MSR Speaking Qualifier

Title: Joint Surface Reconstruction from Monocular Vision and LiDAR   Abstract: In recent years, dense reconstruction gains popularity because of its broad applications in inspection, mapping, and planning. Cameras or LiDARs are generally deployed for 3D dense reconstruction. However, current reconstruction pipelines based on cameras or LiDARs have significant limitations in achieving an accurate and [...]

Title: Automatic detection of human affective behavior in dyadic conversations Abstract: Emotion is communicated through face, voice, and body motion in interpersonal contexts. Yet, most approaches to automatic detection emphasize a single modality (especially face or voice), ignore social context, and focus on well-defined signs of emotion (e.g., smile).  This thesis addresses multimodal, interpersonal emotion [...]

Title: Structured Representations for Behaviors of Autonomous Robots   Abstract: Autonomous robot behavior can be captured in many ways: as code, as modules of code, in an unstructured form such as a neural net, or in one of several more structured formats such as a graph, table, or tree. This talk explores structured representations that [...]

Title: Vision and Improved Learned-Trajectory Replay for Assistive-Feeding and Food-Plating Robots   Abstract: Food manipulation offers an interesting frontier for robotics research because of the direct application of this research to real-world problems and the challenges involved in robust manipulation of deformable food items. In this talk, we focus on the challenges associated with robots manipulating [...]

Title: Leveraging Multimodal Sensory Data for Robust Cutting   Abstract: Cutting food is a challenging task due to the variety of material properties across food items. In addition, different events occur during the slicing process that need to be monitored and detected for robust execution, such as when a knife has completely cut through a [...]

Title: Toward Safe Reinforcement Learning in the Real World   Abstract: Control for mobile robots in slippery, rough terrain at high speeds is difficult. One approach to designing controllers for complex, non-uniform dynamics in unstructured environments is to use model-free learning-based methods. However, these methods often lack the necessary notion of safety which is needed [...]

Title: Data-Efficient Stage Classification and Failure Detection for Robotic Screwdriving   Abstract: Screwdriving is one of the most common assembly methods, yet its full automation is still challenging, especially for small screws. A critical reason is that existing techniques perform poorly in process monitoring and failure prediction. In addition, most solutions are essentially data-driven, thereby [...]

Title: Environment Generalization in Deep Reinforcement Learning Abstract: A key challenge in deep reinforcement learning (RL) is environment generalization: a policy trained to solve a task in one environment often fails to solve the same task in a slightly different test environment. In this work, we propose the ``Environment-Probing'' Interaction (EPI) policy, which allows the agent [...]

Title: Human Torso Pose Forecasting for the Real World   Abstract: Anticipatory human intent modeling is important for robots operating alongside humans in dynamic or crowded environments. Humans often telegraph intent through posture cues, such as torso or head cues. In this paper, we describe a computationally lightweight approach to human torso pose recovery and forecasting [...]

Title: Object-level visual SLAM for plant modeling   Abstract: A 3D model that can capture the finer details of the plant structure as well as the overall field statistics, plays an important role in automating agriculture. However, modeling and mapping an agricultural field is challenging due to dynamics, illumination conditions and limited texture inherent in an outdoor environment. We propose a pipeline that combines the recent [...]