Implementation of Deep Deterministic Policy Gradients for Controlling Dynamic Bipedal Walking
Implementation of Deep Deterministic Policy Gradients for Controlling Dynamic Bipedal Walking
复制标题
DOI：
10.3390/biomimetics4010028
复制
发表时间：
2019-03-22
期刊：
影响因子：
4.5
通讯作者：
Quinn, Roger D.
中科院分区：
文献类型：
--
作者：
Liu, Chujun;Lonsberry, Andrew G.;Quinn, Roger D.
仿真研究了双足机器人矢状面行走控制系统。该双足模型是根据身高1.8米、平均体型的男性的人体测量数据建立的。控制器的核心是一个深度确定性策略梯度（DDPG）神经网络，该网络在物理模拟器GAZEBO中训练，以预测理想的足部位置，以在外部干扰下保持稳定的行走。通过精心选择状态变量和分布式控制系统，降低了DDPG网络的复杂性。在站立阶段的髋关节和脱趾阶段的踝关节的额外控制器有助于在行走期间稳定双足动物。模拟的双足动物可以以大约1m /s的稳定速度行走，并且在运动过程中，躯干向前施加30 kg中心点m/s的脉冲或向后施加40 kg中心点m/s的脉冲时，它可以保持稳定。它也保持稳定行走与一个10公斤的背包或25公斤的前包。控制器在一个1.8米高的模型上训练，但也稳定了1.4-2.3米高的模型，没有变化。
A control system for bipedal walking in the sagittal plane was developed in simulation. The biped model was built based on anthropometric data for a 1.8 m tall male of average build. At the core of the controller is a deep deterministic policy gradient (DDPG) neural network that was trained in GAZEBO, a physics simulator, to predict the ideal foot placement to maintain stable walking despite external disturbances. The complexity of the DDPG network was decreased through carefully selected state variables and a distributed control system. Additional controllers for the hip joints during their stance phases and the ankle joint during toe-off phase help to stabilize the biped during walking. The simulated biped can walk at a steady pace of approximately 1 m/s, and during locomotion it can maintain stability with a 30 kg center dot m/s impulse applied forward on the torso or a 40 kg center dot m/s impulse applied rearward. It also maintains stable walking with a 10 kg backpack or a 25 kg front pack. The controller was trained on a 1.8 m tall model, but also stabilizes models 1.4-2.3 m tall with no changes.