Transfer Learning of Human Preferences for Proactive Robot Assistance in Assembly Tasks

Nemlekar, Heramb; Dhanaraj, Neel; Guan, Angelos; Gupta, Satyandra K.; Nikolaidis, Stefanos

doi:10.1145/3568162.3576965

Cited by 7 publications

(1 citation statement)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Advancement in artificial intelligence (AI) and robotics technologies brings robots out of laboratories, requiring them to perform daily tasks for or with humans [47,54,56,58,67]. To accommodate this requirement, robot task learning aims at teaching robot tasks such as prehabilitation [81], companion [64], assembly tasks [61] according to user preferences. Traditional robot learning algorithms rely on delicately handcrafted reward functions to guide robot behavior.…”

Section: Introductionmentioning

confidence: 99%

FARPLS: A Feature-Augmented Robot Trajectory Preference Labeling System to Assist Human Labelers’ Preference Elicitation

Lyu,

Bai,

Liang

et al. 2024

Proceedings of the 29th International Conference on Intelligent User Interfaces

View full text Add to dashboard Cite

users' familiarities, difficulties of the trajectory pair, and level of disagreements. At the same time, the system monitors labelers' consistency and provides feedback on labeling progress to keep labelers engaged. A between-subjects study (N = 42, 105 pairs of robot pick-and-place trajectories per person) shows that FARPLS can help users establish preference criteria more easily and notice more relevant details in the presented trajectories than the conventional interface. FARPLS also improves labeling consistency and engagement, mitigating challenges in preference elicitation without raising cognitive loads significantly. CCS CONCEPTS• Human-centered computing → HCI design and evaluation methods; Interactive systems and tools.

show abstract