Trust perception based human-robot collaborative indoor target search

Effective human-robot collaboration is the key to improving the efficiency of indoor target search tasks, and the correct perception of human-robot trust is a prerequisite for human-robot collaboration. The trust of collaborators in intelligent agents can significantly improve the search efficiency...

Full description

Saved in:
Bibliographic Details
Main Authors: Feng, Yao, Lei, Bao, Huailin, Zhao
Format: Conference Proceeding
Language:English
Subjects:
Collaboration
Collaboration mechanism
Costs
Human-machine systems
Human-robot cooperation
Intelligent agents
Scene prior knowledge
Semantic relation atlas
Semantics
Simulation
Simulation platform
Target search
Task analysis
Trust
Trust perception
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Effective human-robot collaboration is the key to improving the efficiency of indoor target search tasks, and the correct perception of human-robot trust is a prerequisite for human-robot collaboration. The trust of collaborators in intelligent agents can significantly improve the search efficiency of tasks, while distrust may lead to increased search costs or even failure. We propose a trust based human-machine collaborative search mechanism, which integrates semantic relationship graph as scene prior knowledge in the search. We also build a human-machine collaborative simulation platform for target search and use it to verify the effectiveness of our method. The simulation results show that when the trust between humans and machines is at a high level, the shorter the time to complete the search task, the higher the efficiency.
ISSN:2189-8723
DOI:10.1109/ICIIBMS60103.2023.10347887