摘要
针对机器人工程中任务分配问题,开展了面向煤炭事故救援场景的多机器人任务分配实验。通过分析多机器人系统的任务分配问题,将其归类为单任务/多任务机器人、单机器人/多机器人任务以及瞬时/时延分配等八类模型,并对比了集中式与分布式任务分配方法的优缺点。实验针对煤炭救援场景的特殊需求,采用车辆路径规划模型进行任务建模,结合离散粒子群优化算法求解任务分配方案。重点探讨了粒子群优化算法在离散优化问题中的近似处理、二进制粒子、惩罚函数、修正速度-位置方程等四种改进策略,并分析了其适用性与局限性。实验为多机器人任务分配的工程应用供了理论支撑和方法参考。
Aiming at the pedagogical challenges in multi-robot systems(MRS)for robotics engineering education,this study designs a comprehensive experimental framework for multi-robot task allocation(MRTA)in coal accidentrescue scenarios.Through systematic analysis of MRTA problems,we classify them into eight categories based on single-task/multi-task robots(ST/MT),single-robot/multi-robot tasks(SR/MR),and instantaneous/time-delayed allocation(IA/TA),while comparatively evaluating centralized versus distributed allocation methodologies.To address the unique requirements of coal emergency response,a vehicle routing problem(VRP)model is adopted for task modeling,integrated with a discrete particle swarm optimization(PSO)algorithm for solution derivation.The study focuses on four key PSO improvement strategies for discrete optimization:approximation processing,binary particle encoding,penalty function mechanisms,and modified velocity-position equations,with critical analysis of their applicability and limitations.This experiment provides both theoretical foundations and methodological references for MRTA pedagogy and practice,demonstrating dual value in engineering applications and educational innovation.
作者
张泊明
李富强
ZHANG Boming;LI Fuqiang(Xuhai College,China University of Mining and Technology,Xuzhou Jiangsu 221008,China)
出处
《佳木斯大学学报(自然科学版)》
2025年第10期24-27,共4页
Journal of Jiamusi University:Natural Science Edition
基金
中国矿业大学徐海学院教学改革项目(YA2314)。
关键词
多机器人系统
任务分配
车辆路径规划
粒子群优化
事故救援
multi-robot systems
task allocation
vehicle routing problem
particle swarm optimization
accident rescue
