Task allocation method for anchoring robot with multiple drilling units and multiple tasks in coal mine roadways

To address the optimal allocation problem of multiple drilling and bolting tasks in a multi-drill anchor robot system for coal mine roadway excavation, a hybrid optimization method (GAVNS) integrating Genetic Algorithm (GA) and Variable Neighborhood Search (VNS) is proposed. Based on roadway geometric parameters and the spatial layout characteristics of multi-drill anchor robots, a layered support process for excavation roadways is developed, dividing the overall support tasks into four independent support layers to achieve layered, parallel, and efficient execution of support tasks. For the optimal configuration of drill quantities in different support layers, a comprehensive optimization model is established considering spatial constraints such as drill working space and feasible operational domains, as well as temporal constraints of support tasks, ensuring efficient operations in each support layer and enabling parallel operation with the excavation robot. Addressing the multi-drill multi-task allocation problem within different support layers, a multi-drill multi-task allocation model is constructed, with a fitness function based on segmented integer encoding and diversified neighborhood structures designed to effectively combine the global search capability of GA with the local optimization advantage of VNS, improving algorithm convergence speed and solution quality. Validation through simulation of coal mine roadway support tasks demonstrates that the proposed GAVNS algorithm outperforms traditional genetic algorithms and single VNS in terms of task completion time, load balancing, and resource utilization. This method has been applied to the coal mine intelligent excavation robot system developed by the team, achieving efficient permanent support operations. It has been in operation for over 10 months in large-section excavation roadways, achieving daily excavation advances exceeding 50 m and monthly advances exceeding 1000 m.