adaptive dynamic programming

In this paper, a three-dimensional robust guidance law for a surface to air missile considering actuator saturation and first order dynamic for autopilot system has been designed in order to enhance the performance of defense systems. To attain this goal, first, modeling of the system in 3D spherical coordination using engagement basics has been derived and after that, appropriate cost function for collision of missile and target considering actuator constraints and in absence of target maneuver information has been formulated. Hamilton-Jacobi-Isaacs (HJI) differential equation inequality should be solved according to robust control literature for achieving guidance law which unfortunately does not have closed form solution in our case study problem. Therefore, to overcome this challenge, using adaptive dynamic programming theory for solving acquired HJI, algorithm for designing robust guidance law has been presented. Simplification solving of differential inequality and also guaranteeing robustness of the controller are the most important feature of the proposed algorithm. Numerous simulations for targets with different maneuvering capabilities and comparison of the proposed method with conventional augmented proportional navigation, show effectiveness of designed 3D robust guidance law.

Lateral vehicle’s control with constant longitudinal velocity using adaptive dynamic programming, backstepping and zero-sum games theory is investigated in this paper. The nonlinear dynamics is considered and the steering torque is chosen to be the control input instead of the steering angle. At first, a subsystem is created by augmenting the lateral vehicle’s dynamics with lane keeping ones considering the steering angle as the control input and the road curvature as a disturbance. Utilizing adaptive dynamic programming, neural networks and zero-sum games theory, the optimal control law is obtained and then, the results exerted on the second subsystem which is the dynamics of the steering angle and a control law is captured for which using the backstepping control method. Finally, performance of the proposed algorithm is demonstrated by applying it on a typical vehicle model.