Evasive Guidance Law Against Proportional Navigation Considering Dynamic Lag

Abstract

Terminal-phase evasion serves as a last-resort measure to improve target survivability against missile threats, motivating analytical evasive guidance laws that are both transparent and deployable. However, existing approaches either idealize the target maneuver as a zero-lag input or rely on adjoint integration for online command generation, thereby hindering full exploitation of evasive potential. To address this, an analytical bang-bang evasive guidance law against proportional-navigation (PN)-guided missiles is proposed in this paper, which explicitly incorporates first-order target dynamics and determines the switching direction without online adjoint integration. The key idea is to derive an analytical adjoint representation of this dynamic-lag-aware engagement model and express the associated switching function as a convergent power series, thereby enabling an error-controlled finite-order realization for low-complexity evaluation. Two special cases, corresponding to integer PN gains and matched target-missile time constants, are also derived for further simplification. Representative simulations validate numerical consistency with the adjoint-integration baseline at substantially lower runtime and demonstrate an average miss-distance gain exceeding 2 m over the zero-lag formulation, with broader applicability indicated by Monte Carlo analysis and a three-dimensional (3D) realistic engagement. Hardware deployment on a TMS320F28377D flight controller achieves an approximately 50-fold speedup within typical flight-control timing budgets, supporting practical real-time onboard implementation.