The ability to solve motion-planning queries within a fixed time budget is critical for deploying robotic systems in time-sensitive applications. Semi-static environments, where most of the workspace remains fixed while a subset of obstacles varies between tasks, exhibit structured variability that can be exploited to provide stronger guarantees than general-purpose planners. However, existing approaches either lack formal coverage guarantees or rely on discretizations of obstacle configurations that restrict applicability to realistic domains. This paper introduces COVER, a framework that incrementally constructs coverage-verified roadmaps for semi-static environments. COVER decomposes the arrangement space by independently partitioning the configuration space of each movable obstacle and verifies roadmap feasibility within each partition, enabling fixed-time query resolution for verified regions.We evaluate COVER on a 7-DoF manipulator performing object-picking in tabletop and shelf environments, demonstrating broader problem-space coverage and higher query success rates than prior work, particularly with obstacles of different sizes.

翻译：在固定时间预算内解决运动规划查询的能力对于在时间敏感应用中部署机器人系统至关重要。半静态环境中，大部分工作空间保持不变，而部分障碍物在不同任务之间变化，这种结构化变异性可被利用，从而提供比通用规划器更强的保证。然而，现有方法要么缺乏形式化覆盖保证，要么依赖障碍物配置的离散化，从而限制了在现实领域的适用性。本文提出COVER框架，该框架通过增量构建半静态环境的覆盖验证路线图。COVER通过独立划分每个可移动障碍物的配置空间来分解排列空间，并在每个分区内验证路线图的可行性，从而实现对已验证区域的固定时间查询求解。我们在执行桌面和货架环境中物体抓取任务的7自由度机械臂上评估COVER，结果表明，相比先前工作，该方法具有更广泛的问题空间覆盖范围和更高的查询成功率，尤其在处理不同尺寸的障碍物时表现突出。