In the depths of the ocean, where sunlight barely reaches and pressures are immense, a new frontier in mineral exploration is unfolding. Researchers at the Institute of Deep-Sea Science and Engineering, Chinese Academy of Sciences, led by Zhenzhuo Wei, are pioneering a data-driven cooperative localization algorithm that promises to revolutionize how we explore and extract valuable resources from the seabed. This breakthrough, published in the journal ‘Remote Sensing’ (translated to English), could significantly enhance the efficiency and accuracy of deep-sea mining operations, with profound implications for the energy sector.
The deep-sea landing vehicle (DSLV) swarm exploration system is at the heart of this innovation. Unlike traditional underwater vehicles that float in the water, DSLVs are designed to operate on soft seabeds, offering low noise, high stability, and long operational durations. These characteristics are crucial for extensive data collection and the exploration of seabed mineral resources. However, the complex underwater environment poses significant challenges, particularly when it comes to accurate positioning. Doppler Velocity Log (DVL) measurements, essential for navigation, can be seriously interrupted, leading to unstable localization performance. Additionally, track slippage on soft seabeds further degrades the accuracy of the cooperative localization system.
To address these issues, Wei and his team have developed a novel algorithm that combines a velocity prediction model with an Unscented Kalman Filter (UKF) to improve positioning accuracy. The velocity prediction model, constructed using multi-output least squares support vector regression (MLSSVR), is optimized using a genetic algorithm (GA) to enhance its robustness. “The genetic algorithm helps in fine-tuning the hyperparameters of the MLSSVR model, making it more accurate and reliable,” explains Wei. This model predicts the motion velocity of the DSLV on soft seabed sediments, even during DVL measurement interruptions.
The outputs of the MLSSVR are then fed into a DSLV position estimation framework based on the UKF. This integration allows the system to compensate for cooperative localization errors caused by track slippage, significantly improving the accuracy and reliability of the DSLV cooperative localization system. “Our simulations have shown that this method can effectively mitigate the errors caused by track slippage, ensuring more precise and reliable positioning,” Wei adds.
The implications of this research for the energy sector are vast. As the demand for critical minerals continues to rise, driven by the transition to renewable energy and the electrification of industries, the ability to accurately and efficiently explore deep-sea resources becomes increasingly important. The proposed algorithm could lead to more efficient mining operations, reducing costs and environmental impacts. Moreover, the enhanced accuracy of DSLV positioning could open up new areas for exploration, potentially uncovering previously inaccessible mineral deposits.
Looking ahead, this research sets the stage for future developments in deep-sea mining technology. The integration of machine learning and advanced sensor technologies could pave the way for even more robust and autonomous underwater exploration systems. As Wei and his team continue to refine their algorithm and conduct real-world experiments, the future of deep-sea mining looks brighter and more precise than ever before. The potential for this technology to shape the energy sector is immense, offering a glimpse into a future where the depths of the ocean are not just a mystery but a valuable resource for humanity.
