Object detectors based on CNN(Convolutional Neural Network) have been widely used in industrial production. In the production and transportation process of continuous casting slab, it is necessary to dynamically track the slab logistics and orderly process the tracking information when the slab is on or off the production line. In this paper, the continuous casting slab data set on the continuous casting slab transfer roller table is firstly established, which includes photos of slabs, crown blocks and crown blocks holding slabs under various lighting conditions on the continuous casting roller table. Then an improved YOLOv5 frame called VAAB-YOLOv5(YOLOv5 with VCIoU, the Attention Network, ACON-C Activate Function and BiFPN) is proposed for the detection of continuous casting slabs. VAAB-YOLOv5 uses a new loss called vertex center IoU (VCIoU) loss, which introduces the feature of distance between vertex and center point. VCIoU loss improves the detection accuracy without increasing the calculation cost. In addition, this framework also improves the detection accuracy by improving the network structure and activation function, including adding the attention mechanism, introducing the ACON-C activation function, and improving FPN. At last, the experimental results show that the mAP@50:95 index of the proposed method on the self-established continuous casting slab dataset is 87.75%, which is 1.58% higher than the YOLOv5s algorithm. The VAAB-YOLOv5 improves the detection accuracy while meeting the speed requirements of slab detection on the continuous casting site.

This paper proposes a fixed-time backstepping control strategy based on adaptive super-twisting disturbance observers (ASTDOs) for a class of non-integral cascade high-order uncertain nonlinear systems. First, the ASTDOs are designed to estimate the system disturbances in fixed time, and the adaptive method relaxes the assumption about the system disturbances, which is that the upper bound of the first and second derivatives of the disturbance is unknown. Then we present a backstepping control strategy via an improved distributed fast terminal sliding mode, which not only guarantees that the sliding mode surface of each step is independent of one another, but also makes the system states converge in fixed time. Furthermore, to avoid the ‘differential explosion’ problem and the singularity problem, the dynamics surface control method is constituted. Theoretical analysis shows that the closed-loop system is semi-globally uniformly ultimately fixed-time bounded. Finally, simulation results show the validity of the proposed control strategy.