Reinforcement learning has emerged as a popular tool for solving control tasks, with multiple works focusing on the complex and dynamic task of locomotion. However, the naive application of reinforcement learning to this problem often produces maladaptive policies that exploit the model or reward function. This results in behavior that...
A chair, once placed, will stay put until moved. Or will it? With the rise of technology being embeddable into everyday objects, what if that chair could move itself? Such robotic furniture has been featured in advertisements, art, and Human-Robot Interaction (HRI) research. Existing methods for operating robotic furniture have...
When studying robot systems, it is common to ask about optimal approaches to accomplish a given task. In the context of mobile systems, particularly biomimetic systems, optimization tasks are closely related to the relevant dynamics of locomotion. In this thesis, building on prior work from the geometric mechanics community, we...
Human-robot teams are invaluable for mapping unknown environments, exploring difficult-to-reach areas, and manipulating inaccessible equipment. However, guiding autonomous robots requires dealing with these dynamic domains while synthesizing a significant amount of data and balancing competing objectives. Current mission planning methods often involve manually specifying low-level parameters of the mission, such...
In-hand manipulations consist of dexterous motions that come easy to humans but still pose a challenge to robotic systems. It is difficult to control finger motions in long complicated sequences due to high DOFs and intricate contact interactions. For such complex motions, in-hand manipulations have generally been broken into a...
Autonomous agents that sense, decide, act, and coordinate effectively with each other are critical in many real-world domains such as autonomous driving, search and rescue missions, air traffic management, and underwater or deep space exploration. All such domains share a key difficulty: though high-level mission goals are clear to system...
Human-safe Cobots are a popular replacement for industrial robots to automate manufacturing processes such like cutting and deburring that all involve accurate contour following across a workpiece surface while maintaining sufficient tool contact. Cobot arms are less expensive and human-safe but they have larger error motion in comparison when only...
Deep learning has recently revolutionized robot perception in many canonical robotic applications, such as autonomous driving. However, a similar transformation has yet to occur in more harsh environments including underwater and underground. This is due in part to the difficulty in deploying robots in these environments, which lack large real...
Labeling videos is costly, time-consuming and tedious. These costs can escalate in applications such as medical diagnosis or autonomous driving where we need domain expertise for annotation. Few-shot action recognition aims to solve this problem by annotation-efficient learning mechanisms.
This thesis presents MetaUVFS as the first Unsupervised Meta-learning algorithm for...
Relative to visual and audible communication, haptic perception is quiet, does not require eye contact, can be applied at any point on the human body, and can function in parallel with the other four human senses. Vibrotactile feedback is a method of haptic communication using vibrations applied to the skin....