Expensive multiobjective optimization for robotics with consideration of heteroscedastic noise
Expensive multiobjective optimization for robotics with consideration of heteroscedastic noise
复制标题
DOI：
10.1109/iros.2014.6942863
复制
发表时间：
2014-11
期刊：
影响因子：
--
通讯作者：
Ryo Ariizumi;M. Tesch;H. Choset;F. Matsuno
中科院分区：
文献类型：
--
作者：
Ryo Ariizumi;M. Tesch;H. Choset;F. Matsuno
在许多机器人问题中，需要针对多个冲突标准优化策略。然而，这是非常具有挑战性的，由于噪声的存在，这可能是输入相关的，或异方差的，以及由于机器人实验，这是昂贵的时间和/或金钱的评估的数量的限制。本文提出了一种多目标优化（MOO）算法昂贵的机器人噪声函数的评估。我们提出了一种异方差和标准同方差高斯过程回归技术之间的模型选择方法，从噪声样本中创建合适的代理函数，并找到下一步要观察的点。该算法与现有的MOO算法，假设同方差噪声进行比较，然后使用优化的速度和头部稳定性的侧绕步态的蛇机器人。
In many robotic problems, optimization of the policy for multiple conflicting criteria is required. However this is very challenging due to the existence of noise, which may be input dependent, or heteroscedastic, and the restriction in the number of evaluations, due to robotic experiments which are expensive in time and/or money. This paper presents a multiobjective optimization (MOO) algorithm for expensive-to-evaluate noisy functions for robotics. We present a method for model selection between heteroscedastic and standard homoscedastic Gaussian process regression techniques to create suitable surrogate functions from noisy samples and find the point to be observed at the next step. This algorithm is compared against an existing MOO algorithm which assumes homoscedastic noise, and is then used to optimize the speed and head stability of the sidewinding gait of a snake robot.