针对大规模隧道工程中多台多臂凿岩台车协同作业在任务分配与路径规划上的强耦合及多目标冲突难题，提出一种基于改进非支配排序遗传算法Ⅱ（NSGA-Ⅱ）的多目标协同优化方法。结合凿岩施工工艺特点，将施工过程划分为掏槽孔、辅助孔和周边孔3个阶段，并将孔序规划统一建模为同时最小化总完工时间、钻臂负载方差和系统总能耗的多目标优化问题。在此基础上，引入精英种群初始化策略、自适应遗传算子以及基于惩罚函数的约束处理机制，将动态安全距离、作业时序及空间干涉等工程约束有效融入优化框架，采用嵌入式路径优化模块生成钻臂在任务子集内的高效移动路径，并通过优劣解距离法（TOPSIS）对Pareto最优解集进行决策筛选。测试结果表明： 1）所提出的方法能够生成具有空间分区特性、无碰撞且工程中可执行的协同作业方案。2）在典型工况下，与贪心算法和MOEA/D算法相比，采用本文方法的最优总完工时间分别缩短约9.27%和14.68%，负载方差降低至0.69，显著提升了多钻臂作业的并行效率与均衡性。
Coordinated operation of multi-boom rock jumbos in large-scale tunneling involves strong coupling between task allocation and path planning, along with multiple conflicting objectives. To address these challenges, this study proposes a multi-objective co-optimization framework based on an improved non-dominated sorting genetic -algorithm Ⅱ (NSGA-Ⅱ). The framework simultaneously minimizes total makespan, workload variance, and overall energy consumption while enforcing practical engineering constraints through a penalty-based mechanism. Considering drill-and-blast operational characteristics, the construction process is divided into three stages, namely cut holes, auxiliary holes, and perimeter holes, and the drilling sequence is formulated as a multi-objective optimization problem. Algorithm performance is enhanced through a hybrid evolutionary mechanism combining elite population initialization with adaptive genetic operators, and an embedded path optimization module generates efficient movement paths for drill booms within their assigned task subsets. The technique for order of preference by similarity to ideal solution (TOPSIS) is employed for final solution selection. Experimental results demonstrate that: (1)The proposed method produces collaborative, spatially partitioned, collision-free work plans that are practically executable. (2)Compared with the conventional greedy algorithm and the multi-objective evolutionary algorithm based on decomposition, the optimal makespan is reduced by approximately 9.27% and 14.68%, respectively, and the workload variance is reduced to 0.69, markedly improving parallel efficiency and workload balance in multi-boom operations.
