Path planning of gravel soil paving based on cellular automata and improved A* algorithm

Gravel paving is an important part of the construction of core rockfill dams, and its thickness and flatness are key indicators to characterize the paving quality. However, the existing paving operation path planning highly relies on manual experience and has the disadvantage of strong subjectivity, so it is difficult to effectively control the paving quality. Aiming at the above problems, this research proposes a path planning method for paving gravel soil based on cellular automata and improved A* algorithm. This method optimizes the selection of the global path by improving the trajectory distance measurement function in the A* algorithm and introducing the obstacle avoidance redundant distance. At the same time, it combines the quality evaluation function to optimize the local dynamic path planning method, so as to realize the global-local hybrid path planning of the paving operation. In addition, by establishing a cellular automata model for the paving quality of gravel soil, the real-time monitoring data of the construction machinery is converted into the quality information of each grid of the working surface, so the quality information can be updated in real time. Engineering application shows that the method proposed in this study can control the paving thickness at about 0.27 m on the premise of ensuring construction efficiency, effectively avoiding the occurrence of ultra-thick and ultra-thin phenomena, and the flatness is improved by 22.6% compared with manual work. © 2021, China Water Power Press. All right reserved.