This paper pioneers the use of the symmetrical Hybrid-RAG strategy in the ship welding process domain, addressing the problems of fragmented, unstructured knowledge storage, as well as the limitations of traditional Retrieval-Augmented Generation (RAG), particularly high retrieval noise and low accuracy when answering complex procedural queries. This study proposes an intelligent three-stage symmetric “Generate–Retrieve–Generate” framework for the ship welding process (SWP-Chat), supported by dual retrieval engines: a Neo4j knowledge graph for symbolic reasoning and a vector database for semantic retrieval. Unlike approaches that rely solely on LLM-based process planning, SWP-Chat uses the LLM to generate a logical form, then executes Cypher queries on Neo4j, enabling transparent traceability, precise entity–relation constraints, and deterministic retrieval. Meanwhile, the vector channel supplements unstructured or contextual welding information to enhance semantic coverage. To further improve efficiency, principal component analysis (PCA) was employed for vector dimensionality reduction, reducing average retrieval latency by 31% while retaining more than 95% variance. In addition, an explainable structural–confidence fusion formula integrates evidence from both engines to produce auditable and trustworthy industrial responses. Experimental evaluation demonstrates that the framework achieves an F1 score of 79.35%, greatly surpassing typical RAG systems.

Resistance is a key index of a ship’s hydrodynamic performance, and studying the design of the bulbous bow is an important method to reduce ship resistance. Based on the ship resistance sample data obtained from computational fluid dynamics (CFD) simulation, this study uses a machine learning method to realize the fast prediction of ship resistance corresponding to different bulbous bows. To solve the problem of insufficient accuracy in the single surrogate model, this study proposes a CBR surrogate model that integrates convolutional neural networks with backpropagation and radial basis function models. The coordinates of the control points of the NURBS surface at the bulbous bow are taken as the design variables. Then, a convergence factor is introduced to balance the global and local search abilities of the whale algorithm to improve the convergence speed. The sample space is then iteratively searched using the improved whale algorithm. The results show that the mean absolute error and root mean square error of the CBR model are better than those of the BP and RBF models. The accuracy of the model prediction is significantly improved. The optimized bulbous bow design minimizes the ship resistance, which is reduced by 4.95% compared with the initial ship model. This study provides a reliable and efficient machine learning method for ship resistance prediction.

Focusing on joint-space time-optimal trajectory planning for industrial robots, this study integrates 3-5-3 piecewise polynomial parameterization with an improved Fire Hawk Optimization algorithm (TFHO). Subject to joint position, velocity, and acceleration limits, segment durations are optimized as decision variables. TFHO employs Tent-chaotic initialization to improve the uniformity of initial solutions and a two-phase adaptive Lévy–Gaussian–Cauchy hybrid mutation to balance early global exploration with late local exploitation, mitigating premature convergence and enhancing stability. On benchmark functions, TFHO attains the lowest mean area under the convergence curve (AUC; lower is better). Wilcoxon signed-rank tests show statistically significant improvements over FHO, PSO, GWO, and WOA $(p\le 0.05)$. Ablation studies indicate a pronounced reduction in run-to-run variability: the standard deviation decreases from 0.3157 (FHO) to 0.0023 with TFHO, a 99.27% drop. In an ABB IRB-2600 simulation case, the execution time is shortened from 12.00 s to 9.88 s (−17.66%) while preserving smooth and continuous kinematic profiles (position, velocity, and acceleration), demonstrating practical engineering applicability.

In modern shipbuilding, the structural complexity of ship components and the constrained workspace make robotic grinding prone to collisions. To improve safety and stability, this paper proposes a collision-free posture optimization method for ship-component operations. First, forward and inverse kinematic models are established, and postures along the path are organized into a directed graph. Feasible postures are then identified under joint-limit and singularity constraints. Directed bounding boxes and the GJK collision detection algorithm are applied to construct a collision-free posture set. An improved A* algorithm is then introduced. It incorporates a multi-source heuristic based on joint-space geometric distance and a safety-distance penalty to compute an optimal posture sequence with minimal joint deviation. This design promotes smooth transitions between consecutive postures. Simulation results show that the proposed method avoids robot–workpiece interference in constrained environments and improves obstacle avoidance and motion smoothness. Compared with the standard A* algorithm, the proposed approach reduces search time by 15.8% and increases the minimum safety distance by nearly fivefold. Compared with a non-optimized posture sequence, cumulative joint variation is reduced by up to 92.5%. The joint amplitude range decreases by an average of 41.2%, and the standard deviation of joint fluctuations decreases by 37.8%. The proposed method provides a generalizable solution for robotic measurement, assembly, and machining in complex and confined environments.