A Bioinspired Advanced Neural Control for Autonomous Walking Robots
Researchers from BRAIN-lab at VISTEC in Thailand developed a bioinspired advanced neural control with proactive behavior learning and short-term memory for autonomous walking robots that enables them to traverse complex terrain.
The proposed control consists of three main modular neural mechanisms to create insect-like gaits, adapt robot joint movement individually with respect to the terrain during the stance phase using only the torque feedback, and generate a short-term memory for proactive obstacle avoidance during the swing phase.
Their proposed control:
- does not require robot and environmental models, exteroceptive feedback, or multiple learning trials
- depends only on proprioceptive feedback and short-term memory
- is general (no robot model is needed), and it can be used for autonomous locomotion of other walking robots