Results 111 to 120 of about 7,404,725 (288)

Safe and Efficient Off-Policy Reinforcement Learning

CoRR, 2016
In this work, we take a fresh look at some old and new algorithms for off-policy, return-based reinforcement learning. Expressing these in a common form, we derive a novel algorithm, Retrace($λ$), with three desired properties: (1) it has low variance; (2) it safely uses samples collected from any behaviour policy, whatever its degree of "off ...
Munos, Remi, Stepleton, Tom, Harutyunyan, Anna, Bellemare, Marc G.   +3 more
openaire   +4 more sources

Success-efficient/failure-safe strategy for hierarchical reinforcement motor learning.

PLoS Computational Biology
Our study explores how ecological aspects of motor learning enhance survival by improving movement efficiency and mitigating injury risks during task failures.
Jan Babič, Tjasa Kunavar, Erhan Oztop, Mitsuo Kawato   +3 more
doaj   +1 more source

Compliant Pneumatic Feet with Real‐Time Stiffness Adaptation for Humanoid Locomotion

Advanced Robotics Research, EarlyView.
A compliant pneumatic foot with real‐time variable stiffness enables humanoid robots to adapt to changing terrains. Using onboard vision and pressure control, the foot modulates stiffness within each gait cycle, reducing impact forces and improving balance. The design, cast in soft silicone with embedded air chambers and Kevlar wrapping, offers durable,
Irene Frizza, Hiroshi Kaminaga, Philippe Fraisse, Gentiane Venture   +3 more
wiley   +1 more source

Robotic Control for Human–Robot Collaborative Assembly Based on Digital Human Model and Reinforcement Learning

Advanced Robotics Research, EarlyView.
This work presents a robotic control method for human–robot collaborative assembly based on a biomechanics‐constrained digital human model. Reinforcement learning is used to generate physiologically plausible human motion trajectories, which are integrated into a virtual environment for robot control learning.
Bitao Yao, Hongzhou Ai, Wenjun Xu, Zude Zhou, Lihui Wang   +4 more
wiley   +1 more source

Multimodal Human–Robot Interaction Using Human Pose Estimation and Local Large Language Models

Advanced Robotics Research, EarlyView.
A multimodal human–robot interaction framework integrates human pose estimation (HPE) and a large language model (LLM) for gesture‐ and voice‐based robot control. Speech‐to‐text (STT) enables voice command interpretation, while a safety‐aware arbitration mechanism prioritizes gesture input for rapid intervention.
Nasiru Aboki, Ilche Georgievski, Marco Aiello   +2 more
wiley   +1 more source

A Multi-Objective Collaborative Optimization Approach for Building Integrated Energy Systems Based on Deep Reinforcement Learning

Applied Sciences
To address the challenges of coordinated optimization in building integrated energy systems (IES) under the dual-carbon targets—characterized by strong multi-energy coupling, significant uncertainty in renewable generation, and stringent safety ...
Limin Wang, Yongkai Wu, Jumin Zhao, Wei Gao, Dengao Li   +4 more
doaj   +1 more source

Hybrid Continuum Robot Designs and Architectures for Healthcare Applications

Advanced Robotics Research, EarlyView.
Hybrid continuum robots represent an emerging class of flexible manipulators that blend materials, structures, and actuation concepts from the established fields of soft and continuum robotics. This review introduces an accessible framework to distinguish key hybridization approaches, surveys current designs aimed at complex clinical applications, and ...
Burak Ozdemir, Muhammad A. Khalid, Manish Chauhan, Pietro Valdastri, James H. Chandler   +4 more
wiley   +1 more source

Artificial Intelligence Enabled Future Wireless Electric Vehicles with Multi-Model Learning and Decision Making Models

Tsinghua Science and Technology
In the contemporary era, driverless vehicles are a reality due to the proliferation of distributed technologies, sensing technologies, and Machine to Machine (M2M) communications.
Gajula Ramesh, Anil Kumar Budati, Shayla Islam, Louai A. Maghrabi, Abdullah Al-Atwai   +4 more
doaj   +1 more source

LLM‐Integrated Human–Robot Interaction System for Microrobots

Advanced Robotics Research, EarlyView.
This paper proposes an LLM‐based control framework for guiding microrobots using human natural language. This framework can convert the natural human speech into safe and executable command sets for reliable navigation in complex environments. The experimental results show high accuracy and robustness in task performance, demonstrating the potential of
Bairong Zhu, Amar Salehi, Tingting Yu
wiley   +1 more source

Safe reinforcement learning in automotive

, 2020
This work applies Logically Constrained Reinforcement Learning (LCRL) frame-work for synthesizing policies for active monitoring and management of sensor modules in autonomous vehicles. LCRL allows synthesizing policies for unknown and continuous-state Markov Decision Processes (MDPs) such that a given linear time property is satisfied.
openaire   +2 more sources