英语新闻听写 研究人员发现自动驾驶汽车技术中的错误（在线收听）

As autonomous vehicles become more popular, some researchers remain worried about potential safety issues.

随着自动驾驶汽车越来越受欢迎，一些研究人员仍然担心潜在的安全问题。

Digital Trends reports researchers from Massachusetts Institute of Technology (MIT) have been looking at the differences between autonomous systems learning in training and real world issues.

麻省理工学院的研究人员一直在研究自主系统在训练与现实问题之间的差异。

A model by MIT and Microsoft researchers is testing what the cars will do when the training they have "learned" doesn't match what is happening on the road.

麻省理工学院和微软研究人员制作的模型正在测试：当汽车“学到"的训练与实际道路情况不匹配时，汽车会发生什么情况。

Author of the paper Ramya Ramakrishnan said in a statement.

论文作者Ramya Ramakrishnan表示：

"Many times, when these systems are deployed, their trained simulations don't match the real-world setting and they could make mistakes, such as getting into accidents."

“很多时候，当系统部署之后，模拟训练与现实环境不匹配时，汽车可能会出错，比如发生事故。”