Dr. Dan Olsen
Students working with Dr. Dan Olsen in the ICE, or Interactive Computing Everywhere, Laboratory see a future of "cubic inch computing," where the most powerful computer is one you can hold in the palm of your hand. Handheld computers, not a new technology, are very convenient because they can be carried anywhere. However, their small size makes them difficult to use. Dr. Olsen and his students have invented a technique which allows users to "spill" the small display on their handheld computer onto a larger surface. The user can then move their work around by grabbing the handheld computer and moving it. The user can scroll the information on the table top using their fingers. Upon completion of the project, the user can pick up the handheld and walk away.
The prototype created by the lab uses a Mitsubishi Diamond Touch table that works by detecting the electric charge, or capacitance, of the human body. The capacitance is transmitted through the fingers or through the frame of the handheld computer.
The ICE Lab tackles a variety of computing problems-from tiny, handheld computers to large autonomous robots. In the field of robotics, the students in the ICE Lab have focused on teaching robots to drive. One challenge of creating a robot which will drive itself is teaching it what areas safe and navigable and areas are not. Using interactive machine learning techniques, the ICE Lab has created a robot that can be taught simply by coloring safe and unsafe areas on the computer screen. Unsafe regions such as grass, snow, people, and cars, are colored in red. Safe regions, including carpet, roads, and sidewalks, are colored in blue. From a few drawing strokes, the robot learns what is safe and what is not. Then, while driving itself, the robot will automatically turn to avoid unsafe regions.
The ICE Lab also invented a "Smart Brush" technique that aids in computer graphics. The Smart Brush watches the image that is being colored or painted. It recognizes where the edges are and stops the flow of paint where it finds the edges. If the user's brush is far away from the edge, the paint stops. As the user gets closer and closer to the edge, the "pressure" builds until the paint finally moves across the edge. This gives the user more control than typical graphics programs. If the user wants to respect the edge, or stay inside the lines, they avoid coming too close to the edge with their brush, and the paint will not spill over. If they want to remove the edge, however, the user just moves closer until the paint slides over one edge in search of the next.
In addition to handheld computer technology, robotics, and graphics, the ICE Lab also delves into the field of family history and technology with the Extensible Lineage Manager, or ELM for short. ELM allows genealogists to easily find information missing from their records. ELM simplifies family history research and dramatically speeds up the family history research process in five simple steps.
A. The user enters the information that he or she has already gathered into the ELM user interface. This information can be entered by hand or imported through GEDCOM. The user then selects the individual of interest for whom they wish to find additional information.
B. Based on what has been entered by the user, ELM estimates dates and places that should be searched to find information missing from the individual of interest's records. There is no need for the user to calculate these dates by hand or record them elsewhere.
C. ELM then consults its online resources and suggests several appropriate search tasks which may result in the discovery of the missing information. ELM provides valuable expertise about the available online resources, telling users which available genealogical records are most likely to yield information regarding the desired time period and location of the individual of interest.
D. With ELM, search tasks can be performed automatically. ELM also sorts the results by how well they correspond to what the user already knows, reducing time spent rummaging through unrelated material.
E. With the click of a button, ELM merges the relevant search results back into the researcher's original information, eliminating the need to enter the information by hand. ELM's simple user interface (pictured above) is accessible to anyone regardless of their level of computer expertise.