Hybrid conditional planning using answer set programming

Abstract

We introduce a parallel offline algorithm for computing hybrid conditional plans, called HCP-ASP, oriented towards robotics applications. HCP-ASP relies on modeling actuation actions and sensing actions in an expressive nonmonotonic language of answer set programming (ASP), and computation of the branches of a conditional plan in parallel using an ASP solver. In particular, thanks to external atoms, continuous feasibility checks (like collision checks) are embedded into formal representations of actuation actions and sensing actions in ASP; and thus each branch of a hybrid conditional plan describes a feasible execution of actions to reach their goals. Utilizing nonmonotonic constructs and nondeterministic choices, partial knowledge about states and nondeterministic effects of sensing actions can be explicitly formalized in ASP; and thus each branch of a conditional plan can be computed by an ASP solver without necessitating a conformant planner and an ordering of sensing actions in advance. We apply our method in a service robotics domain and report experimental evaluations. Furthermore, we present performance comparisons with other compilation based conditional planners on standardized benchmark domains. This paper is under consideration for acceptance in TPLP.

Human-Robot Collaborative Assembly Planning using Hybrid Conditional Planning

“…As a novel contribution, we extend hybrid conditional planning (HCP) [17] to include commonsense reasoning and a richer set of communication actions for collaborative tasks: (i) robot asking for confirmation (e.g., whether the human will assemble the part she is holding), (ii) requesting human to perform some action (e.g., human to unhold a part), (iii) asking for help (e.g., human to assemble a part that is not reachable to the robot), (iv) offering help (e.g., when the assembly part is too heavy or needs precision, or when the task is too tedious for the human), and (v) initiating/ending a conversation (e.g., greeting/acknowledging) and providing explanations. With such a general HCP framework, with sensing actions, robots can identify which assembly part the human workmate is holding; with commonsense reasoning, robots can conclude that a heavy assembly part cannot be moved by a human; and with communication skills, robots can communicate with humans in different ways for safer and effective collaborations.…”

“…In our experiments, we have used the HCP planner HCP-ASP [17] for generating conditional plans, and RRT* motion planner [11] from OMPL [15] for the reachability checks embedded into action descriptions. All experiments are performed on a Linux server using 12 2.4GHz Intel E5-2665 CPU cores and up to 64GB memory.…”

“…We propose a formal method for collaborative assembly planning, using hybrid conditional planning [17], with the following contributions. The proposed formal framework allows planning of hybrid sensing actions and various communication actions, in addition to hybrid actuation actions, based on the formal logical descriptions of these actions.…”

Human Robot Collaborative Assembly Planning: An Answer Set Programming Approach

“…In this study, we use the hybrid conditional planner HCP-ASP (Yalciner et al 2017), based on a parallel algorithm that calls the ASP solver CLINGO to compute the branches. The actuation actions and sensing actions are represented in ASP, and the feasibility checks are embedded into these action descriptions by external atoms, as suggested by Yalciner et al (2017).…”

“…We propose a novel method for collaborative assembly planning, utilizing hybrid conditional planning (HCP-ASP) (Yalciner et al 2017) based on answer set programming (ASP) (Brewka et al 2016).…”