The JHU/APL Path Planning team has developed path planning techniques to look for paths that balance the utility and risk associated with different routes through a minefield. Extending on previous years' efforts, we investigated real-world Naval mine avoidance requirements and developed a tactical decision aid (TDA) that satisfies those requirements. APL has developed new mine path planning techniques using graph based and genetic algorithms which quickly produce near-minimum risk paths for complicated fitness functions incorporating risk, path length, ship kinematics, and naval doctrine. The TDA user interface, a Java Swing application that obtains data via Corba interfaces to path planning databases, allows the operator to explore a fusion of historic and in situ mine field data, control the path planner, and display the planning results. To provide a context for the minefield data, the user interface also renders data from the Digital Nautical Chart database, a database created by the National Geospatial-Intelligence Agency containing charts of the world's ports and coastal regions. This TDA has been developed in conjunction with the COMID (Cooperative Organic Mine Defense) system. This paper presents a description of the algorithms, architecture, and application produced.
[1]
Christine D. Piatko,et al.
Stochastic search and graph techniques for MCM path planning
,
2002,
SPIE Defense + Commercial Sensing.
[2]
I-Jeng Wang,et al.
Path planning for mine countermeasures
,
2003,
SPIE Defense + Commercial Sensing.
[3]
Christine D. Piatko,et al.
Minefield path planning: architecture and algorithms obeying kinematic constraints
,
2004,
SPIE Defense + Commercial Sensing.
[4]
Thomas Bäck,et al.
An Overview of Evolutionary Computation
,
1993,
ECML.
[5]
Richard Bellman,et al.
ON A ROUTING PROBLEM
,
1958
.