Learning highly diverse robot throwing movements through quality diversity search
Learning highly diverse robot throwing movements through quality diversity search
复制标题
通过质量多样性搜索学习高度多样化的机器人投掷动作
DOI：
--
复制
发表时间：
2017
期刊：
影响因子：
--
通讯作者：
S. Doncieux
中科院分区：
文献类型：
--
作者：
Seungsu Kim;S. Doncieux
机器人的行为取决于它所处的环境。进化方法可以适用于寻找对变化的条件[4，10]稳健的控制器，但很难保证单一的行为将适应任何新的情况。另一种方法是建立一个行为库，从中寻找适应上下文的行为[1，2]。拥有大量的行为集而不是单一的行为集进一步允许提取关于机器人、环境和任务的信息，从而为自建对机器人有意义的理解开辟了道路[3]，并为获得可以更快、更有效地解决新任务的更高级别的表示[14]开辟了道路。
The behavior of a robot depends on its environment. Evolutionary approaches can be adapted to look for controllers that are robust to changing conditions [4, 10], but it is hard to guarantee that a single behavior will adapt to any new situation. An alternative approach consists in building a repertoire of behaviors in which to look for behaviors adapted to the context [1, 2]. Having a large set of behaviors instead of a single one further allows for extracting information about the robot, the environment and the task, thus opening the way to a self-built understanding of what makes sense for the robot [3] and to the acquisition of higher level representations that can solve new tasks faster and more efficiently [14].