Reactive Task and Motion Planning under Temporal Logic Specifications
This addresses the challenge of seamless task execution in robotics for applications like human-robot collaboration, though it appears incremental.
The paper tackles the problem of task-and-motion planning under human interventions by introducing a dynamically reconfigurable algorithm that minimizes replanning steps, resulting in superior manipulation performance in simulated and real-world tasks.
We present a task-and-motion planning (TAMP) algorithm robust against a human operator's cooperative or adversarial interventions. Interventions often invalidate the current plan and require replanning on the fly. Replanning can be computationally expensive and often interrupts seamless task execution. We introduce a dynamically reconfigurable planning methodology with behavior tree-based control strategies toward reactive TAMP, which takes the advantage of previous plans and incremental graph search during temporal logic-based reactive synthesis. Our algorithm also shows efficient recovery functionalities that minimize the number of replanning steps. Finally, our algorithm produces a robust, efficient, and complete TAMP solution. Our experimental results show the algorithm results in superior manipulation performance in both simulated and real-world tasks.