Replanning for Situated Robots

Abstract

Planning enables intelligent agents, such as robots, to act so as to achieve their long term goals. To make the planning process tractable, a relatively low fidelity model of the world is often used, which sometimes leads to the need to replan. The typical view of replanning is that the robot is given the current state, the goal, and possibly some data from the previous planning process. However, for robots (or teams of robots) that exist in continuous physical space, act concurrently, have deadlines, or must otherwise consider durative actions, things are not so simple. In this paper, we address the problem of replanning for situated robots. Relying on previous work on situated temporal planning, we frame the replanning problem as a situated temporal planning problem, where currently executing actions are handled via Timed Initial Literals (TILs), under the assumption that actions cannot be interrupted. We then relax this assumption, and address situated replanning with interruptible actions. We bridge the gap between the low-level model of the robot and the high-level model used for planning by the novel notion of a bail out action generator, which relies on the low-level model to generate highlevel actions that describe possible ways to interrupt currently executing actions. Because actions can be interrupted at different times during their execution, we also propose a novel algorithm to handle temporal planning with time-dependent durations.

Safe Multi-Agent Pathfinding with Time Uncertainty

“…Temporal planning (Coles et al, 2008;Coles & Coles, 2014;Cashmore et al, 2019) is a popular and well-researched sub-area of AI planning in which actions have durations. Cimatti et al (2018) recently defined the Strong Temporal Planning with Uncertain action Durations (STPUD) problem, which is a temporal planning problem in which action durations are uncontrollable.…”

Planning and Acting While the Clock Ticks

“…Even introducing actions with controllable durations raises questions about dispatchability (Morris 2016) that need to be addressed. One possibility is to leverage the replanning compilation suggested for situated temporal planning (Cashmore et al 2019), which can also be used for our concurrent planning and execution formulation. Additionally, uncertain action durations can be handled as part of the planning process (Cimatti et al 2018).…”

“…This is useful when there are temporal constraints such as deadlines and when planning time might affect the feasibility of meeting such deadlines. For example, situated tem-poral planning was shown to be useful for online replanning for robots (Cashmore et al 2019).…”

The Story So Far on Narrative Planning

“…Ultimately, a narrative is successful to the degree it accomplishes an author's intent. While similar to the process of anticipating a plan's execution dynamics (as in robotic task planning; e.g., Cashmore et al 2019), it is further complicated by the nature of storytelling: the domain of discourse is different when plans are narrated, as plan effects are manifest in the mental states of the audience. Thus, a narrative plan's success is not directly observable and must be empirically assessed.…”