Scholarlink

<p>When accidents occur during the construction of power cables, blindly troubleshooting without accurately identifying the locations of underground cable locations will lead to increased workload. As such, a new method is proposed to locate underground cable seeker by combining visual and inertial sensors, generating precise motion trajectories. An adaptive gray threshold strategy is employed to address the poor quality of feature points extracted in conventional visual navigation. In addition, a loosely coupled algorithm based on dual Kalman filters is employed to overcome the issues of excessive filtering cycles and low data utilization in conventional filtering algorithms. This method uses inertial and visual sensors to estimate the cable motion trajectory separately and fuses the two trajectories together through filtering, effectively eliminating trajectory errors. In the simulation, this method can guide the carrier more than 500s, during which the carrier can travel a distance of 125&#xa0;m. The final distance velocity error of this method is 42.64&#xa0;cm and 2.55×10<sup>−3&#xa0;</sup>m/s, while that generated by the conventional method is 156.47&#xa0;cm and 7.8×10<sup>−3</sup>&#xa0;m/s. This method has significantly higher accuracy compared to conventional navigation algorithms.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

An Enhanced Positioning Method for Underground Cable Seeker Based on Vision-Inertial Sensors

  • 期刊论文

  • 发表日期：

    2025年3月28日

查看全文
  • Zongsheng Xu,
  • Yan Feng,
  • Yufang Lin,
  • Linfeng Dan,
  • Xinyu Pei,
  • Xiongfeng Huang

摘要

When accidents occur during the construction of power cables, blindly troubleshooting without accurately identifying the locations of underground cable locations will lead to increased workload. As such, a new method is proposed to locate underground cable seeker by combining visual and inertial sensors, generating precise motion trajectories. An adaptive gray threshold strategy is employed to address the poor quality of feature points extracted in conventional visual navigation. In addition, a loosely coupled algorithm based on dual Kalman filters is employed to overcome the issues of excessive filtering cycles and low data utilization in conventional filtering algorithms. This method uses inertial and visual sensors to estimate the cable motion trajectory separately and fuses the two trajectories together through filtering, effectively eliminating trajectory errors. In the simulation, this method can guide the carrier more than 500s, during which the carrier can travel a distance of 125 m. The final distance velocity error of this method is 42.64 cm and 2.55×10−3 m/s, while that generated by the conventional method is 156.47 cm and 7.8×10−3 m/s. This method has significantly higher accuracy compared to conventional navigation algorithms.