Masters Thesis Proposals

 Learning Adaptative Robot Behaviours for Individual Human Preferences

Navigating around humans in real-world environments is complex. In addition, people's preferences for how robots should behave around humans can differ. Adapating to individual preferences could therefore allow for more seamless interactions between humans and robots. However, how to predict these preferences and adapt robot behaviour accordingly remains underexplored. This project therefore incorporates both designing and implementing robot navigation algorithms as well as collecting data about human preferences. 

 

This masters thesis will be part of the WASP-NEST PerCorSo project; https://wasp-sweden.org/nest-project-percorso/.

 

Useful skills

• Programming experience (python) -- ROS and robot experience an advantage
• Interest in Human Robot Interaction and/or user studies

References

Linard, A., Torre, I., Bartoli, E., Sleat, A., Leite, I., & Tumova, J. (2023, October). Real-time rrt* with signal temporal logic preferences. In 2023 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) (pp. 8621-8627). IEEE.

Holk, S., Marta, D., & Leite, I. (2024, May). Polite: Preferences combined with highlights in reinforcement learning. In 2024 IEEE International Conference on Robotics and Automation (ICRA) (pp. 2288-2295). IEEE.

Munzer, T., Toussaint, M., & Lopes, M. (2017, May). Preference learning on the execution of collaborative human-robot tasks. In 2017 IEEE International Conference on Robotics and Automation (ICRA) (pp. 879-885). IEEE.

 

Gender Perceptions in Human-Robot Interaction

Gendering robots and thereby mirroring social norms may lead to the reproduction and perpetuation of gender biases. A recent UNESCO report, for example, has demonstrated that the gendering of voice assistants led to the production and reproduction of gender stereotypes, especially the notion that women should be submissive, polite, and patient.  Recent studies investigated how social robots could be designed to go against current digital assistants’ gender norms and suggest that feminist robots can play a role in reducing gender biases and harmful stereotyping. 

 

Useful skills

• Programming experience (python)
• Interest in Human Robot Interaction and/or user studies

Start: Anytime

Contact: Rebecca Stower

Keywords: Human-Robot Interaction, gender, social norms, robot failures, user studies

 

References

Winkle, K., McMillan, D., Arnelid, M., Harrison, K., Balaam, M., Johnson, E., & Leite, I. (2023, March). Feminist human-robot interaction: Disentangling power, principles and practice for better, more ethical HRI. In Proceedings of the 2023 ACM/IEEE International Conference on Human-Robot Interaction (pp. 72-82).

Winkle, K., Melsión, G. I., McMillan, D., & Leite, I. (2021, March). Boosting robot credibility and challenging gender norms in responding to abusive behaviour: A case for feminist robots. In Companion of the 2021 ACM/IEEE international conference on human-robot interaction (pp. 29-37).

Galatolo, A., Melsión, G. I., Leite, I., & Winkle, K. (2023). The right (wo) man for the job? exploring the role of gender when challenging gender stereotypes with a social robot. International Journal of Social Robotics, 15(11), 1933-1947.

Miranda, L., Castellano, G., & Winkle, K. (2024, March). A Case for Diverse Social Robot Identity Performance in Education. In Companion of the 2024 ACM/IEEE International Conference on Human-Robot Interaction (pp. 28-35).

 

Drone Failures in Human Drone Interactions

Unmanned Aerial Vehicles (UAVs) such as drones are becoming increasingly common and have applications in contexts such as exploration and search and rescue, where humans might not be able to easily access such spaces. In this sense, drones are not only a technical tool, but can be perceived as a social agent capable of communicating information to a user about the environment. Yet, studies which focus on the social aspect of interactions between humans and drones are scarce. Furthermore, there are opportunities for errors to occur in such interactions (e.g., imperfect sensors, battery drain, other malfunctions) which can affect the quality and success of the human-drone communication. Consequently, this research project will explore how technological breakdowns of drone robots are perceived by human users. 

Useful skills

• Programming experience (python)
• Interest in Human Robot Interaction

Start: Anytime

Contact: Rebecca Stower

Keywords: Human-Robot Interaction, robot failures, user studies

References

D. Tezza and M. Andujar, "The State-of-the-Art of Human–Drone Interaction: A Survey," in IEEE Access, vol. 7, pp. 167438-167454, 2019, doi: 10.1109/ACCESS.2019.2953900.

Viviane Herdel, Lee J. Yamin, and Jessica R. Cauchard. 2022. Above and Beyond: A Scoping Review of Domains and Applications for Human-Drone Interaction. In Proceedings of the 2022 CHI Conference on Human Factors in Computing Systems (CHI '22). Association for Computing Machinery, New York, NY, USA, Article 463, 1–22. https://doi.org/10.1145/3491102.3501881