Underwater Optical Camera Communication (UOCC) has emerged as a promising paradigm for short-range, high-bandwidth, and secure data exchange in autonomous underwater vehicles (AUVs). UOCC performance strongly depends on exposure time and International Standards Organization (ISO) sensitivity—two parameters that govern photon capture, contrast, and bit detection fidelity. However, optical propagation in aquatic environments is highly susceptible to turbidity, scattering, and illumination variability, which severely degrade image clarity and signal-to-noise ratio (SNR). Conventional systems with fixed imaging settings cannot adapt to time-varying conditions, limiting communication reliability. While validating the feasibility of deep learning for exposure prediction, this baseline lacked environmental awareness and generalization to dynamic scenarios. To overcome these limitations, we introduce a Real-to-Sim-to-Deployment framework that couples a physically calibrated emulation platform with a Hybrid CNN-MLP Model (HCMM). By fusing optical images, environmental states, and camera configurations, the HCMM achieves substantially improved parameter prediction accuracy, reducing RMSE to 0.23–0.33. When deployed on embedded hardware, it enables real-time adaptive reconfiguration and delivers up to 8.5 dB SNR gain, surpassing both static-parameter systems and the prior CNN baseline. These results demonstrate that environment-aware multimodal learning, supported by reproducible optical channel emulation, provides a scalable and robust solution for practical UOCC deployment in positioning, inspection, and laser-based underwater communication.

In order to improve the current problem of poor usability, comfort, and low user motivation in traditional wheelchairs. The physiological status, psychological emotions, and lifestyle of elderly people with mild to moderate disability and caregivers was analyzed through the method of user observation, user interviews and literature research, so as to clarify the multi-level and diverse needs of users and the design positioning. Three usage modes of sitting and driving, reclining and lying flat, and the intelligent control mode of assistant path planning, intelligent obstacle avoidance has been conceived, and the main dimensions and structure have been optimized according to the design positioning. In conclusion, from the perspective of the needs of users and caregivers, an intelligent wheelchair solution that combines aesthetic appearance and ease using functions, nursing burden reducing, and positive emotions enhancing of users’ can be obtained, which providing a certain reference for the design of intelligent products for disability assistance.

Underwater optical camera communication (UOCC) relies on image-based signal reception, where the quality of recorded stripes directly determines the achievable optical signal-to-noise ratio (SNR). However, fixed camera parameters—such as exposure level and International Standards Organization (ISO) sensitivity—are often inadequate under dynamically varying aquatic conditions including turbidity, flow velocity, and ambient illumination. To overcome these limitations, we propose a multimodal deep model that fuses visual cues with environmental context to predict scene-optimal camera parameters at capture time. A ResNet50 backbone extracts semantic representations from raw stripe images, while environmental factors—including turbidity, flow speed, ambient illumination, and LED power—are jointly encoded through a parallel embedding architecture. These modalities are fused within a regression network to infer settings that maximize imaging clarity and stability. The experimental results demonstrate that the proposed model achieves robust accuracy. Crucially, the capture-time parameter selection improves stripe visibility and delivers an average ∼3dB gain in optical SNR across diverse conditions. This capture-time optimization sustains a higher and more stable SNR band than both the original fixed settings and a representative learning-based post-processing baseline (DnCNN with horizontal-banding suppression). Beyond accuracy, the approach is computationally light and portable: context-aware parameter prediction at capture time eliminates per-frame processing, providing a practical route to real-time, resource-constrained UOCC with enhanced image quality and robustness.

Existing indoor positioning methods in visible light communication systems typically rely on large databases, powerful signal processing units, and additional sensors such as gyroscopes, which limit their applicability in efficiency-sensitive scenarios. To address this issue and achieve an optimal balance between positioning accuracy, inference time, and model size, we propose uniform and power-of-two quantization on a multi-head ResNet50 (UPU-MH-ResNet50) algorithm in this paper. The proposed algorithm does not require additional sensors and can accurately estimate the three-dimensional coordinates of the receiver, including X and Y coordinates and rotation angle. It effectively mitigates the impact of random angle changes on positioning accuracy, significantly enhancing the system’s robustness. Additionally, the algorithm is further optimized through model quantization to reduce computational costs. We conducted validation on a self-built experimental testbed in a laboratory environment with dimensions of 2.6m×2.6m×2.2m and a receiver rotation angle range of ±30 ^∘ . Within this tilt range, 90% of the 3D positioning errors are controlled within 2 cm. Moreover, the proposed 4-bit UPU-MH-ResNet50 achieves a 7.7× reduction in model size and a 2.7× speedup in inference, where the acceleration and efficiency results are measured from the GPU-based implementation using real captured data collected on the developed testbed, indicating high energy efficiency.

During the start stage of ship welding, obtaining the three-dimensional coordinates of welding target points is affected by confined installation space, surface reflection, and deployment constraints. This paper proposes a low-complexity point-wise three-dimensional localization method based on two-dimensional visual planar guidance and one-dimensional point-laser distance constraints. A direct computation model of the laser incident point in the robot base coordinate system is established from the tool center point pose, the extrinsic parameters of the point-laser module, and real-time ranging data, enabling target-point coordinate estimation without dense three-dimensional reconstruction. A dual-stage stabilization strategy is introduced by combining ranging-level filtering, spatial coordinate smoothing, and outlier suppression. Image error-based visual closed-loop alignment is further used as a pre-measurement step to ensure that the point laser acts on the target region. Experimental results show that, after workplane-level extrinsic correction, independent validation points achieve a mean three-dimensional Euclidean error of 1.54 mm with a standard deviation of 0.28 mm. The average planar error in closed-loop alignment experiments is 1.124 mm. Passive binocular depth measurement on the current platform still yields an RMSE of 6.16 mm after linear correction. A simulated fillet-weld task verifies the feasibility of the complete perception-to-execution workflow. The proposed method provides a low-complexity coordinate acquisition route for discrete welding target points before arc ignition.