Scholarlink

<p>Autonomous Underwater Vehicles (AUVs) face persistent challenges in localization compared to their counterparts on the ground due to limitations with methods like Global Positioning System (GPS). We propose a novel system for localization, <i>Pisces</i>, that leverages the Angle of Arrival (AoA) and Received Signal Strength Ratio (RSSR) of robot-mounted blue LED signals. This method provides a spectrally efficient training-free solution for estimating 3D underwater positions. The system remains effective despite high water turbidity with a relatively low impact on marine life compared to similar acoustic methods. Pisces is less complex, computationally efficient, and uses less power than camera-based solutions. Pisces enables robust relative localization, especially in swarms of robots with the potential for additional applications like docking. We demonstrate high localization accuracy with a Mean Absolute Error (MAE) of 0.031 m at 0.32 m separation and 0.16 m MAE at 1 m separation. Moreover, it achieved this with minimal power consumption, utilizing only 11 mA of transmitter LED current and performing 3D localization within 10 ms for distances up to 3 m.</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

Low power, non-intrusive 3D localization for underwater mobile robots

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年5月23日

查看全文
  • Suryansh Sharma,
  • Daniel Van Passen,
  • R. Venkatesha Prasad,
  • Kaushik Chowdhury

摘要

Autonomous Underwater Vehicles (AUVs) face persistent challenges in localization compared to their counterparts on the ground due to limitations with methods like Global Positioning System (GPS). We propose a novel system for localization, Pisces, that leverages the Angle of Arrival (AoA) and Received Signal Strength Ratio (RSSR) of robot-mounted blue LED signals. This method provides a spectrally efficient training-free solution for estimating 3D underwater positions. The system remains effective despite high water turbidity with a relatively low impact on marine life compared to similar acoustic methods. Pisces is less complex, computationally efficient, and uses less power than camera-based solutions. Pisces enables robust relative localization, especially in swarms of robots with the potential for additional applications like docking. We demonstrate high localization accuracy with a Mean Absolute Error (MAE) of 0.031 m at 0.32 m separation and 0.16 m MAE at 1 m separation. Moreover, it achieved this with minimal power consumption, utilizing only 11 mA of transmitter LED current and performing 3D localization within 10 ms for distances up to 3 m.