Abstract: |
As the complexity of exploration scenarios and task requirements continue to increase, single autonomous exploration has gradually been unable to meet the efficiency and stability requirements of tasks. Therefore, research on multi UAV collaborative autonomous exploration has received more and more attention. Among them, the multirotors are often used for autonomous exploration tasks in complex and small scale scenes due to their agility, which has higher requirements for the real time performance and autonomy of the collaborative exploration system. Therefore, this paper focuses on the multi UAV collaborative exploration system of multirotors, and focuses on the unknown environment exploration strategy and multi UAV task allocation strategy. It introduces the current research progress of decision making technology in detail and analyzes and summarizes its characteristics. Specifically, this paper classifies the unknown environment exploration strategies as methods based on frontier, sampling, and the combination of the two, respectively introduces them and analyzes their characteristics. Moreover, this paper classifies multi UAV task allocation strategies as methods based on markets and auctions, introduces in detail the modeling process of two types of methods for task allocation problems, and focuses on the specific application of collaborative exploration work in the real world, especially for the strategies considering communication limits, and discuss collaborative exploration in limited scenarios challenges and current approaches. Finally, the research status of multi UAV collaborative exploration technology is summarized, and the future research focus is prospected with reference significance. |