Permanent magnet wall-climbing robots are widely used in the maintenance of tanks in the petrochemical industry and the overhaul of large-pressure pipelines in the hydropower industry. One of the difficulties is to achieve the safe and reliable adsorption of wall-climbing robots. Based on the Halbach array, a double-layer superposition magnetic circuit magnetization method is designed in this paper. Under the same constraints, the adsorption force of the permanent magnetic chuck is increased by at least $8\%$ compared with the traditional magnetic circuit design method. Under the working air gap of 1∼9 mm, the average magnetic energy utilization rate is increased by at least $16.46\%$. This approach not only improves the magnetic energy utilization of the permanent magnetic chuck but also improves the adsorption safety of the wall-climbing robot.

Climbing manufacturing robots can create a revolutionary manufacturing paradigm for large and complex components, while the motion control of climbing manipulation-oriented robots (CMo-Rs) is still challenging considering anti-slippage problems. In this study, a CMo-R with full-scenery climbing capability and redundant load-bearing mobility is designed based on magnetic adsorption. A four-wheel kinematic model considering the slipping phenomenon is established. An adaptive kinematic control algorithm based on slip estimation using Lyapunov theory is designed for uncertain inclined planes. For comparison, the traditional PID-based algorithm without slip consideration is implemented as well. Numeric simulations are conducted to tackle the trajectory tracking problems for both circular and linear trajectories on the horizontal plane (HP), 50° inclined plane (50° IP), 60° inclined plane (60° IP), and vertical plane (VP). The results prove that our approach achieves better tracking accuracy. It demonstrated applicability in various climbing scenarios with uncertain inclined planes. The results of experiments also validate the feasibility, applicability, and stability of the proposed approach.

Ship classification, as an important problem in the field of computer vision, has been the focus of research for various algorithms over the past few decades. In particular, convolutional neural networks (CNNs) have become one of the most popular models for ship classification tasks, especially using deep learning methods. Currently, several classical methods have used single-scale features to tackle ship classification, without paying much attention to the impact of multiscale features. Therefore, this paper proposes a multiscale feature fusion ship classification method based on evidence theory. In this method, multiple scales of features were utilized to fuse the feature maps of three different sizes (40 × 40 × 256, 20 × 20 × 512, and 10 × 10 × 1024), which were used to perform ship classification tasks separately. Finally, the multiscales-based classification results were treated as pieces of evidence and fused at the decision level using evidence theory to obtain the final classification result. Experimental results demonstrate that, compared to classical classification networks, this method can effectively improve classification accuracy.