论文部分内容阅读
联结CPG(connectionist central pattern generator,CCPG)模型适于控制机器人生成步态,但是传统的CCPG模型无法很好地生成3维步态.为此,本文根据生物学原理,提出了一个改进的神经元模型和一个改进的层次化CCPG(hierarchical CCPG,HCCPG)模型.HCCPG模型能够生成相位协调的多自由度运动控制信号,从而解决了传统CCPG模型的步态生成问题.基于该模型,提出了一个统一方法来生成机器人的2维、3维步态.对转弯步态的特性进行了系统化深入分析,以便更好地利用该步态来适应狭窄的弯道环境.本文提出的HCCPG模型以及得到的步态特性,有助于提高机器人的环境适应能力.
The coupled CCPG (CCPG) model is suitable for controlling the robot to generate gait, but the traditional CCPG model can not generate 3-dimensional gait well.Therefore, based on the biological principle, this paper presents an improved neuron Model and an improved hierarchical CCPG (HCCPG) model.The HCCPG model can generate phase-coordinated multi-degree-of-freedom motion control signals to solve the gait generation problem of the traditional CCPG model.Based on this model, a unified Method to generate the 2-dimensional and 3-dimensional gait of the robot.A systematic and in-depth analysis of the characteristics of turning gait in order to make better use of this gait to adapt to the narrow curve environment.The proposed HCCPG model and the obtained Gait characteristics, help to improve the robot’s ability to adapt to the environment.