Research on Improved Self-learning Method of Conveying Volume Prediction for Distribution Robot Based on Data-driven

The improved self-learning method of conveying volume prediction for the distribution robot driven by filling volume data is studied to improve the correction effect of self-learning for the predictive value. Through qualitative and quantitative analysis of the influence of the filling volume on the conveying volume, it is found that the reduction of the concrete filling volume in the hopper will lead to the reduction of the screw conveying volume of the distribution robot. The relation model between filling volume and screw conveying volume is established by data fitting method, and the self-learning method for predicting conveying volume is improved accordingly. The experimental results show that the improved self-learning method can not only adapt to the change of distribution conditions to quickly modify the predictive value of the screw, but also provide a more accurate control target value for the weight control system of the distribution robot. Then, the initial distribution deviation caused by the poor adaptation of the target value is improved.