Scheduling Gain-Based Adaptive NN Control of Underactuated Bridge Crane

Bridge crane is an important industrial device, and is widely applied to many fields such as automobile making and marine equipment. However, the existing control methods have encountered some challenges in the applications owing to the underactuation, strong nonlinearity and external disturbance. This work proposes an adaptive control scheme for the underactuated bridge crane with uncertainties. The non-strict feedback structure is simplified a strict-feedback one by employing the property of neural networks (NNs). Moreover, the computation complexity is reduced by virtue of the approximation ability of NNs. A scheduling gain and a robust item are added to the control input to improve the robustness of algorithm. The unknown control direction problem is tackled by using the adaptive control technique. Further, a scheduling gain-based adaptive control method is proposed by consisting of the NNs and the adaption law to ensure all signals are bounded. The simulation experiments demonstrate the effectiveness of the proposed scheme.