Latest Articles for Robotics

A new method uses ultrasound to recognize actions while protecting privacy.

A new method helps identify and address failures in automated systems using simulations.

FindingEmo offers insights into recognizing emotions through social scenes.

Discover how proximal Bellman mappings enhance decision-making in reinforcement learning.

A new PD controller enhances the performance of flexible joint robots.

Winged UAVs can now safely perch on vertical surfaces using natural mechanics.

A new method combines learning and safety for smarter robot control.

A flexible method for creating unique indoor environments through text descriptions.

Integrating vision-language models with reinforcement learning improves machine learning efficiency.

Exploring new methods to improve image segmentation and instance detection.

Researchers develop bipedal robots that learn walking through practice and animal movements.

A look into how robots are evolving with new technology.

This article examines how observation type impacts robot task learning.

A new framework enhances LLMs' decision-making by leveraging past experiences.

A look at the challenges in assessing RL agents amid changing environments.

A novel method enhances Vision-and-Language Navigation using training data from driving videos.

A new method enhances robots' ability to follow spoken directions accurately.

A new method to enhance depth completion using sparse depth data during testing.

A computer vision method that detects objects efficiently without human input.

A platform for AI agents to interact with real environments using geospatial data.

A new method for safely controlling unpredictable systems.

SGS-SLAM improves 3D mapping accuracy and speed using Gaussian representations.

New methods improve safety of systems in uncertain environments.

Understanding the role of time in automated systems enhances their functionality.

A new method enhances how agents explore complex three-dimensional spaces.

A new approach enhances Hamiltonian system modeling despite noisy data.

Robots can learn human preferences better through behavior and language analysis.

Learn how optimal transport enhances data movement and analysis in various fields.

Research reveals optimal swimming techniques for better thrust and speed.

New methods enhance energy efficiency and performance for spiking neural networks.

A new system allows data scientists to train robots without complex software.

A new method improves decision-making under uncertainty using probability measures.

A method combining quick decision-making with diverse robot behaviors.

Research examines the use of VLMs to assess robot actions.

New method for safe robotic exploration while learning effectively.

A new method enhances safety for robots navigating complex environments.

A method to improve cooperation in multi-agent systems with sparse rewards.

Introducing a new method for detecting errors in robotic actions during few-shot imitation.

SEABO generates rewards from expert data, simplifying offline imitation learning.

New algorithms tackle challenges in continuous state-action spaces of Reinforcement Learning.