Ph.D Thesis

Ph.D StudentShima Tal
SubjectEnd-Game Guidance against a Highly Maneuvering Target
DepartmentDepartment of Aerospace Engineering
Supervisor PROFESSOR EMERITUS Josef Shinar (Deceased)


This study, motivated by ballistic missile defense, analyzes an interception scenario between a guided missile with noisy measurements and a target without information as an imperfect information zero sum pursuit-evasion game. The research, aimed towards improved homing accuracy, took several directions.

Since in a ballistic missile defense scenario velocities and maneuverabilities are not constant, a time-varying linear dynamic model is introduced. Simulations show that the guidance law using this time varying model provides a significantly improved homing accuracy compared to a time-invariant model.

Due to the noisy measurements an estimator is included in the guidance loop. The inherent convergence delay of the process creates great estimation errors. In this study the estimation process is modeled by a pure information delay of the target maneuver. Based on this model a new guidance law, compensating for the delay, is developed.

In the synthesis of a new guidance law the elements of the time-varying model and the delay compensation are integrated and its implementation is tested in a three-dimensional non-linear time varying interception scenario with noisy measurements, using a Kalman filter type estimator. The results demonstrate substantial improvements in the homing accuracy over previous guidance laws, representing a potential breakthrough in guidance law design.

This study also presents an efficient application of a multiple model
adaptive estimator applied in a planar interception scenario. Compared to a Kalman filter it provides substantial improvements in the homing accuracy.