首页> 外国专利> A learning method and a learning apparatus for integrating the space detection result of another autonomous vehicle with the space detection result of the own autonomous vehicle acquired by V2V communication, and a test method and a test apparatus using the learning method and the learning apparatus. DEVICE FOR INTEGRATING OBJECT DETECTION INFORMATION ACQUIRED THROUGH V2V COMMUNICATION FROM OTHER AUTONOMOUS VEHICLE WITH OBJECT DETECTION INFORMATION GENERATED BY PRESENT AUTONOMOUS VEHICLE, AND TESTING METHOD AND TESTING DEVICE USING THE SAME}

A learning method and a learning apparatus for integrating the space detection result of another autonomous vehicle with the space detection result of the own autonomous vehicle acquired by V2V communication, and a test method and a test apparatus using the learning method and the learning apparatus. DEVICE FOR INTEGRATING OBJECT DETECTION INFORMATION ACQUIRED THROUGH V2V COMMUNICATION FROM OTHER AUTONOMOUS VEHICLE WITH OBJECT DETECTION INFORMATION GENERATED BY PRESENT AUTONOMOUS VEHICLE, AND TESTING METHOD AND TESTING DEVICE USING THE SAME}

机译:一种学习方法和学习设备,用于将另一辆自动驾驶汽车的空间检测结果与通过V2V通信获取的自己的自动驾驶汽车的空间检测结果进行积分,以及一种使用该学习方法和学习设备的测试方法和测试设备。将通过V2V通信从其他自治车辆获取的目标检测信息与当前自主车辆生成的目标检测信息集成在一起的装置,测试方法和测试装置,使用相同的方法

摘要

PROBLEM TO BE SOLVED: To provide a learning method, a testing method, a learning device, and a testing device capable of mutually sharing a space detection result using CNN by V2V technology. SOLUTION: The learning method includes a step in which a learning device has a concatenation network 210 to generate one or more pair feature vectors, and a learning device has a discrimination network 220 to apply an FC operation to the pair feature vectors. , A step of generating a discriminant vector and a box regression vector, and the learning device has a loss unit 230 to generate an integrated loss by referring to the discriminant vector, the box regression vector, and the GT corresponding thereto, and use the integrated loss. And performing at least some of the parameters included in the DNN by performing back propagation. [Selection diagram] Figure 2
机译:解决的问题:提供一种能够通过V2V技术使用CNN相互共享空间检测结果的学习方法,测试方法,学习设备和测试设备。解决方案:学习方法包括以下步骤:学习设备具有连接网络210来生成一个或多个对特征向量,学习设备具有判别网络220以便将FC操作应用于对特征向量。 ,生成判别向量和框回归向量的步骤,并且学习设备具有损失单元230,以通过参考判别向量,框回归向量和与其对应的GT来生成积分损失,并使用积分失利。并且通过执行反向传播来执行DNN中包含的至少一些参数。 [选择图]图2

著录项

相似文献

  • 专利
  • 外文文献
获取专利

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号