Improvement of A* Algorithm and Its Path Smoothing Optimization in Nuclear Facility Radioactive Environment

In the path planning of nuclear facilities in radioactive environments, the traditional A* algorithm has been widely used due to its good versatility and heuristic search. However, the algorithm still has two significant shortcomings. The search efficiency is low in radiation fields, and the path it generates is in a broken line shape, which is not conducive to the actual walking of staff and makes it difficult to meet the needs of engineering applications. In response to the above problems, the traditional A* algorithm was optimized by algorithm structure and path morphology. Firstly, in the cost function design, the heuristic function of the A* algorithm was defined as the estimated dose from the current node to the end point. The actual cost was the actual cumulative dose from the start point to the current node. Because the actual cumulative dose G(n) of the A* algorithm usually grew rapidly with the search process, which might mask the role of the heuristic function H(n), resulting in the loss of directionality and decreased efficiency of the search. To address this problem, a dynamic weighting scheme based on the distance between the start point, the end point, and the current node was proposed to balance the influence weights of G(n) and H(n). In this way, the cumulative dose on the path was reduced while ensuring the search efficiency, and the balance between search efficiency and cumulative dose was achieved. Secondly, the initial path was smoothed by a method based on the principle of uniform subdivision. In this paper, the dose value at the edge of the grid was supplemented by the inverse distance weighted interpolation method to more accurately evaluate the cumulative dose changes before and after path smoothing. To verify the effectiveness of the proposed algorithm, this paper built a model based on the actual radiation source library scene based on the Geant4 simulation platform, and calculates the dose rate distribution in the radiation field. The results show that the modified A* algorithm proposed in this paper is superior to the traditional A* algorithm in terms of cumulative dose and search efficiency. The total cumulative dose of the smoothed and optimized path is reduced by 12.00% compared with the unsmoothed path, and the path length is shortened by 4.39%. This shows that the improved path is more in line with the actual walking needs of staff while reducing radiation exposure. In summary, the path planning and optimization method proposed in this paper improves the calculation efficiency of the search process in a complex radiation field. At the same time, it effectively reduces the cumulative radiation dose received by the staff, and the output path is more in line with the feasibility and safety of actual operations. It has strong practical engineering application value and promotion potential.