首页>
外国专利>
Method for training and testing an algorithm for predicting agents in a vehicle environment
Method for training and testing an algorithm for predicting agents in a vehicle environment
展开▼
机译:车辆环境中预测智能体算法的训练和测试方法
展开▼
页面导航
摘要
著录项
相似文献
摘要
The invention relates to a method for training and testing an algorithm for predicting agents in a vehicle environment, which is carried out by means of a machine trained trajectory autoencoder (1). After training the trajectory autoencoder (1), similar scenes (S1 to Sn) are found using the trajectory autoencoder (1) for a complete data set (D) consisting of all scenes (S1 to Sn) to be searched for similar scenes (S1 to Sm),latent representations (R1 to Rn) generated. Furthermore, a scene to be searched for (S),for which similar scenes (S1 to Sm) are to be found in the dataset (D), coded and for the scene to be searched (S) a latent representation (R) is formed. Using a similarity metric (3), the latent representation (R) of the scene to be searched (S) is compared with all other latent representations (R1 to Rn) in the dataset (D) and similar scenes (S1 to Sm) are searched. Using the trained trajectory autoencoder (1) and found similar scenes (S1 to Sm), an existing trajectory prediction algorithm is trained, trained and/or tested.
展开▼