The Personalized Assistant that Learns (PAL) started as a DARPA program to create intelligent software assistants for military officers. One of the main efforts within the PAL program was the Cognitive Agent that Learns and Organizes (CALO) project, a loose collection of artificial intelligence tools developed by SRI International along with several universities including Carnegie Mellon University, the University of Massachusetts, and Stanford.
As the User Experience Lead for PAL’s military deployment at General Dynamics, my job was to take CALO’s research-oriented tools and find ways to translate them into an operational environment. My team worked closely with SRI International and the Army’s 3rd Armored Cavalry Regiment to identify operational use-cases, design and develop new interfaces, and evaluate prototypes. We traveled to more than a dozen military installations in the United States and spent more than three months in Iraq deploying and evaluating PAL in a mission-critical environment.
Building software for government agencies can be especially challenging, however the tremendously positive response we received from the warfighters in theater ultimately made this one of my most rewarding projects to date. One officer asserted that he knew people who would be alive today if PAL had existed during his first deployment.
- Learning by Demonstration for a Collaborative Planning Environment
AI Magazine, 2012 PDF, requires login
- Learning by Demonstration Technology for Military Planning and Decision Making: A Deployment Story
Innovative Applications of Artificial Intelligence (IAAI), 2011, PDF
- Learning by Demonstration to Support Military Planning and Decision Making
Innovative Applications of Artifical Intelligence (IAAI), 2009, PDF