In robotic applications, the control, and actuation deal with a continuous description of the system and environment, while high-level planning usually works with a discrete description. This paper considers the problem of bridging the low-level control and high-level planning for robotic systems via sensor data. In particular, we propose a discretization algorithm that identifies free polytopes via lidar point cloud data. A transition graph is then constructed where each node corresponds to a free polytope and two nodes are connected with an edge if the two corresponding free polytopes intersect. Furthermore, a distance measure is associated with each edge, which allows for the assessment of quality (or cost) of the transition for high-level planning. For the low-level control, the free polytopes act as a convenient encoding of the environment and allow for the planning of collision-free trajectories that realizes the high-level plan. The results are demonstrated in high-fidelity ROS simulations and experiments with a drone and a Segway.

翻译：在机器人应用中,控制和激活涉及对系统和环境的连续描述,而高级规划则通常使用离散描述。本文件审议了通过传感器数据连接对机器人系统进行低层次控制和高层次规划的问题。特别是,我们建议了一种独立的算法,通过利达尔点云数据来识别免费的多端托盘。然后在每个节点与自由的多管和两个节点相对应的地方构造一个边缘连接,如果两个对应的免费多托盘相互交织。此外,与每个边缘相连的距离测量法可以用来评估高层规划过渡的质量(或成本),对于低层次控制,自由的多端机可以方便地对环境进行编码,并能够规划实现高层计划的无碰撞轨迹。结果表现在高密度的ROS模拟中,以及用无人机和Segway进行实验。