At ATL, we develop decision-aids, information fusion, and inference technology to ensure that engaged military units have a clear understanding of the situation at hand.
Modern military operations require situation understanding by humans, robots, and other automated systems. The prevalence of mechanical and human information sources provides a wealth of data, but it is situation assessment processing, including information fusion, inferential logic, and contextual reasoning, that transform this data into actionable knowledge. We focus on the development of situation understanding technologies in two research foci: Perception and Information Fusion. We are developing revolutionary decision-aiding capabilities in intelligence preparation of the battlespace, information needs assessment, geospatial fusion, human representation and context extraction, hard/soft fusion, scene understanding, robotic navigation, full motion video context extraction and search & retrieval mechanisms, and knowledge-based computer vision - critical technology discriminators for Lockheed Martin and visionary enablers for the military.
Situation Understanding is the goal of every fighting force in today's military. Our computing systems can assemble actionable information from data to help realize this goal. Sharing pertinent information throughout military echelons becomes an even greater challenge in the Joint war fighting arena against an asymmetric threat. Automated context generation and exploitation is a pervasive theme in all of our Situation Understanding programs. We apply this paradigm to diverse applications, like the automated sharing of situation assessment information between nodes in distributed tactical networks, in video processing for more accurate object recognition and identification, and to operational plans, to direct our monitoring of the plan’s execution. Our context-driven systems provide up-to-the-minute situation understanding, supporting continuous, high optempo military activities.
Technologies under development include:
- Knowledge-based perception for scene understanding applied to crowd navigation, obstacle avoidance, and unmanned ground vehicle end-user situation awareness
- Human terrain network representations that permit continuous update from the fusion of intelligence data and enable multiple types of effects diffusion algorithms to be executed across appropriate nodes and relationships
- Soft fusion techniques that assess human-generated information context, bias and accuracy, and exploit that information in the processes of soft fusion and hard/soft fusion.
Teachnologies ready for transition include:
- Real-Time Multi-sensor Multi-target Data Fusion for military aircraft pilots and decision aids, from on-board and off-board sensors.
- Plan execution monitoring, anomaly detection and information needs assessment capabilities.
- Agent-based automated context generation services for nodes in tactical net-centric environments, where nodes may be warfighters, legacy systems, data sources, and platforms, to enable the automatic and continuous sharing of situation assessment.
Situation Understanding is one of several Business Areas for the Contextual Systems Laboratory