Synergistic Temporal Logic Planning for an Aerial Gripper

In this paper, we present an innovative Linear Temporal Logic (LTL) planner specifically designed for robots encountering intricate constraints on their actions and mobility modes. Our method employs a novel location-agnostic workspace abstraction, focusing on resource allocation. This results in the creation of a concise multigraph representing the system's capabilities. We introduce a synergistic framework for motion and action planning, complemented by an interface layer that coordinates discrete and continuous variables. By separating geometric data from the discrete level, our system achieves efficient online replanning. Complex constraints are carefully integrated throughout graph synthesis and motion planning stages, ensuring safe execution. We validate the effectiveness of our planner by applying it to an aerial gripper, demonstrating robust real-world cargo transportation capabilities. The video illustration can be viewed at https://youtu.be/45DzOUT_1Eg?t=27.