Adapting Robotic Explanations for Robotic Failures in Human Robot Collaboration

My research focuses on adapting robotic failure explanations to enhance human-robot collaboration (HRC). I examine how different explanation types and explanation progression strategies impact failure resolution and user satisfaction by conducting a user study with multiple interaction rounds featuring repeated robotic failures and varying explanations. I also created a novel multimodal dataset of human responses to these failures and explanations. By analyzing human behavioral responses, I developed a predictor to anticipate user confusion following a specific robotic explanation at a robotic failure. This predictor enables an adaptive mechanism to dynamically adjust explanations based on user needs, fostering efficient and natural collaboration. This research aims to significantly improve overall user experience in HRC, making collaborations with robots smoother and more intuitive even when failures occur.