The Effects of Target Location Uncertainty in Game Theoretic Solutions to Optimal Trajectory Formulations
Abstract
The integration of a variety of Intelligence, Surveillance, and Reconnaissance (ISR) assets is vital to the acquisition of knowledge critical to battlefield success. Missions performed by a penetrating Unmanned Aerial Vehicle (UAV) are of particular interest. The environment in which UAVs must operate includes Surface-to-Air Missile (SAM) sites with extended ranges, among other threats. SAM location uncertainty, terrain obscuration, and radar/sensor capabilities all contribute to the complexity of the situation. This thesis provides a game theoretic approach to determine optimal UAV strategies against enemy SAM sites. It is shown that most characteristics of the UAV or SAM have negligible effects on both image quality (Iq) and probability of kill (Pk) (probability of the SAM shooting down the UAV). Instead, SAM location uncertainty has the largest influence. After only 0.5 miles of uncertainty, the Pk of a UAV assuming perfect knowledge of the SAM location rises to 0.56. When the uncertainty rises to about four miles, the Pk rises to 0.99. When the UAV takes uncertainty into account, the results are not much better. Assuming that the SAM may be at one of three possible locations, the result is an average Pk of 0.49 or 0.79, depending on which optimization routine was used. Extending this situation to five, seven, and nine possible SAM locations results in an increase in Pk to 0.99 at seven locations. An even more realistic scenario involving the UAV optimizing a path through a large area of varying probabilities results in an 85% chance of getting shot down if the SAM is located within a five-mile radius of the center of the area. Outside of this area, the UAV is guaranteed to get shot down with a Pk of 0.99. Other techniques and methods must be explored and used in combination with Radar Cross Section (RCS) management to ensure the continuing collection of valuable ISR imagery in the coming years.
Document Details
- Document Type
- Technical Report
- Publication Date
- Jun 01, 2005
- Accession Number
- ADA436163
Entities
People
- Daniel M. Morales
Organizations
- Massachusetts Institute of Technology