Meghan Booker
YOU?
Author Swipe
View article: FLEET: Formal Language-Grounded Scheduling for Heterogeneous Robot Teams
FLEET: Formal Language-Grounded Scheduling for Heterogeneous Robot Teams Open
Coordinating heterogeneous robot teams from free-form natural-language instructions is hard. Language-only planners struggle with long-horizon coordination and hallucination, while purely formal methods require closed-world models. We pres…
View article: Constrained Natural Language Action Planning for Resilient Embodied Systems
Constrained Natural Language Action Planning for Resilient Embodied Systems Open
Replicating human-level intelligence in the execution of embodied tasks remains challenging due to the unconstrained nature of real-world environments. Novel use of large language models (LLMs) for task planning seeks to address the previo…
View article: EmbodiedRAG: Dynamic 3D Scene Graph Retrieval for Efficient and Scalable Robot Task Planning
EmbodiedRAG: Dynamic 3D Scene Graph Retrieval for Efficient and Scalable Robot Task Planning Open
Recent advances in Large Language Models (LLMs) have helped facilitate exciting progress for robotic planning in real, open-world environments. 3D scene graphs (3DSGs) offer a promising environment representation for grounding such LLM-bas…
View article: ConceptAgent: LLM-Driven Precondition Grounding and Tree Search for Robust Task Planning and Execution
ConceptAgent: LLM-Driven Precondition Grounding and Tree Search for Robust Task Planning and Execution Open
Robotic planning and execution in open-world environments is a complex problem due to the vast state spaces and high variability of task embodiment. Recent advances in perception algorithms, combined with Large Language Models (LLMs) for p…
View article: Perceive With Confidence: Statistical Safety Assurances for Navigation with Learning-Based Perception
Perceive With Confidence: Statistical Safety Assurances for Navigation with Learning-Based Perception Open
Rapid advances in perception have enabled large pre-trained models to be used out of the box for transforming high-dimensional, noisy, and partial observations of the world into rich occupancy representations. However, the reliability of t…
View article: Online Learning for Obstacle Avoidance
Online Learning for Obstacle Avoidance Open
We approach the fundamental problem of obstacle avoidance for robotic systems via the lens of online learning. In contrast to prior work that either assumes worst-case realizations of uncertainty in the environment or a stationary stochast…
View article: Switching Attention in Time-Varying Environments via Bayesian Inference of Abstractions
Switching Attention in Time-Varying Environments via Bayesian Inference of Abstractions Open
Motivated by the goal of endowing robots with a means for focusing attention in order to operate reliably in complex, uncertain, and time-varying environments, we consider how a robot can (i) determine which portions of its environment to …
View article: Learning to Actively Reduce Memory Requirements for Robot Control Tasks
Learning to Actively Reduce Memory Requirements for Robot Control Tasks Open
Robots equipped with rich sensing modalities (e.g., RGB-D cameras) performing long-horizon tasks motivate the need for policies that are highly memory-efficient. State-of-the-art approaches for controlling robots often use memory represent…
View article: Effects of Hacking an Unmanned Aerial Vehicle Connected to the Cloud
Effects of Hacking an Unmanned Aerial Vehicle Connected to the Cloud Open
Control systems with commercial and even military applications are utilizing more networked technologies to perform tasks associated with navigation and communication. Increasingly, these systems are experiencing cyber-attacks due to the i…