移动机器人地形约束条件下的自主探索算法综述OA

Autonomous exploration in unknown environments is a core technology that enables long-duration autonomous operation of mobile robots and is valuable in broad applications,including disaster rescue,underground inspection,and mining and tunnel operations.However,in unstructured three-dimensional environments,severe terrain variations,incomplete sensory information,and high environmental uncertainty make terrain assessment and traversability modeling major bottlenecks to improving the safety and efficiency of autonomous exploration systems.Focusing on autonomous exploration under terrain constraints,this paper presents a systematic review of recent research on mobile robots in this area.First,the overall technical framework and typical system architectures of autonomous exploration are summarized,and the role of terrain traversability modeling in target point selection and path planning is analyzed.Subsequently,four representative technical paradigms—frontier-based methods,viewpoint-based value methods,hybrid exploration strategies,and intelligent exploration approaches—are comprehensively reviewed,with emphasis on their fundamental principles,applicable scenarios,and performance characteristics and limitations in complex terrains.Finally,future development trends of autonomous exploration in complex unstructured three-dimensional environments are discussed,highlighting key research directions such as multi-kinematic-platform adaptability,terrain perception fusion,and intelligent decision-making.