张会文, 张伟, 周维佳. 基于交叉熵优化的高斯混合模型运动编码[J]. 机器人, 2018, 40(4): 569-576. DOI: 10.13973/j.cnki.robot.180146
引用本文: 张会文, 张伟, 周维佳. 基于交叉熵优化的高斯混合模型运动编码[J]. 机器人, 2018, 40(4): 569-576. DOI: 10.13973/j.cnki.robot.180146
ZHANG Huiwen, ZHANG Wei, ZHOU Weijia. Encoding Motor Skills with Gaussian Mixture Models Optimized bythe Cross Entropy Method[J]. ROBOT, 2018, 40(4): 569-576. DOI: 10.13973/j.cnki.robot.180146
Citation: ZHANG Huiwen, ZHANG Wei, ZHOU Weijia. Encoding Motor Skills with Gaussian Mixture Models Optimized bythe Cross Entropy Method[J]. ROBOT, 2018, 40(4): 569-576. DOI: 10.13973/j.cnki.robot.180146

基于交叉熵优化的高斯混合模型运动编码

Encoding Motor Skills with Gaussian Mixture Models Optimized bythe Cross Entropy Method

  • 摘要: 针对模仿学习中运动的表征和泛化问题,提出了交叉熵优化算法,用于混合模型参数的推断.该算法易于实施、计算效率高.更重要的是,它能够自动确定混合模型中最优成分的个数.为了产生泛化的运动轨迹,提出了交叉熵回归算法.为了进一步提高这种算法对动态环境的适应能力,引入了任务参数化的概念并提出了任务参数交叉熵回归算法.最后设计了一个新颖的锤击任务,验证了所提出的算法在理论上的正确性和优越性.基于机器人物理仿真软件Gazebo的仿真实验表明了算法在实际应用中的可行性.

     

    Abstract: Aiming at the movement representation and generalization problems in imitation learning, a cross entropy optimization algorithm is proposed to infer parameters in mixture models. The proposed algorithm is easy to implement and computationally efficient. More importantly, it can automatically determine the optimal component number in the mixture models. In order to produce generalized motion trajectories, a cross entropy regression algorithm is proposed. To further improve the adaptability of the algorithm in dynamic environments, the concept of task parametrization is introduced and a task-parameterized cross entropy regression algorithm is proposed. Finally, a novel hammer-over-a-nail task is designed, which verifies the theoretical correctness and superiority of the proposed methods. Simulation experiments based on robot physical simulation software Gazebo show the feasibility of the proposed algorithms in piratical applications.

     

/

返回文章
返回