SBIR-STTR Award

We are proposing a cognitive agent architecture that leverages Interactive Machine Learning, which will allow ground controllers and crewmembers the ability to train and retrain their cognitive agents during a mission based on new/novel mission data. Interactive Machine Learning (IML) is a human-centered paradigm in which end-users, e.g. crewmembers and ground-based Subject Matter Experts, iteratively build and refine the ML model through iterative cycles of input and review. Model refinement is driven by user input that may come in many forms, such as onboard data and communications, crew preferences, modified mission parameters along with a description of features and selection of high-level model parameters. IML is distinct from classical machine learning in that human intelligence is applied through iterative teaching and model refinement in a relatively tight loop. With this approach, our goal is to develop an architecture that provides end-users with the ability to interactively explore and adapt the training space with the goal of guiding the adaptation of the cognitive agent toward an intended behavior. This approach will allow crewmembers the ability to control how the cognitive agent learns and adapts during a long duration exploration mission, assuring that its performance improves and does not degrade over time. This work will present crewmembers and SME centered approach to applying IML methods to the design of a system that learns and adapts based on crewmembers inputs to the system. By leveraging IML we are able to address one of the primary challenges associated with the use of cognitive agents during a long-duration mission, the inability to adapt and learn from observation, instruction and interaction as missions proceed. Potential NASA Applications (Limit 1500 characters, approximately 150 words): We expect the IML approach to developing training models for multi-agent assignment along with the ability for end-users to retrain the system will be of interest to a number of groups within NASA, e.g., Gateway habitat. Of course, future Mars expeditions could certainly make use of our cognitive agent, as those are the types of scenarios we’ve used in our development. In particular, our agent will be of interest at JSC to the EVA Exploration Office, the EVA Strategic Planning and Architecture group, and the Exploration Mission Planning Office. Potential Non-NASA Applications (Limit 1500 characters, approximately 150 words): The proposed cognitive architecture will benefit a number of TRACLabs commercial customers. For example, Baker Hughes has already expressed interest in licensing some of the new capabilities being developed in previous cognitive agent efforts, particularly the ontology and anomaly management aspects. We expect the ability for end-users to direct the adaptation of the system will be of interest. Duration: 6

The focus of the Phase 2 effort is to expand the Intelligent Machine Learning (IML) approach taken in Phase 1 to develop a Human-Centered Intelligent Virtual Agent (IVA). IML is an approach that moves the development of machine learning models away from engineers and puts the development of the model in the hands of the end-user. A Human-Centered IVA is focused on continuously improving the Machine Learning (ML) models while also providing effective communication between the human crewmembers and the IVA. Human-centered IVA is a perspective on artificial intelligence and ML that algorithms must be designed with awareness of being part of a larger system that includes end-users. This approach allows the IVA to incorporate the knowledge, insight, and feedback of the end-users allowing for tuning and refinement of the ML models. The Human-centered IVA will assist crewmembers in various tasks e.g., crew scheduling, procedure creation, and anomaly detection and resolution during a long-duration mission. ADAPT's Human-Centered IVA will provide the computationally heavy-lifting while still receiving inputs and insights from the crewmembers. This allows for the expansion of processes and information to a larger scale without compromising data integrity or mission success due to a lack of ground assistance. To provide crewmembers with a Human-Centered IVA this effort leverages the supervised learning algorithms Decision Tree, Random Forest, Ada Boost, Gradient Boost, Extreme Gradient Boost, Categorical Boost, and Associative Rule Models which were shown in Phase I to be succesful within an IML approach. Additionally, this phase will focus on providing an explainable interface that allows the end-user to query the IVA for the reason behind its prediction. This will be accomplished using an interactive visualization Graphical User Interface. Potential NASA Applications (Limit 1500 characters, approximately 150 words): We expect the Human-centered Intelligent Virtual Agents (IVA) approach to improving model predictions throughout all phases of a long-duration mission will be of interest to several groups within NASA. The ARTEMIS program for example could make use of IVAs to assist the crew in similar scenarios used during the development. Additionally, this work will be of interest to the EVA Exploration Office, the EVA Strategic Planning and Architecture group, and the Exploration Mission Planning Office. Potential Non-NASA Applications (Limit 1500 characters, approximately 150 words): The proposed cognitive architecture will benefit several TRACLabs commercial customers. We expect the ability of end-users to direct the adaptation of the system will be of interest. For example, Baker Hughes has already expressed interest in licensing some of the new capabilities being developed in previous cognitive agent efforts, particularly the ontology and anomaly management aspects. Duration: 24