June 16, 2023
Title:
Neural Fields for Robotic Manipulation
Abstract:
Neural fields, or coordinate-based neural networks, have garnered significant attention for their ability to parametrize physical properties across space and time, thereby offering novel solutions to complex visual computing challenges. In this thesis, we explore the potential applications of neural fields within the domain of robotics, focusing on three pivotal projects that exemplify the transformative capabilities of this approach. Firstly, we demonstrate how neural fields can be employed to synthesize synthetic data, generating rich and varied training samples that enhance the robustness and accuracy of machine learning models for robotic perception. Secondly, we present a novel application of neural fields for generating orthographic views of objects and environments, a critical capability for affordance-based robotic manipulation systems. Through the precise prediction of orthographic views, robots are empowered to perform 6-DoF manipulation tasks without relying on prone-to-error depth sensors. Lastly, we investigate the scalability of neural fields for city-scale modeling, providing a comprehensive framework that integrates urban data to simulate intricate cityscapes. This large-scale modeling approach has the potential to drive advancements in autonomous navigation, urban planning, and informed decision-making for robotic deployments in urban environments. Collectively, these projects underscore the versatility and value of neural fields as cutting-edge computational tools for advancing the field of robotics.