Latest Articles for Autonomous Systems

A new method enhances learning from environments in visual reinforcement systems.

New method enhances reliability of predictions by addressing uncertainty in machine learning.

A look into determining safe movement areas for autonomous systems with unknown behaviors.

A new method improves how robots assess their surroundings with minimal labeling.

ISAC combines sensing and communication for better wireless technology.

Exploring fine-grained causal relationships to enhance decision-making in machine learning.

New method enhances uncertainty quantification in adversarially trained models.

Exploring the complexities of aligning AI systems with diverse human interests.

MASA learns object tracking using unlabeled images, improving adaptability in diverse situations.

This article examines how language models improve robot communication and interaction.

WGBoost offers advanced probabilistic predictions and better uncertainty handling in various fields.

A new approach enhances model reliability in crucial predictions.

Exploring the impact of ISAC and CF-mMIMO on communication and sensing technologies.

A look at errors in SLAM and the role of Jacobians in optimization.

New methods improve accuracy in depth estimation using synthetic and real-world data.

Learn how autonomous systems plan actions effectively over time.

ABTrack enhances visual tracking speed and efficiency across various devices.

A new technique enhances neural network prediction reliability through geometric adjustments.

A guide to detecting satellite problems during lunar and Martian missions.

A new method improves machine learning model adaptability in dynamic situations.

A new method for teaching machines through videos despite visual differences.

A new method improves safety for drones using resilient estimation and control barrier functions.

Visual Overlap Prediction improves image retrieval accuracy and efficiency in complex environments.

New methods enhance drone navigation and safety in complex environments.

Robots cooperate using only visual input, enhancing movement and coordination.

A new approach enhances decision-making in dynamic environments using continuous-time reinforcement learning.

A new method helps identify weak points in deep learning models quickly.

Introducing methods for safer decision-making in reinforcement learning environments.

Study reveals methods for efficient target search by mobile agents in circular areas.

TADPoLe trains agents using text-based rewards for natural task execution.

New methods improve model recognition across varied data types.

New methods improve detection of small objects in computer vision.

Enhancing prediction sets when data conditions shift without needing labels.

A new method improves motion planning efficiency in robotics and autonomous systems.

BlackVIP enables efficient use of AI models without accessing their internals.

Grad-CAPS improves action smoothness in reinforcement learning for robotics and autonomous systems.

An overview of the interactions in Cyber Physical Systems and their implications.

A new merging algorithm improves 3D instance segmentation accuracy significantly.

Enhancing detection with RGB and depth images to tackle real-world challenges.

A new method enhances robot navigation by merging visual and geometric data.