
Birmingham researchers use artificial intelligence to advance the dexterity of robots
Link to the article about our GPAtlasRRT algorithm for tactile recovering of the shape of novel objects. https://www.birmingham.ac.uk/university/colleges/eps/news/2018/7/birmingham-researchers-artificial-intelligence-advance-dexterity-robots.aspx #rrt #gp #atlas #robotics #robot #tactile #exploration #touch #novelobjects #vito #boris #softhand #humanoid #ai #artificialintelligence #dexterity #universityofbirmingham #irlab #uob #pisa

Feature-Based Transfer Learning for Robotic Push Manipulation
This work presents a data-efficient approach to learning transferable forward models for robotic push manipulation. Our approach extends previous work on modular contact-based predictors by leveraging information on the pushed object's local surface features. We test the hypothesis that, by conditioning predictions on local surface features, we can achieve generalisation across objects of different shapes. Our approach involves learning motion models that are specific to con