There is no unified planning standard for unstructured roads,and the morphological structures of these roads are complex and varied.It is important to maintain a balance between accuracy and speed for unstructured roa...There is no unified planning standard for unstructured roads,and the morphological structures of these roads are complex and varied.It is important to maintain a balance between accuracy and speed for unstructured road extraction models.Unstructured road extraction algorithms based on deep learning have problems such as high model complexity,high computational cost,and the inability to adapt to current edge computing devices.Therefore,it is best to use lightweight network models.Considering the need for lightweight models and the characteristics of unstructured roads with different pattern shapes,such as blocks and strips,a TMB(Triple Multi-Block)feature extraction module is proposed,and the overall structure of the TMBNet network is described.The TMB module was compared with SS-nbt,Non-bottleneck-1D,and other modules via experiments.The feasibility and effectiveness of the TMB module design were proven through experiments and visualizations.The comparison experiment,using multiple convolution kernel categories,proved that the TMB module can improve the segmentation accuracy of the network.The comparison with different semantic segmentation networks demonstrates that the TMBNet network has advantages in terms of unstructured road extraction.展开更多
The complex terrain environment in the hilly land directly affects the operational reliability of agricultural robots.In order to study the impact of road irregularity on walking chassis vibration,the 3CYLZ-750 remote...The complex terrain environment in the hilly land directly affects the operational reliability of agricultural robots.In order to study the impact of road irregularity on walking chassis vibration,the 3CYLZ-750 remote-controlled weeding machine which is applied to orchards was taken as the object of study,and the rear roller was selected as the object of observation to reveal the rules under which the vibration of the track chassis changes as there is a sudden change in road surface elevation.A column-type test-to-pass method based on unit excitation was proposed in this study.The excitation behavior and action process were analyzed by category.A critical acceleration prediction model was built and verified by virtual simulation and hard road surface excitation testing.The results showed that at the forward velocity of 0-2.5 km/h and exciter height of 20-100 mm,the vertical vibration acceleration of the target roller was significantly affected by Track Contact Point Centrifugal Acceleration(TCPCA).As TCPCA increased,the change rate of vertical vibration acceleration decreased,reaching a minimum of[−13.8,28.8];as TCPCA decreased,the vertical vibration acceleration tended to increase positively at a maximum variation range of[−13.3,42.2].The measured and simulated macroscopic change rules were consistent with the theoretical analysis,further verifying the correctness of variable extraction,and providing a research basis for the accurate modification and improvement of the model.The research conclusions can lay a theoretical foundation for analyzing the walking reliability of the track chassis,and provide a design basis and technical support for the development of a tracked agricultural robot chassis for the hilly land in the future.展开更多
基金Supported by National Natural Science Foundation of China(Grant Nos.62261160575,61991414,61973036)Technical Field Foundation of the National Defense Science and Technology 173 Program of China(Grant Nos.20220601053,20220601030)。
文摘There is no unified planning standard for unstructured roads,and the morphological structures of these roads are complex and varied.It is important to maintain a balance between accuracy and speed for unstructured road extraction models.Unstructured road extraction algorithms based on deep learning have problems such as high model complexity,high computational cost,and the inability to adapt to current edge computing devices.Therefore,it is best to use lightweight network models.Considering the need for lightweight models and the characteristics of unstructured roads with different pattern shapes,such as blocks and strips,a TMB(Triple Multi-Block)feature extraction module is proposed,and the overall structure of the TMBNet network is described.The TMB module was compared with SS-nbt,Non-bottleneck-1D,and other modules via experiments.The feasibility and effectiveness of the TMB module design were proven through experiments and visualizations.The comparison experiment,using multiple convolution kernel categories,proved that the TMB module can improve the segmentation accuracy of the network.The comparison with different semantic segmentation networks demonstrates that the TMBNet network has advantages in terms of unstructured road extraction.
基金This study was financially supported by the Guangdong Provincial Key Project R&D Program(Grant No.2019B090922001)the Guangdong Provincial Postdoctoral Research Center Construction Project(Grant No.[2020]No.122).
文摘The complex terrain environment in the hilly land directly affects the operational reliability of agricultural robots.In order to study the impact of road irregularity on walking chassis vibration,the 3CYLZ-750 remote-controlled weeding machine which is applied to orchards was taken as the object of study,and the rear roller was selected as the object of observation to reveal the rules under which the vibration of the track chassis changes as there is a sudden change in road surface elevation.A column-type test-to-pass method based on unit excitation was proposed in this study.The excitation behavior and action process were analyzed by category.A critical acceleration prediction model was built and verified by virtual simulation and hard road surface excitation testing.The results showed that at the forward velocity of 0-2.5 km/h and exciter height of 20-100 mm,the vertical vibration acceleration of the target roller was significantly affected by Track Contact Point Centrifugal Acceleration(TCPCA).As TCPCA increased,the change rate of vertical vibration acceleration decreased,reaching a minimum of[−13.8,28.8];as TCPCA decreased,the vertical vibration acceleration tended to increase positively at a maximum variation range of[−13.3,42.2].The measured and simulated macroscopic change rules were consistent with the theoretical analysis,further verifying the correctness of variable extraction,and providing a research basis for the accurate modification and improvement of the model.The research conclusions can lay a theoretical foundation for analyzing the walking reliability of the track chassis,and provide a design basis and technical support for the development of a tracked agricultural robot chassis for the hilly land in the future.
