TY - CHAP
T1 - Balancing Communication and Behavior
AU - Sreedharan, Sarath
AU - Kulkarni, Anagha
AU - Kambhampati, Subbarao
N1 - Publisher Copyright:
© 2022, Springer Nature Switzerland AG.
PY - 2022
Y1 - 2022
N2 - In the previous sections, we considered how in human-robot teaming scenarios, the robot behavior influences and is influenced by the human’s mental model of the robot. We have been quantifying some of the interaction between the behavior and human’s model in terms of three interpretability scores, each of which corresponds to some desirable property one would expect the robot behavior to satisfy under cooperative scenarios. With these measures defined, one of the strategies the robot can adopt is to specifically generate behavior that optimizes these measures. Though for one of those measures we also investigated an alternate strategy, namely to use communication to alter the human’s mental model to improve explicability of the current plan. Interestingly, these strategies are not mutually exclusive, they can in fact be combined to capitalize on the individual strengths of each to create behaviors that are unique. In this chapter, we will start by focusing on how one can combine explicable and explanation generation and will look at a compilation based method to generate such plans. In general, communication is a strategy we can use for the other two measures as well. As such, we will end this chapter with a brief discussion on how we could use the idea of combining communication and selective behavior generation to improve legibility and predictability scores.
AB - In the previous sections, we considered how in human-robot teaming scenarios, the robot behavior influences and is influenced by the human’s mental model of the robot. We have been quantifying some of the interaction between the behavior and human’s model in terms of three interpretability scores, each of which corresponds to some desirable property one would expect the robot behavior to satisfy under cooperative scenarios. With these measures defined, one of the strategies the robot can adopt is to specifically generate behavior that optimizes these measures. Though for one of those measures we also investigated an alternate strategy, namely to use communication to alter the human’s mental model to improve explicability of the current plan. Interestingly, these strategies are not mutually exclusive, they can in fact be combined to capitalize on the individual strengths of each to create behaviors that are unique. In this chapter, we will start by focusing on how one can combine explicable and explanation generation and will look at a compilation based method to generate such plans. In general, communication is a strategy we can use for the other two measures as well. As such, we will end this chapter with a brief discussion on how we could use the idea of combining communication and selective behavior generation to improve legibility and predictability scores.
UR - http://www.scopus.com/inward/record.url?scp=85139485131&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85139485131&partnerID=8YFLogxK
U2 - 10.1007/978-3-031-03767-2_7
DO - 10.1007/978-3-031-03767-2_7
M3 - Chapter
AN - SCOPUS:85139485131
T3 - Synthesis Lectures on Artificial Intelligence and Machine Learning
SP - 95
EP - 105
BT - Synthesis Lectures on Artificial Intelligence and Machine Learning
PB - Springer Nature
ER -