Multi-expert learning of adaptive legged locomotion

Scientists from the University of Edinburgh (UK) and Zhejiang University (China) present a hierarchical motor control using deep reinforcement learning. It is called Multi-Expert Learning Architecture - MELA - which fuses multiple expert neural networks into a synthesized expert and learns versatile motor skills. MELA can generate adaptive motor behaviors across different locomotion modes and survive in unknown situations, which is essential for truly autonomous robots to operate effectively in the wild. F