Reinforcement learning for long-horizon robot manipulation mapped in 130-study review
A new scoping review of 130 studies finds that reinforcement learning for long-horizon robotic manipulation is dominated by hierarchical methods, ...
Tag: Robotic Manipulation
Robots That Learn From Old Data: Review Maps Offline Reinforcement Learning Under Data Scarcity
A new narrative review maps how offline reinforcement learning methods behave when robotic manipulation datasets are small, narrow and noisy.
Greedy Sampling Shrinks the Search: New Planner Speeds Robot Path Optimization
Researchers have introduced Greedy RRT*, a sampling-based motion planner that uses a greedy informed set to shrink the search region ...
Robots Learn to Feel Pressure Through Cameras Alone, Thanks to Multimodal AI
A new multimodal AI framework fuses RGB images, segmentation masks, depth maps, and text prompts to estimate robotic gripper pressure ...
Tiny AI brain lets a $10 microcontroller remember hidden objects and grab them
Researchers have built Tiny-STM, a 68.4-kilobyte LSTM controller with a simple coordinate memory that lets an ESP32 microcontroller track objects ...
Robots Master Grasping in Cluttered Environments
In a groundbreaking study published in the journal “Autonomous Robots,” researchers led by S. Lachhiramka, along with colleagues Pradeep J ...
