Latest Articles for Autonomous Systems

A method for developing resilient policies for drones in uncertain environments.

A novel approach using hypergraphs improves predictions of agent trajectories in various environments.

This study explores a new method for optimizing multiple goals in self-adaptive systems.

A multirotor UAV designed for safe landings despite rotor failures represents a major step forward.

A look into Enhanced Outlier Logit method for better class recognition.

PointNCBW offers a reliable way to verify ownership of point cloud datasets.

A new method improves language agents' decision-making through self-reflection.

Explore the role of Control Barrier Functions in maintaining safety across various systems.

A new approach to reinforce learning efficiency using heavy-ball momentum in actor-critic methods.

A new system helps indoor blimps navigate autonomously using visual localization.

Integrating ontologies enhances planning in uncertain and complex environments.

Caution-Aware Transfer enhances safety and performance in reinforcement learning applications.

A new method improves the safety verification of neural networks in critical systems.

New strategies improve learning efficiency without frequent resets.

A new approach enhances uncertainty quantification in machine learning models.

A new approach to improve model accuracy amidst label noise.

Addressing domain adaptation issues to improve 3D object detection systems.

A technique to help robots recover from attacks while staying mission compliant.

Discover how hybrid control systems tackle challenges in autonomous technology.

New methods improve tracking accuracy and consistency in computer vision applications.

This article discusses the role of LLMs in spacecraft missions using the Kerbal Space Program.

A new system using event cameras enhances robot navigation and mapping accuracy.

ConVis aims to minimize inaccuracies in multimodal large language models.

A new navigation method improves robot safety in crowded environments.

INK provides a reliable method for identifying out-of-distribution samples in machine learning.

A new method enhances object detection performance by addressing classification and localization challenges.

A new approach helps AUVs make timely decisions underwater.

A method to help diverse agents work together effectively in uncertain environments.

Innovative methods enhance object detection performance on resource-limited devices.

A new machine learning method prioritizes safety in autonomous systems.

This study assesses various visual models for understanding complex 3D scenes.

New method improves safety and performance in autonomous technologies.

StratXplore enhances robot navigation through improved error recovery strategies.

A new framework enhances object detection by identifying out-of-distribution instances using prototypes.

A new library enhances MPPI algorithms for robotics applications.

New tool enhances spacecraft navigation and tracking in cislunar space.

A comprehensive dataset aimed at improving autonomous taxiing for aircraft safety.

An overview of ICRL and its role in learning expert behavior.

A look at using RL to train agents for maritime cyber security.

A new approach to network packet scheduling for time-sensitive applications.