Standing Balance Control of a Bipedal Robot Based on Behavior Cloning

Abstract

Bipedal robots have gained increasing attention for their human-like mobility which allows them to work in various human-scale environments. However, their inherent instability makes it difficult to control their balance while they are physically interacting with the environment. This study proposes a novel balance controller for bipedal robots based on a behavior cloning model as one of the machine learning techniques. The behavior cloning model employs two deep neural networks (DNNs) trained on human-operated balancing data, so that the trained model can predict the desired wrench required to maintain the balance of the bipedal robot. Based on the prediction of the desired wrench, the joint torques for both legs are calculated using robot dynamics. The performance of the developed balance controller was validated with a bipedal lower-body robotic system through simulation and experimental tests by providing random perturbations in the frontal plane. The developed balance controller demonstrated superior performance with respect to resistance to balance loss compared to the conventional balance control method, while generating a smoother balancing movement for the robot.

A Survey on Deep Reinforcement Learning Algorithms for Robotic Manipulation

“…A new balancing controller for bipedal robots based on a behavior cloning model is presented by Bong et al [ 110 ]. The controller first predicts the wrench needed to keep the bipedal robot balanced using two deep neural networks trained on data from human-operated balancing devices and then utilizes robot dynamics to compute joint torques for both legs.…”

Evolution, Design, and Future Trajectories on Bipedal Wheel-legged Robot: A Comprehensive Review

“…Advanced methods like decoupling control, Electrical-Jet assistance, and robotic jumping have improved dynamic balance capabilities [7], [32], [85]. Compliant walking controllers enhance navigation in challenging terrains [86], bridging the gap between robots and authentic bipedal locomotion for enhanced stability and efficiency [87]. Features such as the six degree of freedom (DOF) parallel link mechanism improve leg inertia and position tracking at higher frequencies [42], [87].…”