Full coverage cutting path planning of robotized roadheader to improve cutting stability of the coal lane cross-section containing gangue

Abstract

The purpose of this work is to improve the cutting stability of robotized roadheader through the full coverage cutting path planning of the coal lane cross-section containing gangue. Cutting gangue will bring serious pick wear and severe vibration, which will reduce the service life of robotized roadheader. Therefore, the strategy that avoiding the gangue and cutting the remaining coal-rock was recommended. Firstly, the environment grid map of the cross-section containing gangue was established and the grid attribute was assigned. To describe the arbitrary position of gangue, a random generation method was developed. On this basis, the combination of biologically inspired neural network (BINN) and floating template method was proposed to overcome the shortcomings of traditional BINN and the full coverage cutting path planning simulation was carried out. The simulation shows a better result that the average repetition rate is approximately 10% under the condition that the cutting coverage rate is more than 95%. Finally, the cutting experiment of the coal-rock sample containing gangue on the robotized roadheader cutting platform was performed. Based on the infrared thermography and cutting signal obtained by the previous round cutting, the cutting path of the second round was planned and the cutting experiment was conducted. The experiment results show that the cutting temperature rise and the cutting vibration of the second round cutting can be effectively reduced by approximately 60% and 90%, which demonstrated that the cutting stability of the cross-section containing gangue can be effectively improved by the cutting path planning strategy of avoiding gangue.