Modeling Human Helpfulness with Individual and Contextual Factors for Robot Planning

Abstract

Robots deployed in human-populated spaces often need human help to effectively complete their tasks. Yet, a robot that asks for help too frequently or at the wrong times may cause annoyance, and a robot that asks too infrequently may be unable to complete its tasks. In this paper, we present a model of humans' helpfulness towards a robot in an office environment, learnt from online user study data. Our key insight is that effectively planning for a task that involves bystander help requires disaggregating individual and contextual factors and explicitly reasoning about uncertainty over individual factors. Our model incorporates the individual factor of latent helpfulness and the contextual factors of human busyness and robot frequency of asking. We integrate the model into a Bayes-Adaptive Markov Decision Process (BAMDP) framework and run a user study that compares it to baseline models that do not incorporate individual or contextual factors. The results show that our model significantly outperforms baseline models by a factor of 1.5X, and it does so by asking for help more effectively: asking 1.2X times less while still receiving more human help on average. p ask Human 3 Human 1 tions help it. e model.

Planning with Hidden Parameter Polynomial MDPs

“…When MDPs are used to describe a system, we assume that the state is fully observable to the agent. However, this assumption cannot be satisfied in cases where the sensors used to determine the state are unreliable (Kaelbling, Littman, and Cassandra 1998;Hsiao, Kaelbling, and Lozano-Perez 2007), or the states are dependent on unknown factors, such as a human's likelihood to assist a robot (Nanavati et al 2021;Costen et al 2022). Partially Observable MDPs (POMDPs) (Kaelbling, Littman, and Cassandra 1998), and specialisations thereof are typically used to describe such systems.…”

“…This is done by parameterising the transition probabilities with a set of continuous parameters, which can more accurately reflect a real system. However, many approaches discretise the parameter space (Nanavati et al 2021) to reduce the complexity of the problem. In other approaches where transition probabilities are assumed to be independent, every transition probability is represented as a latent parameter (Rigter, Lacerda, and Hawes 2021b;Grover, Basu, and Dimitrakakis 2020).…”