Simulation of Dynamic Inversion with Neural Network and its Applications to MAV

The MAV accurate modes are typically unavailable. Most traditional methods for system design are complex and can not get satisfactory effect. Recent work in dynamic inversion with neural network may be applied to control a MAV where the reference commands include position, velocity, attitude and angular rate. This control technology can provide the MAV with an admirably command follow and steady control capability. Neural networks are used directly as the controller as well as indirect designs that are based on a neural network process model. The emulator is compiled by using NNCTRL20 and NNSYSID20 in MATLAB. Then the main result of this simulation is presented. It indicates that the method has high precision and strong stability, which can serve the control of MAV. Thereafter, an amendment about controlling system with dynamic inversion of the MAV is described.