CAJ | 학술논문

针对分布式任务分配中的资源竞争冲突问题以及解的质量和求解代价难以协调的问题，提出了一种基于多模态融合的多机器人分布式自组织任务分配算法。它将基于信息素的隐式交互机制、基于视觉的阈值模型任务分配机制和基于通信的显式协商协调机制等相结合，实现了未知环境下多机器人系统的多模态分布式自组织任务分配。仿真实验结果表明，多模态融合技术扩大了机器人的感知范围，提高了系统可靠性，而且在较少通信的情况下，有效解决了任务分配冲突问题，提高了任务执行效率。
침대분포식임무분배중적자원경쟁충돌문제이급해적질량화구해대개난이협조적문제，제출료일충기우다모태융합적다궤기인분포식자조직임무분배산법。타장기우신식소적은식교호궤제、기우시각적역치모형임무분배궤제화기우통신적현식협상협조궤제등상결합，실현료미지배경하다궤기인계통적다모태분포식자조직임무분배。방진실험결과표명，다모태융합기술확대료궤기인적감지범위，제고료계통가고성，이차재교소통신적정황하，유효해결료임무분배충돌문제，제고료임무집행효솔。
In distributed task allocation , it is hard to balance three problems:conflict, solution quality and solving price .Multimodal synthesis method is put forward for multi-robot distributed and self-organized task allocation . Three models of cooperation mechanism are combined; these three models are conflict problem , which is implicit cooperation mechanism based on pheromone , implicit task allocation realized by threshold method based on robot visual sense , and explicit task allocation based on improved contract net mechanism .The experiment results and their analysis show preliminarily that multimodal synthesis method can expand the scope of the robot perception , improve the system reliability , resolve the conflict problem by less communication , and improve the task execution efficiency .