Development of Soft Sensor for Sensorless Automatic Gantry Crane Using RBF Neural Networks

To attain a good control performance of automatic gantry crane system, sensors are indispensable instrument for feedback signals. However, sensing the payload motion of a real gantry crane, particularly swing motion, is not easy and sometimes costly. Therefore, a sensorless automatic gantry crane system is developed and proposed in this paper. A soft sensor based on artificial neural network is introduced to eliminate the real sensor. Instead, a sensor measuring armature current of DC motor driving the cart is used to provide dynamic information for the soft sensor. A simulation study using dynamic model of lab-scale automatic gantry crane is carried out to evaluate the effectiveness of the proposed soft sensor. The results show that the soft sensor can estimate effectively the unmeasured state. Moreover, the proposed method has robustness to deal with parameter variations