NEURAL NETWORK-BASED FAULT-TOLERANT CONTROL METHOD OF UNDERACTUATED AUTONOMOUS VEHICLE

摘要

The present invention relates to a neural network-based fault-tolerant control method of autonomous vehicle underactuated. The present invention includes reconstruction of kinematic model and cooperative model based on a state transition with an underactuated device to tolerate a free unmanned moving body that lacks control input relative to a degree of freedom and follow a target behavior, compensation for lack of a control input, fault-tolerant input control input estimation using Nussbaum technique, and precise following and fault-tolerant control and underactuated operation on the target, by setting the bound condition of a following error.