ResearchSpace

Enhancing agent safety through autonomous environment adaptation

Show simple item record

dc.contributor.author Rosman, Benjamin S
dc.contributor.author Hayes, B
dc.contributor.author Scassellati, B
dc.date.accessioned 2015-11-16T07:36:46Z
dc.date.available 2015-11-16T07:36:46Z
dc.date.issued 2015-08
dc.identifier.citation Rosman, B.S., Hayes, B and Scassellati, B. 2015. Enhancing agent safety through autonomous environment adaptation. In: IEEE International Conference on Development and Learning and on Epigenetic Robotics, Providence, Rhode Island, USA, August 13-16, 2015 en_US
dc.identifier.uri http://www.benjaminrosman.com/papers/icdl15.pdf
dc.identifier.uri http://hdl.handle.net/10204/8291
dc.description IEEE International Conference on Development and Learning and on Epigenetic Robotics, Providence, Rhode Island, USA, August 13-16, 2015 en_US
dc.description.abstract Exploration and self-directed learning are valuable components of early childhood development. This often comes at an unacceptable safety trade-off, as infants and toddlers are especially at risk from environmental hazards that may fundamentally limit their ability to interact with and explore their environments. In this work we address this risk through the incorporation of a caregiver robot, and present a model allowing it to autonomously adapt its environment to minimize danger for other (novice) agents in its vicinity. Through an approach focusing on action prediction strategies for agents with unknown goals, we create a model capable of using expert demonstrations to learn typical behaviors for a multitude of tasks. We then apply this model to predict likely agent behaviors and identify regions of risk within this action space. Our contribution uses this information to prioritize and execute risk mitigating behaviors, manipulating and adapting the environment to minimize the potential harm the novice is likely to encounter. We conclude with an evaluation using multiple agents of varying goal-directedness, comparing agents’ self-interested performance in scenarios with and without the assistance of a caregiver incorporating our model. Our experiments yield promising results, with assisted agents incurring less damage, interacting longer, and exploring their environments more completely than unassisted agents. en_US
dc.language.iso en en_US
dc.publisher IEEE en_US
dc.relation.ispartofseries Workflow;15678
dc.subject Self-directed learning en_US
dc.subject Early childhood development en_US
dc.subject Environmental hazards en_US
dc.subject Caregiver robot en_US
dc.subject Varying goal-directedness en_US
dc.subject Self-interested performance en_US
dc.title Enhancing agent safety through autonomous environment adaptation en_US
dc.type Conference Presentation en_US
dc.identifier.apacitation Rosman, B. S., Hayes, B., & Scassellati, B. (2015). Enhancing agent safety through autonomous environment adaptation. IEEE. http://hdl.handle.net/10204/8291 en_ZA
dc.identifier.chicagocitation Rosman, Benjamin S, B Hayes, and B Scassellati. "Enhancing agent safety through autonomous environment adaptation." (2015): http://hdl.handle.net/10204/8291 en_ZA
dc.identifier.vancouvercitation Rosman BS, Hayes B, Scassellati B, Enhancing agent safety through autonomous environment adaptation; IEEE; 2015. http://hdl.handle.net/10204/8291 . en_ZA
dc.identifier.ris TY - Conference Presentation AU - Rosman, Benjamin S AU - Hayes, B AU - Scassellati, B AB - Exploration and self-directed learning are valuable components of early childhood development. This often comes at an unacceptable safety trade-off, as infants and toddlers are especially at risk from environmental hazards that may fundamentally limit their ability to interact with and explore their environments. In this work we address this risk through the incorporation of a caregiver robot, and present a model allowing it to autonomously adapt its environment to minimize danger for other (novice) agents in its vicinity. Through an approach focusing on action prediction strategies for agents with unknown goals, we create a model capable of using expert demonstrations to learn typical behaviors for a multitude of tasks. We then apply this model to predict likely agent behaviors and identify regions of risk within this action space. Our contribution uses this information to prioritize and execute risk mitigating behaviors, manipulating and adapting the environment to minimize the potential harm the novice is likely to encounter. We conclude with an evaluation using multiple agents of varying goal-directedness, comparing agents’ self-interested performance in scenarios with and without the assistance of a caregiver incorporating our model. Our experiments yield promising results, with assisted agents incurring less damage, interacting longer, and exploring their environments more completely than unassisted agents. DA - 2015-08 DB - ResearchSpace DP - CSIR KW - Self-directed learning KW - Early childhood development KW - Environmental hazards KW - Caregiver robot KW - Varying goal-directedness KW - Self-interested performance LK - https://researchspace.csir.co.za PY - 2015 T1 - Enhancing agent safety through autonomous environment adaptation TI - Enhancing agent safety through autonomous environment adaptation UR - http://hdl.handle.net/10204/8291 ER - en_ZA


Files in this item

This item appears in the following Collection(s)

Show simple item record