This Small Business Innovation Research (SBIR) Phase I project addresses development of an efficient integrated multi-agent system for distributed environment mapping, monitoring, and anomaly detection for water surface mobile aquatic robots dispersed in a riverine, lake, estuary, or coastal environment. The overall system is comprised of multiple ingredients including (1) distributed data fusion and environment mapping, (2) collaborative localization, (3) bandwidth-efficient information representations and communication protocols, (4) distributed anomaly detection and alarm flagging based on customizable perceived spatio-temporal patterns, (5) multi-agent planning. The focus in Phase 1 is on the first three ingredients above and preliminary work on the fourth item. The system also integrates autonomous behavior for ad-hoc teaming based on communication availability and is designed to be scalable in the number of mobile robots and their sensory complements and mobility characteristics. The system integrates a light-weight distributed layer that is specifically optimized to operate on low-bandwidth or intermittent communication channels. The system is designed to be modular and facilitate extensibility to a wide range of platforms and operating environments and ease of integrability into third-party hardware and software systems (such as autopilots, operator ground stations, and higher-level mission planners) via a range of supported communication interfaces. The broader impact/commercial potential of this project includes a broad range of applications that would benefit from the use of teams of geographically dispersed mobile agents such as autonomous water surface aquatic robots in riverine, lake, estuary, or coastal environments. An integrated and flexible system for autonomous multi-agent environment mapping, monitoring, and anomaly detection provides a key enabling capability for effective applicability of collaborating teams of autonomous mobile robots. The efficacy of the system is further enhanced by robustness to intermittent or limited availability of GPS and inter-agent communication links and extensibility to a variety of mobile agent platforms and sensor complements. While the proposed project focuses on application to water surface robots, the approach is extensible to a wide range of aerial, ground, and space robot platforms. The proposed system has a wide market in a range of civilian, commercial, industrial, government, and military applications including bridge and dam monitoring, monitoring and rapid response for industrial effluent spills and oil spills, emergency communications relay, surveying after floods, weather monitoring, water quality monitoring, disaster relief, surveillance, port security, law enforcement, environmental research, marine biodiversity research, commercial ship inspection and surveillance, ocean circulation research, and aquaculture management
