CARL adds stability guarantees to reinforcement-learning robot control
A reinforcement-learning method combining contraction theory was tested in simulation and on three robot platforms, providing stability guarantees and improving ...
A reinforcement-learning method combining contraction theory was tested in simulation and on three robot platforms, providing stability guarantees and improving ...
A new meta-reinforcement learning framework enables quadruped robots to maintain stable locomotion even when multiple joints lock simultaneously, dramatically outperforming ...
A new analysis argues that solid-state batteries could transform robotics, revealing that a single commercial robot consumes an EV-equivalent 60 ...
Researchers combined artificial potential field trajectory planning with a discrete integral terminal sliding mode controller to let a Mecanum-wheeled omnidirectional ...
Researchers developed lightweight neural networks and a new RGB-D dataset that let robots detect accurate grasps for fruits of varying ...
A comprehensive new review traces how diffusion models, the generative engines behind modern image synthesis, have become the dominant force ...
Researchers at DFKI and the University of Bremen present new pytransform3d features for transformation graphs, uncertainty propagation, and 3D visualization, ...
The Rubik's Cube has long served as a demanding benchmark for autonomous systems precisely because it compresses so many difficulties ...
A systematic review of 81 studies finds that artificial intelligence is reducing infectious, toxic, and occupational health risks across the ...
A robotic system that imitates human perception of raised fabric structures and fuzzy decision-making achieves a 93.3 percent grasp success ...
© 2025 Scienmag - Science Magazine
© 2025 Scienmag - Science Magazine