Efficient Learning of Control Policies for Robust Quadruped Bounding using Pretrained Neural Networks
This work addresses efficient and robust locomotion control for quadruped robots, but it is incremental as it builds on existing pretraining and reinforcement learning methods.
The authors tackled the problem of learning robust bounding gaits for quadruped robots by pretraining a neural network with model-based controller data and optimizing it via deep reinforcement learning, achieving successful deployment on a real robot with efficient computing and good locomotion over uneven terrain.
Bounding is one of the important gaits in quadrupedal locomotion for negotiating obstacles. The authors proposed an effective approach that can learn robust bounding gaits more efficiently despite its large variation in dynamic body movements. The authors first pretrained the neural network (NN) based on data from a robot operated by conventional model based controllers, and then further optimised the pretrained NN via deep reinforcement learning (DRL). In particular, the authors designed a reward function considering contact points and phases to enforce the gait symmetry and periodicity, which improved the bounding performance. The NN based feedback controller was learned in the simulation and directly deployed on the real quadruped robot Jueying Mini successfully. A variety of environments are presented both indoors and outdoors with the authors approach. The authors approach shows efficient computing and good locomotion results by the Jueying Mini quadrupedal robot bounding over uneven terrain.