Given a timed automata which admits thick components and a timed word $x$, we present a tester which decides if $x$ is in the language of the automaton or if $x$ is $\epsilon$-far from the language, using finitely many samples taken from the weighted time distribution $\mu$ associated with an input $x$. We introduce a distance between timed words, the {\em timed edit distance}, which generalizes the classical edit distance. A timed word $x$ is $\epsilon$-far from a timed language if its relative distance to the language is greater than $\epsilon$.

翻译：给定一个允许厚分支的时间自动机以及一个时间词$x$，我们提出一种测试器，该测试器通过与输入$x$相关的加权时间分布$\mu$中提取的有限样本，判定$x$是否属于该自动机的语言，或者$x$是否与该语言$\epsilon$-远离。我们引入了一种时间词之间的距离——{\em时间编辑距离}，它推广了经典的编辑距离。一个时间词$x$相对于时间语言是$\epsilon$-远离的，当且仅当其与该语言的相对距离大于$\epsilon$。