Uncertainty quantification of vibro-acoustic coupling problems for robotic manta ray models based on deep learning

Ocean Engineering, Journal Year: 2024, Volume and Issue: 299, P. 117388 - 117388

Published: March 6, 2024

Language: Английский

---

Research on the Identification and Classification of Marine Debris Based on Improved YOLOv8

Journal of Marine Science and Engineering, Journal Year: 2024, Volume and Issue: 12(10), P. 1748 - 1748

Published: Oct. 3, 2024

Autonomous underwater vehicles equipped with target recognition algorithms are a primary means of removing marine debris. However, due to poor visibility, light scattering by suspended particles, and the coexistence organisms debris, current methods have problems such as classification effects, slow speed, weak generalization ability. In response these problems, this article proposes debris identification algorithm based on improved YOLOv8. The incorporates CloFormer module, context-aware local enhancement mechanism, into backbone network, fully utilizing shared weights. Consequently, it enhances high- low-frequency feature extraction from images. proposed C2f-spatial channel reconstruction (C2f-SCConv) module combines SCConv neck C2f reduce spatial redundancy in standard convolutions enhance representation. WIoU v3 is employed bounding box regression loss function, effectively managing low- high-quality samples improve overall model performance. experimental results TrashCan-Instance dataset indicate that compared classical YOLOv8, [email protected] F1 scores increased 5.7% 6%, respectively. Meanwhile, TrashCan-Material dataset, also improve, 5.5% 5%, Additionally, size has been reduced 12.9%. These research conducive maintaining life safety ecosystem stability.

Language: Английский

---

Three-Dimensional Path Following Control of Underactuated AUV Based on Nonlinear Disturbance Observer and Adaptive Line-of-Sight Guidance

IEEE Access, Journal Year: 2024, Volume and Issue: 12, P. 83911 - 83924

Published: Jan. 1, 2024

Language: Английский

---

Three-Dimensional Path Following Control for Underactuated AUV Based on Ocean Current Observer

Drones, Journal Year: 2024, Volume and Issue: 8(11), P. 672 - 672

Published: Nov. 13, 2024

In the marine environment, motion characteristics of Autonomous Underwater Vehicles (AUVs) are influenced by unknown factors such as time-varying ocean currents, thereby amplifying complexity involved in design path-following controllers. this study, a backstepping sliding mode control method based on current observer and nonlinear disturbance (NDO) has been developed, addressing 3D issue for AUVs operating environment. Accounting uncertainties like variable research establishes AUV’s kinematics dynamics models formulates tracking error within Frenet–Serret coordinate system. The kinematic controller is designed through line-of-sight method, dynamic developed using integral method. Furthermore, an real-time estimation velocities, mitigating effects currents navigational performance. Theoretical analysis confirms system’s asymptotic stability, while numerical simulation attests to proposed method’s efficacy robustness path following.

Language: Английский

---

The effect of flame treatment on the compressive strength of composite sandwich panels made of pure epoxy resin H‐shaped structures and hollow glass bead/epoxy resin composite material

Polymer Engineering and Science, Journal Year: 2024, Volume and Issue: 64(6), P. 2736 - 2746

Published: March 25, 2024

Abstract At present, syntactic foam composites (HGB/EP) made of epoxy resin (EP) and hollow glass beads (HGBs) have received extensive attention in the research field lightweight materials because their high compression strength‐to‐density ratio. However, deep‐sea environment has put forward higher requirements for compressive properties materials. The traditional sandwich structure with HGB/EP as core gradually failed to meet mechanical equipment deeper ocean. This paper proposes a comprising pure EP H‐shaped structural component rigid skeleton an composite filling component. was subjected flame treatment using butane jet at different distances. resistance resulting tested, results showed varying degrees improvement strength after treatment. effects on surface adhesion performance were characterized compression‐shear tests, optical microscopy, contact angle infrared spectroscopy, temperature tests. demonstrated that distance 0 cm significantly increased oxygen‐containing functional groups EP, improved wettability surface, adhesive between HGB/EP, enhanced structure. shear test specimens by 33.0%, while 17.5%. Highlights composed material prepared. affects chemical resin. Flame

Language: Английский

---

Hydrodynamic modelling and motion characteristics of flexible swimming limbs for asymmetric swinging underwater soft robots

Applied Ocean Research, Journal Year: 2024, Volume and Issue: 144, P. 103884 - 103884

Published: Jan. 15, 2024

Language: Английский

---

Adaptive maximum correntropy UKF TDOA tracking algorithm based on multi-layer isogradient sound speed profile

Ocean Engineering, Journal Year: 2024, Volume and Issue: 303, P. 117541 - 117541

Published: April 12, 2024

Language: Английский

---

Acoustic camera-based super-resolution reconstruction approach for underwater perception in low-visibility marine environments

Applied Ocean Research, Journal Year: 2024, Volume and Issue: 150, P. 104110 - 104110

Published: July 8, 2024

In low-visibility environments, the underwater perception range of optical cameras is severely restricted, and operations in ocean engineering often rely on sonar. Acoustic are a type forward-looking sonar that have attracted considerable attention because their ability to produce images similar those cameras. However, owing unique imaging mechanism employed by acoustic cameras, resulting suffer from insufficient resolution loss feature details. This issue considerably diminishes precision downstream visual tasks, limiting application this study, we propose deep-learning-based super-resolution reconstruction approach for where process relies only images, without prior assumptions regarding detection scenes. We verified effectiveness proposed method two practical applications: marine debris structure inspection. The experimental results show our can robustly reconstruct high-resolution obtained superior details, which improved vision tasks. aim provide better solutions while exploring

Language: Английский

---

Design of a Multi-Environmentally Adaptable Modular Self-Reconfigurable Robot

IEEE Robotics and Automation Letters, Journal Year: 2024, Volume and Issue: 9(10), P. 8627 - 8634

Published: Aug. 22, 2024

Language: Английский

---

Advanced Power Converters and Learning in Diverse Robotic Innovation: A Review

Energies, Journal Year: 2023, Volume and Issue: 16(20), P. 7156 - 7156

Published: Oct. 19, 2023

This paper provides a comprehensive review of the integration advanced power management systems and learning techniques in field robotics. It identifies critical roles these areas play reshaping capabilities robotic across diverse applications. To begin, it highlights significance efficient usage modern The explains how converters effectively control voltage, manage current shape waveforms, thereby optimizing energy utilization. These ensure that components receive precise voltage levels they require, leading to improved motor performance enabling over behavior. Consequently, this results extended operational times increased design flexibility. Furthermore, explores approaches, emphasizing their substantial impact on perception, decision-making autonomy. discusses application such as reinforcement learning, supervised unsupervised showcasing applications like object recognition, semantic segmentation, sensor fusion anomaly detection. By utilizing methods, robots become more intelligent, adaptable capable autonomous operation various domains. examining interaction between integration, anticipates future where operate with efficiency, adapt tasks drive technological innovation wide range industries.

Language: Английский

---