Complete coverage path planning and performance factor analysis for autonomous bulldozer

Abstract With the development of intelligent machinery, path planning of autonomous bulldozers plays an important role to solve the problems of efficiency in future construction sites. How to plan the path that meets the construction requirements is the key problem. This study investigates the standardized construction workmanship and moving rules of the bulldozer to add rules and path selection strategies to the algorithm. This study establishes the path planning model of bulldozer's leveling, considering the limitations of the traditional bioinspired neural network algorithm, an improved complete coverage path planning method is proposed. An integrated framework is proposed for an automatic bulldozer with a sensor system for a field experiment to verify the feasibility of the method. According to the comparative statistical simulation results, the effectiveness of the method is verified, and the effects of different factors on the path planning results are analyzed.