The Wearable Computing Concept of wearIT@work

There are different approaches to defining wearable computing depending on the research direction and the application domain. In the WearIT@work project we focus on the interaction between the user, the system and the environment. In conventional mobile systems the interaction is based on a modified version of a desktop human computer interface (HCI) and follows the pattern shown in figure 1 (left).

To operate the system, the user needs to focus on the interface. This includes his attention as well as physical activity, in particular the use of his hands. As a consequence he can either interact with the system or with the environment, however not with both at the same time.

Thus to access data on a PDA (Personal Digital Assistant) the user must interrupt whatever he is doing, take the PDA out of his pocket and focus on the device. The above means that the range of applications and situations in which the system is useful is severely restricted.

In general, over a course of a day, mobile devices are actually useful at most between 1% and 5% of the time. In addition, many of today’s devices are bulky and obtrusive which means that for many activities the user does not even take them along.

In contrast, wearable systems are designed to be permanently useful and usable in a wide range of mobile settings. The corresponding interaction concept is illustrated in Figure 1 (right). It allows the user to simultaneously interact with the system and the environment. In addition, there is direct interaction between the system and the environment as well as the possibility of the system mediating the interaction between the user and the environment.

The implementation of the wearable interaction concepts involves four main issues:

1. The system must be able to interact with the environment through an array of different sensors distributed in different parts of the outfit. In particular must be able to develop a certain degree of awareness of user activity, his physiological and emotional state, and the situation around him like environment temperature, humidity, GPS coordinates. This is often referred to as context awareness.
2. The user interface needs to be operated with minimal cognitive effort and with no or little involvement of hands. In general the low cognitive load is achieved through appropriate use of the context information. Thus for example instead of having the user select a function from a complex hierarchy of menus, the system might derive the two most likely options from the context information and present the user with a simple binary choice. In terms of the actual input modality, simple, natural methods such as nod of the head, a simple gesture, or spoken command are preferred.
3. The system uses the context information to proactively support the user. This includes system self-configuration tasks as well as automatic retrieval, delivery, and recording of information that might be relevant to the user in a specific situation. An example is the automatic retrieval of maintenance specs as soon as the user gets near a certain maintenance point.
4. The system should be seamlessly integrated in the outfit so that it neither interferes with the user’s physical activity nor in any unpleasant way affects his appearance. This means that the devices have to evolve together with the clothing to be taken along nearly anywhere.

A trivial example of an existing device that adheres to the above requirements is a modern hearing aid computer. It is unobtrusive, useful during most of the day, requires hardly any cognitive effort to operate, and by definition of its function, mediates the user’s perception of the real world. In addition advanced devices are able to automatically adjust the volume between noisy and quiet settings and even optimize the amplification mode to suit the situation such as conversation or a concert.

In WearIT@work we address more advanced wearable systems. They can detect complex activities such as social interaction, or certain specific work related actions (e.g. in maintenance) and use this information to deliver a variety of services exactly tailored to the user needs in a given situation.