Social navigation framework for assistive robots in human inhabited unknown environments

Kivrak, Hasan; Çakmak, Furkan; Köse, Hatice; Yavuz, Sırma

doi:10.1016/j.jestch.2020.08.008

Cited by 31 publications

(31 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…If models aim to reflect the features of society, we need to question what behaviors should be replicated and promoted. For example, Kivrak et al ( 2020 ) explicitly exclude women in the real-world experiments of their social navigation framework for assistive robots around humans. Their model that aims to yield human-friendly routes was only tested in a corridor where women were excluded based on previous analysis (Jones and Healy, 2006 ), which affirms gender differences in spatial problem solving.…”

Section: Learning—relearning Framework For Socially-aware Robot Namentioning

confidence: 99%

From Learning to Relearning: A Framework for Diminishing Bias in Social Robot Navigation

2021

View full text Add to dashboard Cite

The exponentially increasing advances in robotics and machine learning are facilitating the transition of robots from being confined to controlled industrial spaces to performing novel everyday tasks in domestic and urban environments. In order to make the presence of robots safe as well as comfortable for humans, and to facilitate their acceptance in public environments, they are often equipped with social abilities for navigation and interaction. Socially compliant robot navigation is increasingly being learned from human observations or demonstrations. We argue that these techniques that typically aim to mimic human behavior do not guarantee fair behavior. As a consequence, social navigation models can replicate, promote, and amplify societal unfairness, such as discrimination and segregation. In this work, we investigate a framework for diminishing bias in social robot navigation models so that robots are equipped with the capability to plan as well as adapt their paths based on both physical and social demands. Our proposed framework consists of two components: learning which incorporates social context into the learning process to account for safety and comfort, and relearning to detect and correct potentially harmful outcomes before the onset. We provide both technological and societal analysis using three diverse case studies in different social scenarios of interaction. Moreover, we present ethical implications of deploying robots in social environments and propose potential solutions. Through this study, we highlight the importance and advocate for fairness in human-robot interactions in order to promote more equitable social relationships, roles, and dynamics and consequently positively influence our society.

show abstract

Section: Learning—relearning Framework For Socially-aware Robot Namentioning

confidence: 99%

From Learning to Relearning: A Framework for Diminishing Bias in Social Robot Navigation

2021

View full text Add to dashboard Cite

show abstract

“…Model-based prediction methods predict the navigation intentions of humans based on explicitly modeling the relation between online measurements, such as velocity and heading direction, and the expected outcome (e.g., see [ 12 , 13 ]). Sometimes, such models take into account the interaction between people and robots, such as the work proposed in [ 14 , 15 , 16 , 17 , 18 , 19 ]. Future states of the environment are typically modeled as a pose of the person of interest or an occupancy grid of the map where certain cells are expected to be occupied or not [ 5 ].…”

Section: Related Workmentioning

confidence: 99%

“…For example, at crossings there are several distinct paths a human could take. In the present work, we take a different approach than what was discussed in [ 12 , 13 , 14 , 15 , 16 , 17 , 18 , 19 ], as we propose a coarse discretization of the environment based on its semantics. As a result, our model does not predict a single path of a human but, instead, it predicts to which semantic location a person is directed to by simultaneously evaluating probable alternatives.…”

Section: Related Workmentioning

confidence: 99%

A Probabilistic Model for Real-Time Semantic Prediction of Human Motion Intentions from RGBD-Data

Houtman

Bijlenga

Torta

et al. 2021

Sensors

View full text Add to dashboard Cite

For robots to execute their navigation tasks both fast and safely in the presence of humans, it is necessary to make predictions about the route those humans intend to follow. Within this work, a model-based method is proposed that relates human motion behavior perceived from RGBD input to the constraints imposed by the environment by considering typical human routing alternatives. Multiple hypotheses about routing options of a human towards local semantic goal locations are created and validated, including explicit collision avoidance routes. It is demonstrated, with real-time, real-life experiments, that a coarse discretization based on the semantics of the environment suffices to make a proper distinction between a person going, for example, to the left or the right on an intersection. As such, a scalable and explainable solution is presented, which is suitable for incorporation within navigation algorithms.

show abstract

“…Kivrak et al proposed a novel social navigation framework for mobile service robots, maintaining both human safety and comfort in environments where humans perform actions [12]. Their main contribution is that the framework can be used for unknown environments.…”

Section: Related Workmentioning

confidence: 99%

Reward Shaping with Subgoals for Social Navigation

Okudo,

Yamada

2021

Preprint

View full text Add to dashboard Cite

Social navigation has been gaining attentions with the growth in machine intelligence. Since reinforcement learning can select an action in the prediction phase at a low computational cost, it has been formulated in a social navigation tasks. However, reinforcement learning takes an enormous number of iterations until acquiring a behavior policy in the learning phase. This negatively affects the learning of robot behaviors in the real world. In particular, social navigation includes humans who are unpredictable moving obstacles in an environment. We proposed a reward shaping method with subgoals to accelerate learning. The main part is an aggregation method that use subgoals to shape a reinforcement learning algorithm. We performed a learning experiment with a social navigation task in which a robot avoided collisions and then reached its goal. The experimental results show that our method improved the learning efficiency from a base algorithm in the task.

show abstract

Social navigation framework for assistive robots in human inhabited unknown environments

Cited by 31 publications

References 21 publications

From Learning to Relearning: A Framework for Diminishing Bias in Social Robot Navigation

From Learning to Relearning: A Framework for Diminishing Bias in Social Robot Navigation

A Probabilistic Model for Real-Time Semantic Prediction of Human Motion Intentions from RGBD-Data

Reward Shaping with Subgoals for Social Navigation

Contact Info

Product

Resources

About