Ant colony optimization-based method for energy-efficient cutting trajectory planning in axial robotic roadheader

Abstract

The traditional cutting trajectory of the axial robotic roadheader (spiral or reciprocating) is too simple to meet the requirements of adaptive planning. We proposed a method for planning the cutting trajectory of an axial robot roadheader based on ant colony optimization, which enables the machine to automatically adapt to the rock characteristics of the tunnel face, solves the problems of high energy consumption and low cutting efficiency in existing methods. This method uses machine vision to capture the fractures in the tunnel face, construct a grid map of the fractures, and calculate the fracture parameters within each grid, including the numbers, length, width, and density, through the connected domain method. Among them, the fracture parameters is used as input and the geological strength index as output. The BPNN is established to predict GSI of the grid environment. Based on this, an improved ant colony optimization for full coverage path planning is introduced, the cutting trajectory can be planned adaptively according to the geological conditions. The method is validated on a axial robotic roadheader, demonstrating its ability to adaptively plan cutting paths based on the geological conditions of the tunnel face. Compared to traditional methods, it reduces axial robotic roadheader energy consumption. This research enhances the intelligence level of coal mining robots, reduces energy consumption, and increases equipment lifespan.