in some indoor and many outdoor environments. We also have hand-held SatNavs or bike apps such as those by BikeHub.

What these systems do not have though, is context awareness and for those PNSs which now do, the integration with context awareness is quite low.

The integration in other words, between personal navigation systems and context awareness, despite being talked about over 15 years ago, is still in its infancy. We are though getting much closer to having the technologies and capability to create much more intelligent integrated systems, which will make the marriage between PNS and context awareness something of the future. Some features of a PNS include:

  • Factoring in the user’s personal pre-set preferences, the user’s activity and the local environmental conditions, i.e. how busy an airport is or how busy a motorway is. (Some SatNavs do already have elements of these features but lack, for example, the ability to make higher level decisions based on your travel history. SatNavs though in many respects, is the closest thing we presently have to a real context awareness system).
  • Context detection algorithm i.e. collates GPS and location data with the heterogeneous source data (such as that of the transport providers). The location information would in effect be statistical data-driven. There is though the chance for software designers to develop intelligence built on past histories i.e. knowledge driven.
  • Visual odometry and/or odometry – in computer-generated route estimation (odometry meaning the use of sensors for data to estimate positions. With visual odometry, it involves the use of images to help predict locations such as when images from a mobile camera are factored in).

Uses in Wayfinding

When we begin to understand that wayfinding and navigation are not always about getting from one place to another necessarily in the shortest time possible, but in navigating the route in a satisfactory manner – we then begin to actually better understand the way in which context-aware systems can truly benefit us in travel.

A person for example, as they walk through an outdoor attraction, one would assume would wish to avoid the pouring down rain and to walk a covered route in such conditions. In such as scenario, a context system which includes atmospheric and climatic considerations can be extremely useful.

The context system itself might also lead you in a certain route in order for the device itself to best maintain a powerful enough signal through which it can capture data streams from travel providers, from outdoor or indoor beacons and so on. What is clear is that the complexity of such a navigational system is in the multitude of data sources and this certainly creates design challenges. Such a system needs also to be able to understand when the user is in an indoor or outdoor environment.

The idea of context awareness in terms of PNS is that NO explicit user intervention is needed i.e. the system will understand or at least learn to understand the user’s habits. This is how, for a car SatNav, rather than needing to program in that you prefer longer routes that use main roads, the system will detect this. In an airport, an indoor PNS would give you updated flight information, direct you to the gate, give you alerts if you are too far from the gate as boarding commences. United Airlines have teamed up recently with Uber to integrate the flight experience including, for example, to tell you about available ground transportation. Whilst progress, this is not context aware and thus there is so much that can be done moving forward.

For those of you walking, the context awareness system should be able to measure your acceleration using an accelerometer, or complex directions through a gyroscope (a rotational motion sensor) and personal movement and actions via a magnetometer (which are now being used in phones to detect a person’s movement through the use of magnets).

What a Context System needs

When you consider that data will need to be imported and processed using multiple data sources and that the system most likely will need to use multiple sensors, one of the greatest difficulties is that a complex algorithm is needed to process the heterogeneous sourced data and to make sense of it. The output then needs to be pushed to the user in such a way that:

Sensor fusion

Context-aware sensor fusion

Information Fusion

One of the key elements to get context awareness to work is to have information fusion. By this I mean that there has to be a way to fuse and make sense of the different types of data, be it from a gyroscope, camera, historical data and data feeds such as those by an airline feeding in departure changes or connection information.

Indoor Beacons

One of the most exciting and pragmatic moves for indoor navigation and which certainly can aid context driven systems is the development, in recent years, of ‘indoor beacons’. These technologies can communicate with your smartphone in real time and place your position within the environment i.e. the beacons help your mobile device to act as a more accurate indoor navigation system. Companies such as Indoo and Apple’s iBeacon are two such products in the market. Indoo’s system also includes features such as what they call ‘GeoFencing’ – which informs navigators when they go into specific areas.

Using Magnetometers – Video from MagiTact

The use of magnetomoters is particularly interesting and offers a real way forward for improving context awareness in many different environments, including in travel situations. Imagine if your heartbeat is rising because you are stressed as you queue in an airport and then your phone, through which you listen to music, changes to more relaxing music to try and relax you.

Case Study: Everyday Traveller

It might be useful to give an example of context awareness services, in a real travel situation, to explain the type of functionality one would expect to see soon or in the future. In an airport you could for example expect:

  • Guidance to your specific check-in area and to your specific departure gate.
  • Automated alerts for those in the same travel space who match your criteria (i.e. friends, specific business type user).
  • Navigation and directional guidance.
  • Automated arrival information and local guidance i.e. for transportation options.
  • Heartbeat checks and automated music according to your mood.
  • Automatic route guidance based on your available time and historical preferences.
PNS diagram for context awareness navigation

Personal Navigation System for context awareness navigation – by Paul Symonds (2014 – Travelwayfinding.com)

The functionality of the Systems for Wayfinding

Context awareness designed systems for aiding in navigation, be it for the visually impaired, for those trying to get around unfamiliar environments, or for those of you just trying to navigate complex and large locations, there are a number of factors which ideally should be included in the design of such a system.

  • Global positioning (including for indoor environments).
  • Object avoidance.
  • Speedometer and audio direction giving (many systems now provide this).
  • Cardinal directional guidance.
  • Weather guidance – wayfinding routes often need to be adapted for climatic conditions – see the definition of Wayfinding by Symonds (2014) which sees wayfinding as an experience. Rain outside a building can change the route to being one which is inside and undercover. On a sunny day, the decision can differ and an outdoor route preferred.
  • User preferences (i.e. preference for more distant green areas or for very direct routes).
  • Socially connected (i.e. so that you can meet up with colleagues or friends who happen to be navigating the same travel space).
  • Integrated systems to provide environmental information such as present weather and timetables.
The following two tabs change content below.
Dr Paul Symonds has a PhD in Wayfinding from Cardiff Metropolitan University in the UK. Paul works with the signage industry, airports and other locations providing wayfinding audits, consultancy and training.