Search

In order to improve the detection accuracy of the navigation line by the unmanned automatic marking vehicle (UAMV) in the complex construction environment. Solve the problem of unqualified road markings drawn by the UAMV due to inaccurate detection during construction. A navigation line detection algorithm based on and improved YOLOv4 and improved Haar-like feature named YOLOv4-HR is proposed in this paper. Firstly, an image enhancement algorithm based on improved Haar-like features is proposed. It is used to enhance the images of the training set, make the images contain more semantic information, which improves the generalization ability of the network; Secondly, a multi-scale feature extraction network is added to the YOLOv4 network, which made model has a stronger learning ability for details and improves the accuracy of detection. Finally, a verification experiment is carried out on the self-built data set. The experimental results show that, compared with the original YOLOv4 network, the method proposed in this paper improves the AP value by 14.3% and the recall by 11.89%. The influence of factors such as the environment on the detection effect of the navigation line is reduced, and the effect of the navigation line detection in the visual navigation of the UAMV is effectively improved.

For the problem of simplifying pattern-based modeling procedures, an improved pattern-based modeling method is put forward via pattern classification for a class of complex processes. It is a pure data-driven modeling method using statistical attributes of the processes. At the beginning of the paper, a method of system dynamics description based on pattern moving is introduced. Then, an improved method of pattern-moving-based prediction modeling is put forward, and it simplifies the pattern-moving-based modeling method by integrating an initial model and a classification mapping. It consists of two parts: system pattern construction and pattern classification. And a constructive classification neural network (CCNN) is designed to describe system dynamics by classifying the system pattern, and its generalization is discussed. Finally, simulations using data of an actual production process demonstrate the feasibility of the proposed modeling method, and the effectiveness of the CCNN is verified using comparison experiments.