In the field of mobile robot path planning, traditional genetic algorithms face issues such as slow convergence, lack of dynamic adaptability, and uncertain mutation directionality. To address these issues, a dichotomy-based multi-step method was used during the population initialization phase to enhance both the quality and diversity of the population. The selection strategy was improved for tournament selection, which reduces the monopolistic dominance of exceptional individuals and preserves population diversity through multiple rounds of grouping competitions. The crossover strategy adopted an adaptive crossover point number, with parents chosen based on a dynamic threshold. The mutation strategy combines a two-layer encoding approach: the first layer uses random mutations to enhance diversity, while the second layer performs goal-oriented mutations, with the selection probability of each layer dynamically adjusted during iterations. Bezier curve optimization of the optimal path was conducted. Experimental results show that the improved genetic algorithm demonstrates significant advantages in terms of path length, smoothness, iteration speed, and computational efficiency.

To address the frequent cleaning requirements of casting molds in bridge tower construction, a wall-climbing cleaning robot based on negative pressure adsorption is designed to safely and efficiently replace manual labor for cleaning tasks. The primary focus of this paper is the establishment of a theoretical model for negative pressure adsorption, along with an analysis of potential adhesion hazards. Initially, the robot’s chassis was designed, followed by the development of a theoretical model for the rotational-flow suction unit that incorporates two critical parameters: the number of blades and their thickness. This model was validated through computational fluid dynamics (CFD) and experimental methods. The findings indicate that, with fewer blades, an increase in blade quantity significantly improves the distribution of nonlinear velocity in the z-plane, resulting in a substantial enhancement of suction force up to a certain limit. As the number of blades increases, the thickness of the blades primarily influences the volume of air within the rotating domain, thereby affecting the suction force; thinner blades are preferable. Moreover, this study reveals that square suction units provide greater suction force compared to circular ones, attributable to their superior negative pressure effect and larger adsorption area. The most critical adhesion risk identified is leakage at the edges of the suction unit.

Path planning for autonomous mobile robots in dynamic environments is a critical challenge, particularly when robots must navigate both known and unknown obstacles in real-time. Traditional methods like ant colony optimization (ACO) have shown promise in global path planning but often suffer from slow convergence and the risk of getting trapped in local optima. Similarly, while the dynamic window approach (DWA) excels in real-time obstacle avoidance, it exhibits delayed response and higher failure rates in scenarios with fast-moving obstacles. To address these challenges, this study proposes a hybrid path planning framework, IACO-QDWA, which integrates an improved ACO (IACO) for global path optimization with a Q-learning-enhanced DWA (QDWA) for dynamic obstacle avoidance. IACO improves global path search through goal-oriented initialization and adaptive pheromone updates, while QDWA enhances local decision-making by optimizing the state-action space. The experiments demonstrate that IACO-QDWA achieves a reasonable balance between path quality, obstacle avoidance success rate, and real-time performance. It exhibits stable obstacle avoidance behavior and reliable navigation capabilities in highly dynamic environments.

It is common to find severe perspective distortion in a pipeline’s image in medium-diameter pipeline defect detection by the panoramic image unwrapping method, resulting in low-quality image unwrapping and stitching, which is caused by the camera’s optical axis being completely deviated from the pipeline’s center. To solve this problem, a novel correction method for reducing perspective distortion in pipeline images was proposed for pipeline defect detection. Firstly, the method enhances the edges of unevenly illuminated regions within a pipeline to facilitate image segmentation and identify key points necessary for correcting perspective distortion. Then, a six-feature-point extraction method was proposed for a circle target to establish the projection relationship between the extracted feature and mapped points on the reference circle. Finally, a perspective matrix was constructed to complete the perspective transformation correction of the distorted images. The results show that the average correction rate and the average relative error of the proposed correction method can reach 90.85% and 1.31%, respectively. The study innovatively used the enhancement of uneven illumination to find distorted edge information. It proposed an extraction method using a reference circle and six key feature points to build a mapping model. It can provide a novel method which can be used to obtain a superior image for pipeline detection and lay a solid foundation for subsequent high-quality pipeline image stitching.