GEneric NEural control for Self-organized emergent behavIor of limbed Systems (GENESIS, Smart motor control software)
Flexible, Soft Object Manipulation
Online Neural Learning and Adaptation
Bio-inspired Robot Structure Design with Hybrid Rigid-Soft Material
Neural control for autonomous climbing and obstacle avoidance of a gecko-inspired climbing robot on steep slopes.
Autonomous online adaptation of a walking robot under bioinspired adaptive locomotion control.
Novel hybrid soft-rigid foot with dry adhesive material for a gecko-inspired climbing robot.
Small-sized and lightweight quadruped robot serving as a generic robot platform for research and education in the fields of robot locomotion, bionic control, and machine learning.
For more details, see Sun et al., Front. Neurorobot., 2020
Novel fast online learning based on error feedback for adaptive robot motor control.
For more details, see Thor and Manoonpong, IEEE Transactions on Neural Networks and Learning Systems, 2019
Exploiting neural dynamics of a reservoir computing-based recurrent neural network and haptic feedback to classify multiple terrains.
For more details, see Borijindakul et al., Lecture Notes in Computer Science, 2019
A bio-inspired climbing robot with flexible pads and claws that can climb on rough walls.
For more details, see Ji et al., J Bionic Eng, 2018
Adaptive neural control for self-organized locomotion and obstacle negotiation of quadruped robots.
For more details, see Sun et al., IEEE International Symposium on Robot and Human Interactive Communication, 2018 [Video1, Video2]