Collaborative Planning with Encoding of Users' High-Level Strategies

Abstract

The generation of near-optimal plans for multi-agent systems with numerical states and temporal actions is computationally challenging. Current off-the-shelf planners can take a very long time before generating a near-optimal solution. In an effort to reduce plan computation time, increase the quality of the resulting plans, and make them more interpretable by humans, we explore collaborative planning techniques that actively involve human users in plan generation. Specifically, we explore a framework in which users provide high-level strategies encoded as soft preferences to guide the low-level search of the planner. Through human subject experimentation, we empirically demonstrate that this approach results in statistically significant improvements to plan quality, without substantially increasing computation time. We also show that the resulting plans achieve greater similarity to those generated by humans with regard to the produced sequences of actions, as compared to plans that do not incorporate user-provided strategies.

NL2LTL – a Python Package for Converting Natural Language (NL) Instructions to Linear Temporal Logic (LTL) Formulas

“…Related Work Existing works fall under two buckets: one which admits support for a range of LTL formulas but compromises on the expressiveness of the input (Hahn et al 2022;Schmitt 2022;Narizzano et al 2018;Narizzano and Vuotto 2017), and the other which admits natural language inputs but were built for a particular domain like robotics and are not readily useful as a general purpose package for practitioners (Wang et al 2020;Wang 2020;Nikora and Balcom 2009;Dwyer, Avrunin, and Corbett 1998;Kim, Banks, and Shah 2017;Lignos et al 2015). Furthermore, among these works, other than (Wang 2020;Narizzano and Vuotto 2017;Schmitt 2022), none have publicly available code and are therefore not readily usable for practitioners.…”

RADAR-X: An Interactive Mixed Initiative Planning Interface Pairing Contrastive Explanations and Revised Plan Suggestions

“…The spectrum of work in human-in-the-loop planning (Kambhampati and Talamadupula 2015) ranges from the more traditional mixed-initiative settings (Ferguson et al 1996;Ai-Chang et al 2004;Kim, Banks, and Shah 2017) where the planners drove the interaction in these scenarios with the users 'advising' them, to works in decision support systems (Grover et al 2020;Sengupta, Chakraborti, and Kambhampati 2018;Mishra et al 2019) where the user is responsible for the plan while the system provides support. In (Grover et al 2020), the authors propose a proactive aspect to decision support systems and design the system's capabilities based on principles in Human-Computer Interaction (HCI).…”

Supervised Bayesian specification inference from demonstrations

“…Within the context of stochastic sequential decision-making problems, Kasenberg and Scheutz (2017) and Xu et al (2019) proposed specification inference models conditioned on observations of both state and actions of the trajectories provided by the demonstrators. Similarly, within symbolic planning domains, prior research focused on identifying the minimal LTL formulas that explain the difference between sets of plan traces (Camacho and McIlraith, 2019; Kim et al, 2017). Finally, Chou et al (2022) proposed an optimization-based method that infers a compact specification as a temporal logic formula with parametric propositions from demonstrations.…”