A nonlinear constant bearing guidance and adaptive autopilot design for BTT missiles

This paper is an integration of two earlier results on missile guidance and autopilot (Chuang, 1993; Chang, 1995). The guidance design is a nonlinear implementation of the constant bearing guidance. The guidance command in the form of acceleration is transformed into acceptable forms for the autopilot design. This design incorporates feedforward neural networks in the structure of adaptive control. Its weighting matrices and other unknown parameters are tuned online, and the approximation error of the neural networks is compensated via the technique of sliding mode control. Stability and performance for the integrated system are also investigated.