Wearable Computing at the MIT Media Lab

I'm kind of amazed when I read this as it sounds a lot like the kinds of things I've been talking about for a while here on MVIS Blog: context-aware mobile computing devices. Seems to me like we should just import this whole team of MIT kids and let them run wild up there in Bothell.

Wearable Computing at the MIT Media Lab

To date, personal computers have not lived up to their name. Most machines sit on the desk and interact with their owners for only a small fraction of the day. Smaller and faster notebook computers have made mobility less of an issue, but the same staid user paradigm persists. Wearable computing hopes to shatter this myth of how a computer should be used. A person's computer should be worn, much as eyeglasses or clothing are worn, and interact with the user based on the context of the situation. With heads-up displays, unobtrusive input devices, personal wireless local area networks, and a host of other context sensing and communication tools, the wearable computer can act as an intelligent assistant, whether it be through a Remembrance Agent, augmented reality, or intellectual collectives.

Memory Glasses

The Memory Glasses is a wearable, proactive, context-aware memory aid based on the MIThril platform and wearable sensors. The primary goal of this project is to produce an effective memory aid and reminder system that requires a minimum of the wearer's attention.

The function of our system is to deliver reminders to the wearer in a timely, situation-appropriate way, without requiring intervention on the part of the wearer beyond the initial request to be reminded. In other words, the system behaves like a reliable human assistant that remembers reminder requests and delivers them under appropriate circumstances. Such a system is qualitatively different from a passive reminder system (such as a paper organizer) or a context blind reminder system (a modern PDA) which records and structures reminder requests but which can not know the user's context.

The Problem

People's lives are increasingly complicated, and the need for effective logistical and memory aids is on the rise. For years, digital technology and computers have promised effective solutions for the problems of organization and memory, but present technology offers little more than an incremental improvement over the paper scheduler and alarm clock of the 19th century. Even the most sophisticated PDA is deaf and blind, and knows no more about the wearer's context than the time of day.

The Solution

The solution we propose is a proactive, context-aware memory aid called the Memory Glasses. By creating a reminder system that is context aware, the role of the system is transformed from a blind, passive organizer to a proactive, perceptive entity akin to a human assistant. The delivery of information is situation conditioned; reminders are associated with context and are delivered when and where appropriate.

There are a number of open questions to be addressed in the construction of a proactive, context aware reminder system. These questions may be broken down as follows:

Defining Useful Context: What is useful context for proactive reminders? Assuming that we cannot know everything about the wearer's state and actions, what are the most important things to know? Of these, which are technologically feasible to sense and classify?

Time and location are obviously useful. In addition, it may be useful to know socially or logistically important features of the wearer's activity state, such as "in a conversation with person X" or "driving to work."

Context Sensing and Classification: How do we sense and classify useful context? What types of sensors, signal processing, and inference techniques are necessary? What are the bandwidth and computational requirements for this classification?

Time and location are relatively easy to sense and classify. Aspects of the wearer's activity state which are independent of time and location are harder. High-bandwidth, computationally expensive computer perception techniques may be used to provide infrastructure-free sensing and classification, and low-bandwidth low-computing power tag readers and tags may be used in cases where tagging infrastructure, people, and objects is feasible.

Wearer Interaction: Humans have limited bandwidth, and a proactive information delivery system must always assume that it is a secondary task. One of the hardest questions is how to present information to the wearer in a way that supports their interaction with and involvement in the world around them. Arguably this is the most important question, since no amount of sophistication and effectiveness in context definition, classification and sensing can be effective if the wearer refuses to use it.

Our recent research suggests that it is possible to support memory recall without the user's conscious attention.