CS MSc Thesis Presentation 8 September 2026

Tuesday, 8 September there will be a master thesis presentation in Computer Science at Lund University, Faculty of Engineering.

The presentation will take place in E:4130 (Lucas).

Note to potential opponents: Register as an opponent to the presentation of your choice by sending an email to the examiner for that presentation (firstname [dot] lastname [at] cs [dot] lth [dot] se). Do not forget to specify the presentation you register for! Note that the number of opponents may be limited (often to two), so you might be forced to choose another presentation if you register too late. Registrations are individual, just as the oppositions are! More instructions for opponents are found here on the LTH thesis project page.

11:00-12:00 in E:4130 (Lucas)

• Presenter: Tianci Wang
• Title: Learning Context-Aware Guidance Behaviors for Inclusive Museum Robots
• Examiner: Jacek Malec
• Supervisors: Davide Tateo (LTH), Andrea Bonarini (Politecnico di Milano)

Human-robot interaction aims to enable robots to respond appropriately to people. In museums, guide robots lead visitors through tours, where changing visitor states may require different guidance behaviours. This thesis investigates whether reinforcement learning can improve museum robot guidance while balancing visitor needs and tour efficiency. We modelled a museum environment and visitor dynamics in MuJoCo. Visitor motion follows a Social Force Model combining goal-directed movement with interactions from nearby people, the robot, and the environment. During tours, visitors follow the robot and listen to explanations. Human states of distraction, impatience, and overwhelm are inferred using fuzzy logic from temporal and spatial factors. Robot uses predefined guidance behaviours, while reinforcement learning optimizes their control parameters. Different algorithms are compared with a fixed baseline. Results show that the optimized strategies shorten tour duration and reduce all three negative visitor states, highlighting the potential of reinforcement learning for adaptive museum guide robots.