摘要
面向灾害等突发事件的应急搜救能力是国家应急管理中的一项核心能力。近年来,无人机、消防机器人等智能机器设备在应急搜救中得到了广泛应用。但在很多复杂任务场景中,机器仍无法完全取代人的作用,需要人与机器来协同完成任务。目前对人机协同搜救方法的研究还很有限。本文对人机协同搜救的主要场景和任务需求进行了分析,针对三类典型任务场景,分别构建问题的优化模型,在其中体现人机搜救协同模式与搜救目标行为特征的内在关系,并设计了进化算法与局部搜索相结合的混合优化算法,对复杂场景下的人机搜救行为进行协同优化调度。最后,选择我国南方山区及丘陵等典型地貌构建搜救问题测试实例,并与文献中已有算法进行比较,研究了本文所提模型及算法的有效性。
The ability of search and rescue is a key to emergency management in response to disastrous events.In recent years,intelligent robots,such as unmanned aerial vehicles and fire-fighting robots,have been widely used in emergency search-and-rescue tasks.However,in most complex tasks,robots need to cooperate with human instead of replacing human.In the literature,there are few studies on human-robot cooperation methods for search and rescue.In this paper,we analyze typical scenarios and requirements of human-robot cooperative search and rescue,and formulate three optimization problems according to typical task scenarios to represent the internal relationship between the modes of human-robot cooperation and the behaviors of targets.We propose a hybrid algorithm combining evolutionary search and local search to efficiently solve the problems.Computational results on test instances constructed on typical terrains in South China demonstrate the effectiveness of the proposed method compared to several popular methods in the literature.
作者
郑宇军
杨潇
杜怡辰
江勋林
ZHENG Yujun;YANG Xiao;DU Yichen;JIANG Xunlin(School of Information Science and Technology,Hangzhou Normal University,Hangzhou 311121,China;Department of Engineering Technology and Application,Army Infantry College of PLA,Nanchang 330100,China)
出处
《南昌工程学院学报》
CAS
2022年第3期1-11,共11页
Journal of Nanchang Institute of Technology
基金
国家自然科学基金资助项目(61872123)
浙江省自然科学杰出青年基金资助项目(LR20F030002)。
关键词
人机协同
应急搜救
优化调度
混合进化算法
human-robot cooperation
emergency search and rescue
optimal scheduling
hybrid evolutionary algorithm
